ReID_baseline.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ReID.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "0cd1884758d943e4a7cbcee44c5d88ad": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_bc23887ba3d5441baedf58e14efdb64d",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_99fe32270325402d892fafe5fd9e7048",
              "IPY_MODEL_17e2590b35d04f6c9a689921fbbc28b9",
              "IPY_MODEL_60853b977f1948dfb764960c623b153a"
            ]
          }
        },
        "bc23887ba3d5441baedf58e14efdb64d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "99fe32270325402d892fafe5fd9e7048": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_9bd23908225a4502bbf864bf6761bbe7",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": "100%",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_c088f1d254534036a2235f629d4bf7a6"
          }
        },
        "17e2590b35d04f6c9a689921fbbc28b9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_c3d7fa5cfd5e423983dea5d3e6cce18c",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 28,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 28,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_9d1566e06a764d97a30df558124a1262"
          }
        },
        "60853b977f1948dfb764960c623b153a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_8ac39e397aff436a85a4ed3a4a028b53",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 28/28 [00:00&lt;00:00, 90.41it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_c6bf132b57754fecb817195548262e22"
          }
        },
        "9bd23908225a4502bbf864bf6761bbe7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "c088f1d254534036a2235f629d4bf7a6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "c3d7fa5cfd5e423983dea5d3e6cce18c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "9d1566e06a764d97a30df558124a1262": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "8ac39e397aff436a85a4ed3a4a028b53": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "c6bf132b57754fecb817195548262e22": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "bce6852264b64ee599e23bce2aea316e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_ccd77c70a8fd48c498f87bc43ec312e7",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_18b8484209dc49418aa6333e8944f248",
              "IPY_MODEL_f88dff762ed549ffba1f894eb61ac4d4",
              "IPY_MODEL_502bd8ba34a949f88775fa07300031b3"
            ]
          }
        },
        "ccd77c70a8fd48c498f87bc43ec312e7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "18b8484209dc49418aa6333e8944f248": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_c2d4bba57b5f4c38ab64661df0d27c5e",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": "100%",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_70c6202ed76142788a55b1911ece23cd"
          }
        },
        "f88dff762ed549ffba1f894eb61ac4d4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_045b787c9b3b44f8858255cbffad9fe5",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 181,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 181,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_c5c7053f3be546e78f32ab9eda61d173"
          }
        },
        "502bd8ba34a949f88775fa07300031b3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_5dda229fdcb84adfb845966c02128e8a",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 181/181 [00:01&lt;00:00, 97.73it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_3ac5aba209a342ce91ddc5369a205a3f"
          }
        },
        "c2d4bba57b5f4c38ab64661df0d27c5e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "70c6202ed76142788a55b1911ece23cd": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "045b787c9b3b44f8858255cbffad9fe5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "c5c7053f3be546e78f32ab9eda61d173": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "5dda229fdcb84adfb845966c02128e8a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "3ac5aba209a342ce91ddc5369a205a3f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "44a792a0d89d438bb8d7e13098f5563d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_1cf90ccdb0a249ea906da9bc4e645a19",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_ce7ad7ccf2154cc3bbca8edb6ef3cdf7",
              "IPY_MODEL_8ddc904e7b474adcb803c66c319572d0",
              "IPY_MODEL_3a2953e5e8f74e139d117bf343aa9dc5"
            ]
          }
        },
        "1cf90ccdb0a249ea906da9bc4e645a19": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "ce7ad7ccf2154cc3bbca8edb6ef3cdf7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_f7a8eb6d27c84977bb50b2d6a4876baf",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": "100%",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_13a2160214f44257a97503c647c821a1"
          }
        },
        "8ddc904e7b474adcb803c66c319572d0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_7f5688001e4d4180970a4d8a4930d77a",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 28,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 28,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_89cfeb4e98e24353a114d1e400d6b232"
          }
        },
        "3a2953e5e8f74e139d117bf343aa9dc5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_532d981b923c4323a1409b6840bcefc6",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 28/28 [00:00&lt;00:00, 1008.63it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_657eb0dd50164caea73671544e78d4f1"
          }
        },
        "f7a8eb6d27c84977bb50b2d6a4876baf": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "13a2160214f44257a97503c647c821a1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "7f5688001e4d4180970a4d8a4930d77a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "89cfeb4e98e24353a114d1e400d6b232": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "532d981b923c4323a1409b6840bcefc6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "657eb0dd50164caea73671544e78d4f1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "a96508424b81468594e66d2bdb7eef2c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_e5c0b960e93d457dbfd3fc1afec2add4",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_ac0c92b66d814930989d242661c1c41a",
              "IPY_MODEL_55ba1aa8f97940f9b569f7cf24eb4ba3",
              "IPY_MODEL_3c8534f2dc3d4889be8e53ae8e34a102"
            ]
          }
        },
        "e5c0b960e93d457dbfd3fc1afec2add4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "ac0c92b66d814930989d242661c1c41a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_a3475060715743c9b1a2086b9d933d79",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": "100%",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_0901239051704a9aa6435a63ef3eb6a7"
          }
        },
        "55ba1aa8f97940f9b569f7cf24eb4ba3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_2af63ec015ab4990a06c685202955d2e",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 181,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 181,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_0929a6b4f40b4ed19f2f0897fef40b4c"
          }
        },
        "3c8534f2dc3d4889be8e53ae8e34a102": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_a7d6f429edc94dd399ba7dce98e7fbc6",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 181/181 [00:00&lt;00:00, 4442.83it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_3045bc3462cb4a729759f43b0c8471c0"
          }
        },
        "a3475060715743c9b1a2086b9d933d79": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "0901239051704a9aa6435a63ef3eb6a7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "2af63ec015ab4990a06c685202955d2e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "0929a6b4f40b4ed19f2f0897fef40b4c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "a7d6f429edc94dd399ba7dce98e7fbc6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "3045bc3462cb4a729759f43b0c8471c0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RrGoRz-HCSHK",
        "outputId": "1df21ef1-0772-4ce3-ba73-c7d55c8cc9a1"
      },
      "source": [
        "!wget --load-cookies /tmp/cookies.txt \"https://docs.google.com/uc?export=download&confirm=$(wget --quiet --save-cookies /tmp/cookies.txt --keep-session-cookies --no-check-certificate 'https://docs.google.com/uc?export=download&id=1qgRXM0ZiX5_L0V-e1cr_clc2uHDFVcyz' -O- | sed -rn 's/.*confirm=([0-9A-Za-z_]+).*/\\1\\n/p')&id=1qgRXM0ZiX5_L0V-e1cr_clc2uHDFVcyz\" -O starter_code.zip && rm -rf /tmp/cookies.txt"
      ],
      "execution_count": 197,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2021-11-29 18:35:32--  https://docs.google.com/uc?export=download&confirm=&id=1qgRXM0ZiX5_L0V-e1cr_clc2uHDFVcyz\n",
            "Resolving docs.google.com (docs.google.com)... 173.194.218.100, 173.194.218.101, 173.194.218.102, ...\n",
            "Connecting to docs.google.com (docs.google.com)|173.194.218.100|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Moved Temporarily\n",
            "Location: https://doc-10-74-docs.googleusercontent.com/docs/securesc/bugvjvkoq1tg73lftd3g0qfp33qnegm8/n61124m07ft387pl1jve1860jkph51eh/1638210900000/08196725501150479867/17432634495261139668Z/1qgRXM0ZiX5_L0V-e1cr_clc2uHDFVcyz?e=download [following]\n",
            "--2021-11-29 18:35:40--  https://doc-10-74-docs.googleusercontent.com/docs/securesc/bugvjvkoq1tg73lftd3g0qfp33qnegm8/n61124m07ft387pl1jve1860jkph51eh/1638210900000/08196725501150479867/17432634495261139668Z/1qgRXM0ZiX5_L0V-e1cr_clc2uHDFVcyz?e=download\n",
            "Resolving doc-10-74-docs.googleusercontent.com (doc-10-74-docs.googleusercontent.com)... 173.194.218.132, 2607:f8b0:400c:c14::84\n",
            "Connecting to doc-10-74-docs.googleusercontent.com (doc-10-74-docs.googleusercontent.com)|173.194.218.132|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://docs.google.com/nonceSigner?nonce=ojfbilj587o90&continue=https://doc-10-74-docs.googleusercontent.com/docs/securesc/bugvjvkoq1tg73lftd3g0qfp33qnegm8/n61124m07ft387pl1jve1860jkph51eh/1638210900000/08196725501150479867/17432634495261139668Z/1qgRXM0ZiX5_L0V-e1cr_clc2uHDFVcyz?e%3Ddownload&hash=0dps79413g6ejgus69f5sor3bmehsjbr [following]\n",
            "--2021-11-29 18:35:40--  https://docs.google.com/nonceSigner?nonce=ojfbilj587o90&continue=https://doc-10-74-docs.googleusercontent.com/docs/securesc/bugvjvkoq1tg73lftd3g0qfp33qnegm8/n61124m07ft387pl1jve1860jkph51eh/1638210900000/08196725501150479867/17432634495261139668Z/1qgRXM0ZiX5_L0V-e1cr_clc2uHDFVcyz?e%3Ddownload&hash=0dps79413g6ejgus69f5sor3bmehsjbr\n",
            "Connecting to docs.google.com (docs.google.com)|173.194.218.100|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://doc-10-74-docs.googleusercontent.com/docs/securesc/bugvjvkoq1tg73lftd3g0qfp33qnegm8/n61124m07ft387pl1jve1860jkph51eh/1638210900000/08196725501150479867/17432634495261139668Z/1qgRXM0ZiX5_L0V-e1cr_clc2uHDFVcyz?e=download&nonce=ojfbilj587o90&user=17432634495261139668Z&hash=7k2oavtt5hg39cv1hl7d3g8sihos008t [following]\n",
            "--2021-11-29 18:35:40--  https://doc-10-74-docs.googleusercontent.com/docs/securesc/bugvjvkoq1tg73lftd3g0qfp33qnegm8/n61124m07ft387pl1jve1860jkph51eh/1638210900000/08196725501150479867/17432634495261139668Z/1qgRXM0ZiX5_L0V-e1cr_clc2uHDFVcyz?e=download&nonce=ojfbilj587o90&user=17432634495261139668Z&hash=7k2oavtt5hg39cv1hl7d3g8sihos008t\n",
            "Connecting to doc-10-74-docs.googleusercontent.com (doc-10-74-docs.googleusercontent.com)|173.194.218.132|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 3299042 (3.1M) [application/x-zip-compressed]\n",
            "Saving to: ‘starter_code.zip’\n",
            "\n",
            "starter_code.zip    100%[===================>]   3.15M  --.-KB/s    in 0.02s   \n",
            "\n",
            "2021-11-29 18:35:40 (177 MB/s) - ‘starter_code.zip’ saved [3299042/3299042]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ihOKoGtGXXQA",
        "outputId": "de5824a6-e949-4e50-d926-02b44244081a"
      },
      "source": [
        "!unzip -o starter_code"
      ],
      "execution_count": 198,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  starter_code.zip\n",
            "2f4309617046ecaf0bd5d1862c42ed370a17f46c\n",
            "  inflating: reid-col780-master/README.md  \n",
            "  inflating: reid-col780-master/data/train/001/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/001/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/001/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/001/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/001/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/001/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/001/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/001/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/001/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/001/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/001/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/001/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/001/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/001/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/001/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/001/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/003/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/003/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/003/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/003/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/003/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/003/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/003/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/003/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/003/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/003/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/003/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/003/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/003/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/003/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/003/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/003/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/007/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/007/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/007/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/007/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/007/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/007/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/007/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/007/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/007/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/007/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/007/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/007/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/007/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/007/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/007/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/007/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/008/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/008/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/008/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/008/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/008/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/008/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/008/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/008/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/008/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/008/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/008/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/008/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/008/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/008/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/008/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/008/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/009/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/009/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/009/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/009/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/009/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/009/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/009/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/009/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/009/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/009/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/009/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/009/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/009/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/009/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/009/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/009/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/010/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/010/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/010/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/010/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/010/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/010/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/010/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/010/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/010/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/010/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/010/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/010/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/010/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/010/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/010/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/010/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/011/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/011/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/011/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/011/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/011/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/011/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/011/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/011/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/011/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/011/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/011/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/011/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/011/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/011/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/011/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/011/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/013/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/013/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/013/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/013/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/013/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/013/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/013/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/013/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/013/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/013/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/013/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/013/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/013/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/013/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/013/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/013/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/014/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/014/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/014/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/014/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/014/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/014/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/014/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/014/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/014/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/014/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/014/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/014/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/014/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/014/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/014/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/014/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/015/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/015/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/015/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/015/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/015/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/015/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/015/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/015/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/015/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/015/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/015/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/015/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/015/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/015/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/015/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/015/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/016/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/016/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/016/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/016/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/016/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/016/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/016/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/016/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/016/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/016/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/016/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/016/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/016/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/016/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/016/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/016/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/019/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/019/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/019/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/019/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/019/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/019/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/019/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/019/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/019/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/019/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/019/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/019/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/019/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/019/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/019/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/019/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/020/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/020/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/020/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/020/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/020/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/020/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/020/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/020/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/020/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/020/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/020/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/020/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/020/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/020/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/020/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/020/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/021/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/021/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/021/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/021/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/021/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/021/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/021/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/021/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/021/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/021/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/021/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/021/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/021/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/021/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/021/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/021/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/022/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/022/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/022/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/022/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/022/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/022/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/022/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/022/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/022/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/022/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/022/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/022/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/022/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/022/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/022/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/022/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/024/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/024/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/024/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/024/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/024/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/024/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/024/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/024/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/024/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/024/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/024/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/024/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/024/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/024/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/024/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/024/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/027/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/027/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/027/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/027/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/027/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/027/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/027/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/027/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/027/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/027/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/027/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/027/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/027/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/027/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/027/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/027/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/028/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/028/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/028/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/028/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/028/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/028/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/028/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/028/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/028/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/028/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/028/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/028/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/028/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/028/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/028/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/028/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/029/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/029/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/029/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/029/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/029/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/029/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/029/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/029/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/029/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/029/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/029/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/029/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/029/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/029/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/029/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/029/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/030/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/030/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/030/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/030/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/030/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/030/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/030/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/030/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/030/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/030/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/030/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/030/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/030/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/030/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/030/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/030/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/034/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/034/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/034/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/034/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/034/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/034/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/034/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/034/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/034/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/034/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/034/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/034/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/034/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/034/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/034/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/034/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/036/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/036/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/036/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/036/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/036/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/036/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/036/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/036/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/036/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/036/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/036/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/036/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/036/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/036/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/036/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/036/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/037/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/037/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/037/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/037/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/037/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/037/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/037/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/037/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/037/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/037/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/037/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/037/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/037/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/037/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/037/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/037/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/038/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/038/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/038/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/038/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/038/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/038/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/038/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/038/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/038/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/038/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/038/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/038/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/038/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/038/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/038/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/038/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/041/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/041/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/041/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/041/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/041/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/041/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/041/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/041/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/041/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/041/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/041/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/041/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/041/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/041/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/041/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/041/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/042/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/042/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/042/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/042/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/042/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/042/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/042/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/042/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/042/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/042/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/042/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/042/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/042/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/042/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/042/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/042/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/044/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/044/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/044/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/044/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/044/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/044/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/044/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/044/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/044/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/044/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/044/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/044/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/044/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/044/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/044/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/044/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/045/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/045/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/045/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/045/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/045/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/045/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/045/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/045/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/045/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/045/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/045/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/045/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/045/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/045/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/045/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/045/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/047/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/047/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/047/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/047/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/047/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/047/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/047/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/047/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/047/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/047/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/047/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/047/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/047/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/047/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/047/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/047/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/049/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/049/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/049/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/049/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/049/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/049/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/049/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/049/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/049/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/049/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/049/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/049/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/049/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/049/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/049/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/049/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/050/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/050/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/050/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/050/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/050/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/050/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/050/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/050/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/050/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/050/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/050/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/050/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/050/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/050/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/050/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/050/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/051/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/051/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/051/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/051/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/051/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/051/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/051/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/051/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/051/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/051/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/051/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/051/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/051/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/051/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/051/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/051/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/053/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/053/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/053/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/053/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/053/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/053/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/053/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/053/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/053/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/053/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/053/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/053/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/053/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/053/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/053/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/053/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/054/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/054/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/054/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/054/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/054/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/054/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/054/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/054/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/054/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/054/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/054/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/054/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/054/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/054/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/054/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/054/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/058/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/058/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/058/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/058/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/058/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/058/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/058/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/058/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/058/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/058/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/058/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/058/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/058/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/058/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/058/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/058/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/059/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/059/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/059/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/059/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/059/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/059/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/059/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/059/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/059/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/059/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/059/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/059/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/059/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/059/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/059/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/059/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/060/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/060/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/060/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/060/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/060/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/060/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/060/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/060/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/060/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/060/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/060/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/060/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/060/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/060/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/060/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/060/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/061/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/061/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/061/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/061/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/061/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/061/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/061/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/061/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/061/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/061/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/061/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/061/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/061/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/061/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/061/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/061/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/062/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/062/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/062/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/062/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/062/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/062/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/062/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/062/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/062/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/062/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/062/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/062/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/062/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/062/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/062/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/062/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/063/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/063/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/063/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/063/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/063/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/063/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/063/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/063/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/063/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/063/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/063/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/063/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/063/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/063/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/063/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/063/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/065/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/065/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/065/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/065/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/065/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/065/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/065/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/065/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/065/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/065/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/065/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/065/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/065/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/065/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/065/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/065/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/066/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/066/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/066/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/066/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/066/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/066/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/066/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/066/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/066/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/066/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/066/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/066/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/066/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/066/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/066/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/066/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/067/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/067/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/067/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/067/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/067/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/067/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/067/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/067/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/067/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/067/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/067/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/067/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/067/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/067/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/067/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/067/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/070/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/070/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/070/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/070/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/070/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/070/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/070/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/070/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/070/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/070/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/070/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/070/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/070/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/070/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/070/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/070/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/071/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/071/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/071/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/071/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/071/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/071/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/071/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/071/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/071/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/071/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/071/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/071/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/071/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/071/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/071/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/071/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/077/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/077/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/077/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/077/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/077/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/077/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/077/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/077/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/077/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/077/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/077/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/077/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/077/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/077/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/077/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/077/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/079/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/079/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/079/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/079/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/079/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/079/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/079/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/079/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/079/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/079/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/079/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/079/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/079/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/079/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/079/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/079/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/082/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/082/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/082/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/082/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/082/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/082/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/082/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/082/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/082/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/082/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/082/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/082/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/082/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/082/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/082/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/082/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/084/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/084/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/084/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/084/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/084/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/084/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/084/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/084/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/084/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/084/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/084/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/084/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/084/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/084/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/084/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/084/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/085/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/085/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/085/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/085/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/085/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/085/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/085/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/085/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/085/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/085/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/085/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/085/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/085/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/085/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/085/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/085/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/086/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/086/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/086/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/086/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/086/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/086/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/086/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/086/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/086/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/086/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/086/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/086/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/086/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/086/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/086/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/086/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/087/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/087/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/087/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/087/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/087/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/087/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/087/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/087/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/087/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/087/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/087/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/087/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/087/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/087/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/087/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/087/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/089/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/089/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/089/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/089/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/089/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/089/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/089/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/089/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/089/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/089/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/089/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/089/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/089/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/089/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/089/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/089/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/095/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/095/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/095/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/095/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/095/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/095/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/095/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/095/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/095/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/095/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/095/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/095/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/095/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/095/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/095/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/095/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/096/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/096/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/096/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/096/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/096/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/096/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/096/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/096/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/096/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/096/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/096/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/096/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/096/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/096/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/096/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/096/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/097/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/097/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/097/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/097/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/097/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/097/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/097/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/097/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/097/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/097/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/097/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/097/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/097/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/097/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/097/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/097/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/098/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/098/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/098/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/098/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/098/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/098/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/098/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/098/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/098/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/098/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/098/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/098/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/098/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/098/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/098/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/098/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/099/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/099/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/099/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/099/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/099/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/099/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/099/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/099/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/099/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/099/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/099/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/099/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/099/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/099/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/099/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/099/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/102/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/102/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/102/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/102/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/102/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/102/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/102/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/102/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/102/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/102/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/102/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/102/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/102/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/102/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/102/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/102/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/106/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/106/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/106/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/106/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/106/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/106/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/106/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/106/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/106/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/106/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/106/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/106/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/106/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/106/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/106/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/106/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/109/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/109/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/109/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/109/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/109/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/109/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/109/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/109/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/109/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/109/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/109/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/109/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/109/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/109/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/109/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/109/02_8.png  \n",
            "  inflating: reid-col780-master/data/train/114/01_1.png  \n",
            "  inflating: reid-col780-master/data/train/114/01_2.png  \n",
            "  inflating: reid-col780-master/data/train/114/01_3.png  \n",
            "  inflating: reid-col780-master/data/train/114/01_4.png  \n",
            "  inflating: reid-col780-master/data/train/114/01_5.png  \n",
            "  inflating: reid-col780-master/data/train/114/01_6.png  \n",
            "  inflating: reid-col780-master/data/train/114/01_7.png  \n",
            "  inflating: reid-col780-master/data/train/114/01_8.png  \n",
            "  inflating: reid-col780-master/data/train/114/02_1.png  \n",
            "  inflating: reid-col780-master/data/train/114/02_2.png  \n",
            "  inflating: reid-col780-master/data/train/114/02_3.png  \n",
            "  inflating: reid-col780-master/data/train/114/02_4.png  \n",
            "  inflating: reid-col780-master/data/train/114/02_5.png  \n",
            "  inflating: reid-col780-master/data/train/114/02_6.png  \n",
            "  inflating: reid-col780-master/data/train/114/02_7.png  \n",
            "  inflating: reid-col780-master/data/train/114/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/004/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/012/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/017/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/018/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/023/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/032/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/033/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/046/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/052/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/055/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/073/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/all_imgs/088/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/004/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/012/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/017/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/018/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/023/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/032/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/033/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/046/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/052/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/055/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/073/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/gallery/088/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/query/004/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/query/004/01_2.png  \n",
            "  inflating: reid-col780-master/data/val/query/012/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/query/012/02_6.png  \n",
            "  inflating: reid-col780-master/data/val/query/012/02_7.png  \n",
            "  inflating: reid-col780-master/data/val/query/012/02_8.png  \n",
            "  inflating: reid-col780-master/data/val/query/017/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/query/017/01_6.png  \n",
            "  inflating: reid-col780-master/data/val/query/018/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/query/018/01_7.png  \n",
            "  inflating: reid-col780-master/data/val/query/018/01_8.png  \n",
            "  inflating: reid-col780-master/data/val/query/023/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/query/032/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/query/032/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/query/033/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/query/033/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/query/033/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/query/046/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/query/046/01_3.png  \n",
            "  inflating: reid-col780-master/data/val/query/052/02_1.png  \n",
            "  inflating: reid-col780-master/data/val/query/052/02_2.png  \n",
            "  inflating: reid-col780-master/data/val/query/052/02_3.png  \n",
            "  inflating: reid-col780-master/data/val/query/052/02_4.png  \n",
            "  inflating: reid-col780-master/data/val/query/052/02_5.png  \n",
            "  inflating: reid-col780-master/data/val/query/055/01_4.png  \n",
            "  inflating: reid-col780-master/data/val/query/073/01_1.png  \n",
            "  inflating: reid-col780-master/data/val/query/073/01_5.png  \n",
            "  inflating: reid-col780-master/data/val/query/088/01_1.png  \n",
            "  inflating: reid-col780-master/metrics.py  \n",
            "  inflating: reid-col780-master/model.py  \n",
            "  inflating: reid-col780-master/run-test.py  \n",
            "  inflating: reid-col780-master/utils.py  \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TVUcc7svYQw0"
      },
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "from torch.utils.data import Dataset, DataLoader\n",
        "from torchvision import datasets, transforms\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "from torchvision.io import read_image"
      ],
      "execution_count": 199,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZLM8aXHiZqEw"
      },
      "source": [
        "import os\n",
        "import sys\n",
        "starter_path = os.path.abspath(os.getcwd())+'/reid-col780-master'\n",
        "sys.path.insert(0, starter_path)"
      ],
      "execution_count": 200,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kVWT-EBMbvFP"
      },
      "source": [
        "from utils import get_id"
      ],
      "execution_count": 201,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "id": "p_TWT8duhh2N",
        "outputId": "6e0a3d38-5195-46a6-a589-ff979562ff32"
      },
      "source": [
        "import cv2\n",
        "imgSample = cv2.imread(starter_path+'/data/train/001/01_1.png')\n",
        "plt.imshow(imgSample)\n",
        "imgDim = imgSample.shape\n",
        "print(imgDim)"
      ],
      "execution_count": 202,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(128, 48, 3)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ltHuEov1thyP"
      },
      "source": [
        "### Load Dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uDQH7LCPg6xf"
      },
      "source": [
        "def getCameraID(fileName):\n",
        "  id = int(fileName.split('_')[0])\n",
        "  return id"
      ],
      "execution_count": 203,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VhRBtKi0ganU",
        "outputId": "08c22b2a-3bde-4c5f-919b-f7487d173480"
      },
      "source": [
        "trainImages = os.listdir(starter_path+'/data/train')\n",
        "numClasses = len(trainImages)\n",
        "trainImages = [[label, getCameraID(y), trainImages[label]+'/'+y] for label in range(len(trainImages)) for y in os.listdir(starter_path+'/data/train/'+trainImages[label]) ]\n",
        "trainImagesCam1 = []\n",
        "trainImagesCam2 = []\n",
        "for i in range(len(trainImages)):\n",
        "  img_path = os.path.join(starter_path+'/data/train', trainImages[i][2])\n",
        "  image = read_image(img_path)\n",
        "  if trainImages[i][1]==1:\n",
        "    trainImagesCam1.append([i, image])\n",
        "  else:\n",
        "    trainImagesCam2.append([i, image])\n",
        "print(trainImages)\n",
        "print(len(trainImages))"
      ],
      "execution_count": 204,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[0, 2, '051/02_3.png'], [0, 2, '051/02_6.png'], [0, 1, '051/01_5.png'], [0, 1, '051/01_4.png'], [0, 1, '051/01_2.png'], [0, 2, '051/02_2.png'], [0, 2, '051/02_8.png'], [0, 2, '051/02_5.png'], [0, 1, '051/01_6.png'], [0, 1, '051/01_8.png'], [0, 2, '051/02_7.png'], [0, 1, '051/01_7.png'], [0, 1, '051/01_1.png'], [0, 2, '051/02_4.png'], [0, 1, '051/01_3.png'], [0, 2, '051/02_1.png'], [1, 2, '029/02_3.png'], [1, 2, '029/02_6.png'], [1, 1, '029/01_5.png'], [1, 1, '029/01_4.png'], [1, 1, '029/01_2.png'], [1, 2, '029/02_2.png'], [1, 2, '029/02_8.png'], [1, 2, '029/02_5.png'], [1, 1, '029/01_6.png'], [1, 1, '029/01_8.png'], [1, 2, '029/02_7.png'], [1, 1, '029/01_7.png'], [1, 1, '029/01_1.png'], [1, 2, '029/02_4.png'], [1, 1, '029/01_3.png'], [1, 2, '029/02_1.png'], [2, 2, '028/02_3.png'], [2, 2, '028/02_6.png'], [2, 1, '028/01_5.png'], [2, 1, '028/01_4.png'], [2, 1, '028/01_2.png'], [2, 2, '028/02_2.png'], [2, 2, '028/02_8.png'], [2, 2, '028/02_5.png'], [2, 1, '028/01_6.png'], [2, 1, '028/01_8.png'], [2, 2, '028/02_7.png'], [2, 1, '028/01_7.png'], [2, 1, '028/01_1.png'], [2, 2, '028/02_4.png'], [2, 1, '028/01_3.png'], [2, 2, '028/02_1.png'], [3, 2, '053/02_3.png'], [3, 2, '053/02_6.png'], [3, 1, '053/01_5.png'], [3, 1, '053/01_4.png'], [3, 1, '053/01_2.png'], [3, 2, '053/02_2.png'], [3, 2, '053/02_8.png'], [3, 2, '053/02_5.png'], [3, 1, '053/01_6.png'], [3, 1, '053/01_8.png'], [3, 2, '053/02_7.png'], [3, 1, '053/01_7.png'], [3, 1, '053/01_1.png'], [3, 2, '053/02_4.png'], [3, 1, '053/01_3.png'], [3, 2, '053/02_1.png'], [4, 2, '027/02_3.png'], [4, 2, '027/02_6.png'], [4, 1, '027/01_5.png'], [4, 1, '027/01_4.png'], [4, 1, '027/01_2.png'], [4, 2, '027/02_2.png'], [4, 2, '027/02_8.png'], [4, 2, '027/02_5.png'], [4, 1, '027/01_6.png'], [4, 1, '027/01_8.png'], [4, 2, '027/02_7.png'], [4, 1, '027/01_7.png'], [4, 1, '027/01_1.png'], [4, 2, '027/02_4.png'], [4, 1, '027/01_3.png'], [4, 2, '027/02_1.png'], [5, 2, '045/02_3.png'], [5, 2, '045/02_6.png'], [5, 1, '045/01_5.png'], [5, 1, '045/01_4.png'], [5, 1, '045/01_2.png'], [5, 2, '045/02_2.png'], [5, 2, '045/02_8.png'], [5, 2, '045/02_5.png'], [5, 1, '045/01_6.png'], [5, 1, '045/01_8.png'], [5, 2, '045/02_7.png'], [5, 1, '045/01_7.png'], [5, 1, '045/01_1.png'], [5, 2, '045/02_4.png'], [5, 1, '045/01_3.png'], [5, 2, '045/02_1.png'], [6, 2, '095/02_3.png'], [6, 2, '095/02_6.png'], [6, 1, '095/01_5.png'], [6, 1, '095/01_4.png'], [6, 1, '095/01_2.png'], [6, 2, '095/02_2.png'], [6, 2, '095/02_8.png'], [6, 2, '095/02_5.png'], [6, 1, '095/01_6.png'], [6, 1, '095/01_8.png'], [6, 2, '095/02_7.png'], [6, 1, '095/01_7.png'], [6, 1, '095/01_1.png'], [6, 2, '095/02_4.png'], [6, 1, '095/01_3.png'], [6, 2, '095/02_1.png'], [7, 2, '014/02_3.png'], [7, 2, '014/02_6.png'], [7, 1, '014/01_5.png'], [7, 1, '014/01_4.png'], [7, 1, '014/01_2.png'], [7, 2, '014/02_2.png'], [7, 2, '014/02_8.png'], [7, 2, '014/02_5.png'], [7, 1, '014/01_6.png'], [7, 1, '014/01_8.png'], [7, 2, '014/02_7.png'], [7, 1, '014/01_7.png'], [7, 1, '014/01_1.png'], [7, 2, '014/02_4.png'], [7, 1, '014/01_3.png'], [7, 2, '014/02_1.png'], [8, 2, '050/02_3.png'], [8, 2, '050/02_6.png'], [8, 1, '050/01_5.png'], [8, 1, '050/01_4.png'], [8, 1, '050/01_2.png'], [8, 2, '050/02_2.png'], [8, 2, '050/02_8.png'], [8, 2, '050/02_5.png'], [8, 1, '050/01_6.png'], [8, 1, '050/01_8.png'], [8, 2, '050/02_7.png'], [8, 1, '050/01_7.png'], [8, 1, '050/01_1.png'], [8, 2, '050/02_4.png'], [8, 1, '050/01_3.png'], [8, 2, '050/02_1.png'], [9, 2, '013/02_3.png'], [9, 2, '013/02_6.png'], [9, 1, '013/01_5.png'], [9, 1, '013/01_4.png'], [9, 1, '013/01_2.png'], [9, 2, '013/02_2.png'], [9, 2, '013/02_8.png'], [9, 2, '013/02_5.png'], [9, 1, '013/01_6.png'], [9, 1, '013/01_8.png'], [9, 2, '013/02_7.png'], [9, 1, '013/01_7.png'], [9, 1, '013/01_1.png'], [9, 2, '013/02_4.png'], [9, 1, '013/01_3.png'], [9, 2, '013/02_1.png'], [10, 2, '065/02_3.png'], [10, 2, '065/02_6.png'], [10, 1, '065/01_5.png'], [10, 1, '065/01_4.png'], [10, 1, '065/01_2.png'], [10, 2, '065/02_2.png'], [10, 2, '065/02_8.png'], [10, 2, '065/02_5.png'], [10, 1, '065/01_6.png'], [10, 1, '065/01_8.png'], [10, 2, '065/02_7.png'], [10, 1, '065/01_7.png'], [10, 1, '065/01_1.png'], [10, 2, '065/02_4.png'], [10, 1, '065/01_3.png'], [10, 2, '065/02_1.png'], [11, 2, '066/02_3.png'], [11, 2, '066/02_6.png'], [11, 1, '066/01_5.png'], [11, 1, '066/01_4.png'], [11, 1, '066/01_2.png'], [11, 2, '066/02_2.png'], [11, 2, '066/02_8.png'], [11, 2, '066/02_5.png'], [11, 1, '066/01_6.png'], [11, 1, '066/01_8.png'], [11, 2, '066/02_7.png'], [11, 1, '066/01_7.png'], [11, 1, '066/01_1.png'], [11, 2, '066/02_4.png'], [11, 1, '066/01_3.png'], [11, 2, '066/02_1.png'], [12, 2, '019/02_3.png'], [12, 2, '019/02_6.png'], [12, 1, '019/01_5.png'], [12, 1, '019/01_4.png'], [12, 1, '019/01_2.png'], [12, 2, '019/02_2.png'], [12, 2, '019/02_8.png'], [12, 2, '019/02_5.png'], [12, 1, '019/01_6.png'], [12, 1, '019/01_8.png'], [12, 2, '019/02_7.png'], [12, 1, '019/01_7.png'], [12, 1, '019/01_1.png'], [12, 2, '019/02_4.png'], [12, 1, '019/01_3.png'], [12, 2, '019/02_1.png'], [13, 2, '008/02_3.png'], [13, 2, '008/02_6.png'], [13, 1, '008/01_5.png'], [13, 1, '008/01_4.png'], [13, 1, '008/01_2.png'], [13, 2, '008/02_2.png'], [13, 2, '008/02_8.png'], [13, 2, '008/02_5.png'], [13, 1, '008/01_6.png'], [13, 1, '008/01_8.png'], [13, 2, '008/02_7.png'], [13, 1, '008/01_7.png'], [13, 1, '008/01_1.png'], [13, 2, '008/02_4.png'], [13, 1, '008/01_3.png'], [13, 2, '008/02_1.png'], [14, 2, '038/02_3.png'], [14, 2, '038/02_6.png'], [14, 1, '038/01_5.png'], [14, 1, '038/01_4.png'], [14, 1, '038/01_2.png'], [14, 2, '038/02_2.png'], [14, 2, '038/02_8.png'], [14, 2, '038/02_5.png'], [14, 1, '038/01_6.png'], [14, 1, '038/01_8.png'], [14, 2, '038/02_7.png'], [14, 1, '038/01_7.png'], [14, 1, '038/01_1.png'], [14, 2, '038/02_4.png'], [14, 1, '038/01_3.png'], [14, 2, '038/02_1.png'], [15, 2, '089/02_3.png'], [15, 2, '089/02_6.png'], [15, 1, '089/01_5.png'], [15, 1, '089/01_4.png'], [15, 1, '089/01_2.png'], [15, 2, '089/02_2.png'], [15, 2, '089/02_8.png'], [15, 2, '089/02_5.png'], [15, 1, '089/01_6.png'], [15, 1, '089/01_8.png'], [15, 2, '089/02_7.png'], [15, 1, '089/01_7.png'], [15, 1, '089/01_1.png'], [15, 2, '089/02_4.png'], [15, 1, '089/01_3.png'], [15, 2, '089/02_1.png'], [16, 2, '084/02_3.png'], [16, 2, '084/02_6.png'], [16, 1, '084/01_5.png'], [16, 1, '084/01_4.png'], [16, 1, '084/01_2.png'], [16, 2, '084/02_2.png'], [16, 2, '084/02_8.png'], [16, 2, '084/02_5.png'], [16, 1, '084/01_6.png'], [16, 1, '084/01_8.png'], [16, 2, '084/02_7.png'], [16, 1, '084/01_7.png'], [16, 1, '084/01_1.png'], [16, 2, '084/02_4.png'], [16, 1, '084/01_3.png'], [16, 2, '084/02_1.png'], [17, 2, '036/02_3.png'], [17, 2, '036/02_6.png'], [17, 1, '036/01_5.png'], [17, 1, '036/01_4.png'], [17, 1, '036/01_2.png'], [17, 2, '036/02_2.png'], [17, 2, '036/02_8.png'], [17, 2, '036/02_5.png'], [17, 1, '036/01_6.png'], [17, 1, '036/01_8.png'], [17, 2, '036/02_7.png'], [17, 1, '036/01_7.png'], [17, 1, '036/01_1.png'], [17, 2, '036/02_4.png'], [17, 1, '036/01_3.png'], [17, 2, '036/02_1.png'], [18, 2, '099/02_3.png'], [18, 2, '099/02_6.png'], [18, 1, '099/01_5.png'], [18, 1, '099/01_4.png'], [18, 1, '099/01_2.png'], [18, 2, '099/02_2.png'], [18, 2, '099/02_8.png'], [18, 2, '099/02_5.png'], [18, 1, '099/01_6.png'], [18, 1, '099/01_8.png'], [18, 2, '099/02_7.png'], [18, 1, '099/01_7.png'], [18, 1, '099/01_1.png'], [18, 2, '099/02_4.png'], [18, 1, '099/01_3.png'], [18, 2, '099/02_1.png'], [19, 2, '030/02_3.png'], [19, 2, '030/02_6.png'], [19, 1, '030/01_5.png'], [19, 1, '030/01_4.png'], [19, 1, '030/01_2.png'], [19, 2, '030/02_2.png'], [19, 2, '030/02_8.png'], [19, 2, '030/02_5.png'], [19, 1, '030/01_6.png'], [19, 1, '030/01_8.png'], [19, 2, '030/02_7.png'], [19, 1, '030/01_7.png'], [19, 1, '030/01_1.png'], [19, 2, '030/02_4.png'], [19, 1, '030/01_3.png'], [19, 2, '030/02_1.png'], [20, 2, '034/02_3.png'], [20, 2, '034/02_6.png'], [20, 1, '034/01_5.png'], [20, 1, '034/01_4.png'], [20, 1, '034/01_2.png'], [20, 2, '034/02_2.png'], [20, 2, '034/02_8.png'], [20, 2, '034/02_5.png'], [20, 1, '034/01_6.png'], [20, 1, '034/01_8.png'], [20, 2, '034/02_7.png'], [20, 1, '034/01_7.png'], [20, 1, '034/01_1.png'], [20, 2, '034/02_4.png'], [20, 1, '034/01_3.png'], [20, 2, '034/02_1.png'], [21, 2, '102/02_3.png'], [21, 2, '102/02_6.png'], [21, 1, '102/01_5.png'], [21, 1, '102/01_4.png'], [21, 1, '102/01_2.png'], [21, 2, '102/02_2.png'], [21, 2, '102/02_8.png'], [21, 2, '102/02_5.png'], [21, 1, '102/01_6.png'], [21, 1, '102/01_8.png'], [21, 2, '102/02_7.png'], [21, 1, '102/01_7.png'], [21, 1, '102/01_1.png'], [21, 2, '102/02_4.png'], [21, 1, '102/01_3.png'], [21, 2, '102/02_1.png'], [22, 2, '054/02_3.png'], [22, 2, '054/02_6.png'], [22, 1, '054/01_5.png'], [22, 1, '054/01_4.png'], [22, 1, '054/01_2.png'], [22, 2, '054/02_2.png'], [22, 2, '054/02_8.png'], [22, 2, '054/02_5.png'], [22, 1, '054/01_6.png'], [22, 1, '054/01_8.png'], [22, 2, '054/02_7.png'], [22, 1, '054/01_7.png'], [22, 1, '054/01_1.png'], [22, 2, '054/02_4.png'], [22, 1, '054/01_3.png'], [22, 2, '054/02_1.png'], [23, 2, '041/02_3.png'], [23, 2, '041/02_6.png'], [23, 1, '041/01_5.png'], [23, 1, '041/01_4.png'], [23, 1, '041/01_2.png'], [23, 2, '041/02_2.png'], [23, 2, '041/02_8.png'], [23, 2, '041/02_5.png'], [23, 1, '041/01_6.png'], [23, 1, '041/01_8.png'], [23, 2, '041/02_7.png'], [23, 1, '041/01_7.png'], [23, 1, '041/01_1.png'], [23, 2, '041/02_4.png'], [23, 1, '041/01_3.png'], [23, 2, '041/02_1.png'], [24, 2, '082/02_3.png'], [24, 2, '082/02_6.png'], [24, 1, '082/01_5.png'], [24, 1, '082/01_4.png'], [24, 1, '082/01_2.png'], [24, 2, '082/02_2.png'], [24, 2, '082/02_8.png'], [24, 2, '082/02_5.png'], [24, 1, '082/01_6.png'], [24, 1, '082/01_8.png'], [24, 2, '082/02_7.png'], [24, 1, '082/01_7.png'], [24, 1, '082/01_1.png'], [24, 2, '082/02_4.png'], [24, 1, '082/01_3.png'], [24, 2, '082/02_1.png'], [25, 2, '003/02_3.png'], [25, 2, '003/02_6.png'], [25, 1, '003/01_5.png'], [25, 1, '003/01_4.png'], [25, 1, '003/01_2.png'], [25, 2, '003/02_2.png'], [25, 2, '003/02_8.png'], [25, 2, '003/02_5.png'], [25, 1, '003/01_6.png'], [25, 1, '003/01_8.png'], [25, 2, '003/02_7.png'], [25, 1, '003/01_7.png'], [25, 1, '003/01_1.png'], [25, 2, '003/02_4.png'], [25, 1, '003/01_3.png'], [25, 2, '003/02_1.png'], [26, 2, '086/02_3.png'], [26, 2, '086/02_6.png'], [26, 1, '086/01_5.png'], [26, 1, '086/01_4.png'], [26, 1, '086/01_2.png'], [26, 2, '086/02_2.png'], [26, 2, '086/02_8.png'], [26, 2, '086/02_5.png'], [26, 1, '086/01_6.png'], [26, 1, '086/01_8.png'], [26, 2, '086/02_7.png'], [26, 1, '086/01_7.png'], [26, 1, '086/01_1.png'], [26, 2, '086/02_4.png'], [26, 1, '086/01_3.png'], [26, 2, '086/02_1.png'], [27, 2, '109/02_3.png'], [27, 2, '109/02_6.png'], [27, 1, '109/01_5.png'], [27, 1, '109/01_4.png'], [27, 1, '109/01_2.png'], [27, 2, '109/02_2.png'], [27, 2, '109/02_8.png'], [27, 2, '109/02_5.png'], [27, 1, '109/01_6.png'], [27, 1, '109/01_8.png'], [27, 2, '109/02_7.png'], [27, 1, '109/01_7.png'], [27, 1, '109/01_1.png'], [27, 2, '109/02_4.png'], [27, 1, '109/01_3.png'], [27, 2, '109/02_1.png'], [28, 2, '070/02_3.png'], [28, 2, '070/02_6.png'], [28, 1, '070/01_5.png'], [28, 1, '070/01_4.png'], [28, 1, '070/01_2.png'], [28, 2, '070/02_2.png'], [28, 2, '070/02_8.png'], [28, 2, '070/02_5.png'], [28, 1, '070/01_6.png'], [28, 1, '070/01_8.png'], [28, 2, '070/02_7.png'], [28, 1, '070/01_7.png'], [28, 1, '070/01_1.png'], [28, 2, '070/02_4.png'], [28, 1, '070/01_3.png'], [28, 2, '070/02_1.png'], [29, 2, '061/02_3.png'], [29, 2, '061/02_6.png'], [29, 1, '061/01_5.png'], [29, 1, '061/01_4.png'], [29, 1, '061/01_2.png'], [29, 2, '061/02_2.png'], [29, 2, '061/02_8.png'], [29, 2, '061/02_5.png'], [29, 1, '061/01_6.png'], [29, 1, '061/01_8.png'], [29, 2, '061/02_7.png'], [29, 1, '061/01_7.png'], [29, 1, '061/01_1.png'], [29, 2, '061/02_4.png'], [29, 1, '061/01_3.png'], [29, 2, '061/02_1.png'], [30, 2, '049/02_3.png'], [30, 2, '049/02_6.png'], [30, 1, '049/01_5.png'], [30, 1, '049/01_4.png'], [30, 1, '049/01_2.png'], [30, 2, '049/02_2.png'], [30, 2, '049/02_8.png'], [30, 2, '049/02_5.png'], [30, 1, '049/01_6.png'], [30, 1, '049/01_8.png'], [30, 2, '049/02_7.png'], [30, 1, '049/01_7.png'], [30, 1, '049/01_1.png'], [30, 2, '049/02_4.png'], [30, 1, '049/01_3.png'], [30, 2, '049/02_1.png'], [31, 2, '024/02_3.png'], [31, 2, '024/02_6.png'], [31, 1, '024/01_5.png'], [31, 1, '024/01_4.png'], [31, 1, '024/01_2.png'], [31, 2, '024/02_2.png'], [31, 2, '024/02_8.png'], [31, 2, '024/02_5.png'], [31, 1, '024/01_6.png'], [31, 1, '024/01_8.png'], [31, 2, '024/02_7.png'], [31, 1, '024/01_7.png'], [31, 1, '024/01_1.png'], [31, 2, '024/02_4.png'], [31, 1, '024/01_3.png'], [31, 2, '024/02_1.png'], [32, 2, '063/02_3.png'], [32, 2, '063/02_6.png'], [32, 1, '063/01_5.png'], [32, 1, '063/01_4.png'], [32, 1, '063/01_2.png'], [32, 2, '063/02_2.png'], [32, 2, '063/02_8.png'], [32, 2, '063/02_5.png'], [32, 1, '063/01_6.png'], [32, 1, '063/01_8.png'], [32, 2, '063/02_7.png'], [32, 1, '063/01_7.png'], [32, 1, '063/01_1.png'], [32, 2, '063/02_4.png'], [32, 1, '063/01_3.png'], [32, 2, '063/02_1.png'], [33, 2, '077/02_3.png'], [33, 2, '077/02_6.png'], [33, 1, '077/01_5.png'], [33, 1, '077/01_4.png'], [33, 1, '077/01_2.png'], [33, 2, '077/02_2.png'], [33, 2, '077/02_8.png'], [33, 2, '077/02_5.png'], [33, 1, '077/01_6.png'], [33, 1, '077/01_8.png'], [33, 2, '077/02_7.png'], [33, 1, '077/01_7.png'], [33, 1, '077/01_1.png'], [33, 2, '077/02_4.png'], [33, 1, '077/01_3.png'], [33, 2, '077/02_1.png'], [34, 2, '042/02_3.png'], [34, 2, '042/02_6.png'], [34, 1, '042/01_5.png'], [34, 1, '042/01_4.png'], [34, 1, '042/01_2.png'], [34, 2, '042/02_2.png'], [34, 2, '042/02_8.png'], [34, 2, '042/02_5.png'], [34, 1, '042/01_6.png'], [34, 1, '042/01_8.png'], [34, 2, '042/02_7.png'], [34, 1, '042/01_7.png'], [34, 1, '042/01_1.png'], [34, 2, '042/02_4.png'], [34, 1, '042/01_3.png'], [34, 2, '042/02_1.png'], [35, 2, '047/02_3.png'], [35, 2, '047/02_6.png'], [35, 1, '047/01_5.png'], [35, 1, '047/01_4.png'], [35, 1, '047/01_2.png'], [35, 2, '047/02_2.png'], [35, 2, '047/02_8.png'], [35, 2, '047/02_5.png'], [35, 1, '047/01_6.png'], [35, 1, '047/01_8.png'], [35, 2, '047/02_7.png'], [35, 1, '047/01_7.png'], [35, 1, '047/01_1.png'], [35, 2, '047/02_4.png'], [35, 1, '047/01_3.png'], [35, 2, '047/02_1.png'], [36, 2, '016/02_3.png'], [36, 2, '016/02_6.png'], [36, 1, '016/01_5.png'], [36, 1, '016/01_4.png'], [36, 1, '016/01_2.png'], [36, 2, '016/02_2.png'], [36, 2, '016/02_8.png'], [36, 2, '016/02_5.png'], [36, 1, '016/01_6.png'], [36, 1, '016/01_8.png'], [36, 2, '016/02_7.png'], [36, 1, '016/01_7.png'], [36, 1, '016/01_1.png'], [36, 2, '016/02_4.png'], [36, 1, '016/01_3.png'], [36, 2, '016/02_1.png'], [37, 2, '060/02_3.png'], [37, 2, '060/02_6.png'], [37, 1, '060/01_5.png'], [37, 1, '060/01_4.png'], [37, 1, '060/01_2.png'], [37, 2, '060/02_2.png'], [37, 2, '060/02_8.png'], [37, 2, '060/02_5.png'], [37, 1, '060/01_6.png'], [37, 1, '060/01_8.png'], [37, 2, '060/02_7.png'], [37, 1, '060/01_7.png'], [37, 1, '060/01_1.png'], [37, 2, '060/02_4.png'], [37, 1, '060/01_3.png'], [37, 2, '060/02_1.png'], [38, 2, '085/02_3.png'], [38, 2, '085/02_6.png'], [38, 1, '085/01_5.png'], [38, 1, '085/01_4.png'], [38, 1, '085/01_2.png'], [38, 2, '085/02_2.png'], [38, 2, '085/02_8.png'], [38, 2, '085/02_5.png'], [38, 1, '085/01_6.png'], [38, 1, '085/01_8.png'], [38, 2, '085/02_7.png'], [38, 1, '085/01_7.png'], [38, 1, '085/01_1.png'], [38, 2, '085/02_4.png'], [38, 1, '085/01_3.png'], [38, 2, '085/02_1.png'], [39, 2, '087/02_3.png'], [39, 2, '087/02_6.png'], [39, 1, '087/01_5.png'], [39, 1, '087/01_4.png'], [39, 1, '087/01_2.png'], [39, 2, '087/02_2.png'], [39, 2, '087/02_8.png'], [39, 2, '087/02_5.png'], [39, 1, '087/01_6.png'], [39, 1, '087/01_8.png'], [39, 2, '087/02_7.png'], [39, 1, '087/01_7.png'], [39, 1, '087/01_1.png'], [39, 2, '087/02_4.png'], [39, 1, '087/01_3.png'], [39, 2, '087/02_1.png'], [40, 2, '058/02_3.png'], [40, 2, '058/02_6.png'], [40, 1, '058/01_5.png'], [40, 1, '058/01_4.png'], [40, 1, '058/01_2.png'], [40, 2, '058/02_2.png'], [40, 2, '058/02_8.png'], [40, 2, '058/02_5.png'], [40, 1, '058/01_6.png'], [40, 1, '058/01_8.png'], [40, 2, '058/02_7.png'], [40, 1, '058/01_7.png'], [40, 1, '058/01_1.png'], [40, 2, '058/02_4.png'], [40, 1, '058/01_3.png'], [40, 2, '058/02_1.png'], [41, 2, '037/02_3.png'], [41, 2, '037/02_6.png'], [41, 1, '037/01_5.png'], [41, 1, '037/01_4.png'], [41, 1, '037/01_2.png'], [41, 2, '037/02_2.png'], [41, 2, '037/02_8.png'], [41, 2, '037/02_5.png'], [41, 1, '037/01_6.png'], [41, 1, '037/01_8.png'], [41, 2, '037/02_7.png'], [41, 1, '037/01_7.png'], [41, 1, '037/01_1.png'], [41, 2, '037/02_4.png'], [41, 1, '037/01_3.png'], [41, 2, '037/02_1.png'], [42, 2, '021/02_3.png'], [42, 2, '021/02_6.png'], [42, 1, '021/01_5.png'], [42, 1, '021/01_4.png'], [42, 1, '021/01_2.png'], [42, 2, '021/02_2.png'], [42, 2, '021/02_8.png'], [42, 2, '021/02_5.png'], [42, 1, '021/01_6.png'], [42, 1, '021/01_8.png'], [42, 2, '021/02_7.png'], [42, 1, '021/01_7.png'], [42, 1, '021/01_1.png'], [42, 2, '021/02_4.png'], [42, 1, '021/01_3.png'], [42, 2, '021/02_1.png'], [43, 2, '096/02_3.png'], [43, 2, '096/02_6.png'], [43, 1, '096/01_5.png'], [43, 1, '096/01_4.png'], [43, 1, '096/01_2.png'], [43, 2, '096/02_2.png'], [43, 2, '096/02_8.png'], [43, 2, '096/02_5.png'], [43, 1, '096/01_6.png'], [43, 1, '096/01_8.png'], [43, 2, '096/02_7.png'], [43, 1, '096/01_7.png'], [43, 1, '096/01_1.png'], [43, 2, '096/02_4.png'], [43, 1, '096/01_3.png'], [43, 2, '096/02_1.png'], [44, 2, '007/02_3.png'], [44, 2, '007/02_6.png'], [44, 1, '007/01_5.png'], [44, 1, '007/01_4.png'], [44, 1, '007/01_2.png'], [44, 2, '007/02_2.png'], [44, 2, '007/02_8.png'], [44, 2, '007/02_5.png'], [44, 1, '007/01_6.png'], [44, 1, '007/01_8.png'], [44, 2, '007/02_7.png'], [44, 1, '007/01_7.png'], [44, 1, '007/01_1.png'], [44, 2, '007/02_4.png'], [44, 1, '007/01_3.png'], [44, 2, '007/02_1.png'], [45, 2, '011/02_3.png'], [45, 2, '011/02_6.png'], [45, 1, '011/01_5.png'], [45, 1, '011/01_4.png'], [45, 1, '011/01_2.png'], [45, 2, '011/02_2.png'], [45, 2, '011/02_8.png'], [45, 2, '011/02_5.png'], [45, 1, '011/01_6.png'], [45, 1, '011/01_8.png'], [45, 2, '011/02_7.png'], [45, 1, '011/01_7.png'], [45, 1, '011/01_1.png'], [45, 2, '011/02_4.png'], [45, 1, '011/01_3.png'], [45, 2, '011/02_1.png'], [46, 2, '009/02_3.png'], [46, 2, '009/02_6.png'], [46, 1, '009/01_5.png'], [46, 1, '009/01_4.png'], [46, 1, '009/01_2.png'], [46, 2, '009/02_2.png'], [46, 2, '009/02_8.png'], [46, 2, '009/02_5.png'], [46, 1, '009/01_6.png'], [46, 1, '009/01_8.png'], [46, 2, '009/02_7.png'], [46, 1, '009/01_7.png'], [46, 1, '009/01_1.png'], [46, 2, '009/02_4.png'], [46, 1, '009/01_3.png'], [46, 2, '009/02_1.png'], [47, 2, '059/02_3.png'], [47, 2, '059/02_6.png'], [47, 1, '059/01_5.png'], [47, 1, '059/01_4.png'], [47, 1, '059/01_2.png'], [47, 2, '059/02_2.png'], [47, 2, '059/02_8.png'], [47, 2, '059/02_5.png'], [47, 1, '059/01_6.png'], [47, 1, '059/01_8.png'], [47, 2, '059/02_7.png'], [47, 1, '059/01_7.png'], [47, 1, '059/01_1.png'], [47, 2, '059/02_4.png'], [47, 1, '059/01_3.png'], [47, 2, '059/02_1.png'], [48, 2, '015/02_3.png'], [48, 2, '015/02_6.png'], [48, 1, '015/01_5.png'], [48, 1, '015/01_4.png'], [48, 1, '015/01_2.png'], [48, 2, '015/02_2.png'], [48, 2, '015/02_8.png'], [48, 2, '015/02_5.png'], [48, 1, '015/01_6.png'], [48, 1, '015/01_8.png'], [48, 2, '015/02_7.png'], [48, 1, '015/01_7.png'], [48, 1, '015/01_1.png'], [48, 2, '015/02_4.png'], [48, 1, '015/01_3.png'], [48, 2, '015/02_1.png'], [49, 2, '079/02_3.png'], [49, 2, '079/02_6.png'], [49, 1, '079/01_5.png'], [49, 1, '079/01_4.png'], [49, 1, '079/01_2.png'], [49, 2, '079/02_2.png'], [49, 2, '079/02_8.png'], [49, 2, '079/02_5.png'], [49, 1, '079/01_6.png'], [49, 1, '079/01_8.png'], [49, 2, '079/02_7.png'], [49, 1, '079/01_7.png'], [49, 1, '079/01_1.png'], [49, 2, '079/02_4.png'], [49, 1, '079/01_3.png'], [49, 2, '079/02_1.png'], [50, 2, '071/02_3.png'], [50, 2, '071/02_6.png'], [50, 1, '071/01_5.png'], [50, 1, '071/01_4.png'], [50, 1, '071/01_2.png'], [50, 2, '071/02_2.png'], [50, 2, '071/02_8.png'], [50, 2, '071/02_5.png'], [50, 1, '071/01_6.png'], [50, 1, '071/01_8.png'], [50, 2, '071/02_7.png'], [50, 1, '071/01_7.png'], [50, 1, '071/01_1.png'], [50, 2, '071/02_4.png'], [50, 1, '071/01_3.png'], [50, 2, '071/02_1.png'], [51, 2, '097/02_3.png'], [51, 2, '097/02_6.png'], [51, 1, '097/01_5.png'], [51, 1, '097/01_4.png'], [51, 1, '097/01_2.png'], [51, 2, '097/02_2.png'], [51, 2, '097/02_8.png'], [51, 2, '097/02_5.png'], [51, 1, '097/01_6.png'], [51, 1, '097/01_8.png'], [51, 2, '097/02_7.png'], [51, 1, '097/01_7.png'], [51, 1, '097/01_1.png'], [51, 2, '097/02_4.png'], [51, 1, '097/01_3.png'], [51, 2, '097/02_1.png'], [52, 2, '098/02_3.png'], [52, 2, '098/02_6.png'], [52, 1, '098/01_5.png'], [52, 1, '098/01_4.png'], [52, 1, '098/01_2.png'], [52, 2, '098/02_2.png'], [52, 2, '098/02_8.png'], [52, 2, '098/02_5.png'], [52, 1, '098/01_6.png'], [52, 1, '098/01_8.png'], [52, 2, '098/02_7.png'], [52, 1, '098/01_7.png'], [52, 1, '098/01_1.png'], [52, 2, '098/02_4.png'], [52, 1, '098/01_3.png'], [52, 2, '098/02_1.png'], [53, 2, '062/02_3.png'], [53, 2, '062/02_6.png'], [53, 1, '062/01_5.png'], [53, 1, '062/01_4.png'], [53, 1, '062/01_2.png'], [53, 2, '062/02_2.png'], [53, 2, '062/02_8.png'], [53, 2, '062/02_5.png'], [53, 1, '062/01_6.png'], [53, 1, '062/01_8.png'], [53, 2, '062/02_7.png'], [53, 1, '062/01_7.png'], [53, 1, '062/01_1.png'], [53, 2, '062/02_4.png'], [53, 1, '062/01_3.png'], [53, 2, '062/02_1.png'], [54, 2, '114/02_3.png'], [54, 2, '114/02_6.png'], [54, 1, '114/01_5.png'], [54, 1, '114/01_4.png'], [54, 1, '114/01_2.png'], [54, 2, '114/02_2.png'], [54, 2, '114/02_8.png'], [54, 2, '114/02_5.png'], [54, 1, '114/01_6.png'], [54, 1, '114/01_8.png'], [54, 2, '114/02_7.png'], [54, 1, '114/01_7.png'], [54, 1, '114/01_1.png'], [54, 2, '114/02_4.png'], [54, 1, '114/01_3.png'], [54, 2, '114/02_1.png'], [55, 2, '020/02_3.png'], [55, 2, '020/02_6.png'], [55, 1, '020/01_5.png'], [55, 1, '020/01_4.png'], [55, 1, '020/01_2.png'], [55, 2, '020/02_2.png'], [55, 2, '020/02_8.png'], [55, 2, '020/02_5.png'], [55, 1, '020/01_6.png'], [55, 1, '020/01_8.png'], [55, 2, '020/02_7.png'], [55, 1, '020/01_7.png'], [55, 1, '020/01_1.png'], [55, 2, '020/02_4.png'], [55, 1, '020/01_3.png'], [55, 2, '020/02_1.png'], [56, 2, '044/02_3.png'], [56, 2, '044/02_6.png'], [56, 1, '044/01_5.png'], [56, 1, '044/01_4.png'], [56, 1, '044/01_2.png'], [56, 2, '044/02_2.png'], [56, 2, '044/02_8.png'], [56, 2, '044/02_5.png'], [56, 1, '044/01_6.png'], [56, 1, '044/01_8.png'], [56, 2, '044/02_7.png'], [56, 1, '044/01_7.png'], [56, 1, '044/01_1.png'], [56, 2, '044/02_4.png'], [56, 1, '044/01_3.png'], [56, 2, '044/02_1.png'], [57, 2, '106/02_3.png'], [57, 2, '106/02_6.png'], [57, 1, '106/01_5.png'], [57, 1, '106/01_4.png'], [57, 1, '106/01_2.png'], [57, 2, '106/02_2.png'], [57, 2, '106/02_8.png'], [57, 2, '106/02_5.png'], [57, 1, '106/01_6.png'], [57, 1, '106/01_8.png'], [57, 2, '106/02_7.png'], [57, 1, '106/01_7.png'], [57, 1, '106/01_1.png'], [57, 2, '106/02_4.png'], [57, 1, '106/01_3.png'], [57, 2, '106/02_1.png'], [58, 2, '001/02_3.png'], [58, 2, '001/02_6.png'], [58, 1, '001/01_5.png'], [58, 1, '001/01_4.png'], [58, 1, '001/01_2.png'], [58, 2, '001/02_2.png'], [58, 2, '001/02_8.png'], [58, 2, '001/02_5.png'], [58, 1, '001/01_6.png'], [58, 1, '001/01_8.png'], [58, 2, '001/02_7.png'], [58, 1, '001/01_7.png'], [58, 1, '001/01_1.png'], [58, 2, '001/02_4.png'], [58, 1, '001/01_3.png'], [58, 2, '001/02_1.png'], [59, 2, '022/02_3.png'], [59, 2, '022/02_6.png'], [59, 1, '022/01_5.png'], [59, 1, '022/01_4.png'], [59, 1, '022/01_2.png'], [59, 2, '022/02_2.png'], [59, 2, '022/02_8.png'], [59, 2, '022/02_5.png'], [59, 1, '022/01_6.png'], [59, 1, '022/01_8.png'], [59, 2, '022/02_7.png'], [59, 1, '022/01_7.png'], [59, 1, '022/01_1.png'], [59, 2, '022/02_4.png'], [59, 1, '022/01_3.png'], [59, 2, '022/02_1.png'], [60, 2, '010/02_3.png'], [60, 2, '010/02_6.png'], [60, 1, '010/01_5.png'], [60, 1, '010/01_4.png'], [60, 1, '010/01_2.png'], [60, 2, '010/02_2.png'], [60, 2, '010/02_8.png'], [60, 2, '010/02_5.png'], [60, 1, '010/01_6.png'], [60, 1, '010/01_8.png'], [60, 2, '010/02_7.png'], [60, 1, '010/01_7.png'], [60, 1, '010/01_1.png'], [60, 2, '010/02_4.png'], [60, 1, '010/01_3.png'], [60, 2, '010/02_1.png'], [61, 2, '067/02_3.png'], [61, 2, '067/02_6.png'], [61, 1, '067/01_5.png'], [61, 1, '067/01_4.png'], [61, 1, '067/01_2.png'], [61, 2, '067/02_2.png'], [61, 2, '067/02_8.png'], [61, 2, '067/02_5.png'], [61, 1, '067/01_6.png'], [61, 1, '067/01_8.png'], [61, 2, '067/02_7.png'], [61, 1, '067/01_7.png'], [61, 1, '067/01_1.png'], [61, 2, '067/02_4.png'], [61, 1, '067/01_3.png'], [61, 2, '067/02_1.png']]\n",
            "992\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0NVX3vH18ToC"
      },
      "source": [
        "transform_train_list = [\n",
        "        transforms.ToPILImage(),\n",
        "        transforms.Resize((128,48)),\n",
        "        transforms.ToTensor(),\n",
        "    ]\n",
        "transform_train = transforms.Compose(transform_train_list)\n",
        "target_transform = transforms.Lambda(lambda y: torch.zeros(numClasses, dtype=torch.float).scatter_(dim=0, index=torch.tensor(y), value=1))"
      ],
      "execution_count": 205,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oYlLGwsPtUEv"
      },
      "source": [
        "### Normalization"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k134ZiV8jKeq"
      },
      "source": [
        "imgsCam1 = torch.stack([transform_train(x[1]) for x in trainImagesCam1], dim=3)\n",
        "imgsCam2 = torch.stack([transform_train(x[1]) for x in trainImagesCam2], dim=3)\n",
        "meanCam1 = imgsCam1.view(3, -1).mean(dim=1)\n",
        "meanCam2 = imgsCam2.view(3, -1).mean(dim=1)\n",
        "stdCam1 = imgsCam1.view(3, -1).std(dim=1)\n",
        "stdCam2 = imgsCam2.view(3, -1).std(dim=1)"
      ],
      "execution_count": 206,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7sTKON2nl3-O"
      },
      "source": [
        "tranformNormCam1 = transforms.Normalize(meanCam1, stdCam1)\n",
        "tranformNormCam2 = transforms.Normalize(meanCam2, stdCam2)\n",
        "trainImagesCam1 = [[x[0], tranformNormCam1(transform_train(x[1]))] for x in trainImagesCam1]\n",
        "trainImagesCam2 = [[x[0], tranformNormCam2(transform_train(x[1]))] for x in trainImagesCam2]\n",
        "for x in trainImagesCam1:\n",
        "  trainImages[x[0]][2] = x[1]\n",
        "for x in trainImagesCam2:\n",
        "  trainImages[x[0]][2] = x[1]"
      ],
      "execution_count": 207,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-AvBeN29tapw"
      },
      "source": [
        "### Custom Dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zXTueh6YYtz_"
      },
      "source": [
        "class CustomDataset(Dataset):\n",
        "  def __init__(self, images, transform=None, target_transform=None):\n",
        "        self.images = images\n",
        "        self.transform = transform\n",
        "        self.target_transform = target_transform\n",
        "\n",
        "  def __len__(self):\n",
        "        return len(self.images)\n",
        "\n",
        "  def __getitem__(self, idx):\n",
        "        label = self.images[idx][0]\n",
        "        image = self.images[idx][2]\n",
        "        if self.transform:\n",
        "            image = self.transform(image)\n",
        "        if self.target_transform:\n",
        "            label = self.target_transform(label)\n",
        "        return image, label"
      ],
      "execution_count": 208,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QNnE-cku-j2L"
      },
      "source": [
        "dataset = CustomDataset(trainImages, target_transform=target_transform)\n"
      ],
      "execution_count": 209,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wIgEpMNOb_ae"
      },
      "source": [
        "### Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "K7oap08_cxXG"
      },
      "source": [
        "from torch.nn import Conv2d,Linear, Module, Sequential,BatchNorm2d,MaxPool2d\n",
        "import torch.nn.functional as F\n",
        "import torchvision.models as models\n",
        "from torchsummary import summary"
      ],
      "execution_count": 210,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "51ZFhV1IikYZ"
      },
      "source": [
        "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")"
      ],
      "execution_count": 211,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7gDNEuSda2Rq"
      },
      "source": [
        "class ResidualBlock(Module):\n",
        "  def __init__(self,in_channels,increase_dims=False,is_first=False):\n",
        "        super(ResidualBlock, self).__init__()\n",
        "        self.increase_dims = increase_dims\n",
        "        self.is_first = is_first\n",
        "        self.bn_0 = None\n",
        "        self.conv_up = None\n",
        "        out_channels = in_channels\n",
        "        stride = 1\n",
        "        if increase_dims:\n",
        "          out_channels*=2\n",
        "          stride*=2\n",
        "          # self.conv_up = nn.Conv2d(in_channels=in_channels,out_channels=out_channels,kernel_size=1,stride=2,padding=\"same\")\n",
        "          self.conv_up = nn.Conv2d(in_channels=in_channels,out_channels=out_channels,kernel_size=1,stride=2)\n",
        "        if not is_first:\n",
        "          self.bn_0 = nn.Sequential(\n",
        "              nn.BatchNorm2d(in_channels),\n",
        "              nn.ELU()\n",
        "          )\n",
        "        if increase_dims:\n",
        "          self.conv_1 = nn.Sequential(\n",
        "              # nn.Conv2d(in_channels=in_channels,out_channels=out_channels,kernel_size=3,stride=stride,padding=\"same\"),\n",
        "              nn.Conv2d(in_channels=in_channels,out_channels=out_channels,kernel_size=3,stride=stride),\n",
        "              nn.BatchNorm2d(out_channels),\n",
        "              nn.ELU(),\n",
        "              nn.Dropout(p=0.4)\n",
        "          )\n",
        "        else:\n",
        "          self.conv_1 = nn.Sequential(\n",
        "              nn.Conv2d(in_channels=in_channels,out_channels=out_channels,kernel_size=3,stride=stride,padding=\"same\"),\n",
        "              # nn.Conv2d(in_channels=in_channels,out_channels=out_channels,kernel_size=3,stride=stride),\n",
        "              nn.BatchNorm2d(out_channels),\n",
        "              nn.ELU(),\n",
        "              nn.Dropout(p=0.4)\n",
        "          )\n",
        "        self.conv_2 = nn.Conv2d(in_channels=out_channels,out_channels=out_channels,kernel_size=3,stride=1,padding=\"same\")\n",
        "  def forward(self,x):\n",
        "      if self.bn_0:\n",
        "        y = self.bn_0(x)\n",
        "      else:\n",
        "        y = x\n",
        "      residual = x\n",
        "      if self.increase_dims:\n",
        "        y = F.pad(y, (0, 1, 0, 1))\n",
        "      out = self.conv_1(y)\n",
        "      out = self.conv_2(out)\n",
        "      if self.conv_up:\n",
        "        residual = self.conv_up(residual)\n",
        "      return out+residual"
      ],
      "execution_count": 212,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "66pOGwF4b92z"
      },
      "source": [
        "class ReidModel1(Module):   \n",
        "    def __init__(self, numClasses = 2, inference = False):\n",
        "        super(ReidModel1, self).__init__()\n",
        "        self.inference = inference\n",
        "        self.numClasses = numClasses\n",
        "        self.conv_1 = nn.Sequential(\n",
        "          nn.Conv2d(in_channels=3,out_channels=32,kernel_size=3,stride=1,padding=\"same\"),\n",
        "          nn.BatchNorm2d(32),\n",
        "          nn.ELU()\n",
        "        )\n",
        "        self.conv_2 = nn.Sequential(\n",
        "          nn.Conv2d(in_channels=32,out_channels=32,kernel_size=3,stride=1,padding=\"same\"),\n",
        "          nn.BatchNorm2d(32),\n",
        "          nn.ELU()\n",
        "        )\n",
        "        # self.pool_3 = nn.MaxPool2d(kernel_size=3,stride=2,padding=\"same\")\n",
        "        self.pool_3 = nn.MaxPool2d(kernel_size=3,stride=2)\n",
        "        \n",
        "        self.residual_4 = ResidualBlock(in_channels=32,is_first=True)\n",
        "        self.residual_5 = ResidualBlock(in_channels=32)\n",
        "        self.residual_6 = ResidualBlock(in_channels=32,increase_dims=True)\n",
        "        self.residual_7 = ResidualBlock(in_channels=64)\n",
        "        self.residual_8 = ResidualBlock(in_channels=64,increase_dims=True)\n",
        "        self.residual_9 = ResidualBlock(in_channels=128)\n",
        "\n",
        "        self.dense_10 = nn.Sequential(\n",
        "            nn.Flatten(),\n",
        "            nn.Dropout(p=0.4),\n",
        "            nn.Linear(in_features=128*16*6,out_features=128),\n",
        "            nn.BatchNorm1d(128),\n",
        "            nn.ELU()\n",
        "        )\n",
        "        \n",
        "        self.cosine_layer = nn.utils.weight_norm(nn.Linear(in_features=128,out_features=numClasses,bias=False))\n",
        "\n",
        "    # Defining the forward pass    \n",
        "    def forward(self, x):\n",
        "        out = self.conv_1(x)\n",
        "        out = self.conv_2(out)\n",
        "        out = F.pad(out, (0, 1, 0, 1))\n",
        "        out = self.pool_3(out)\n",
        "        out = self.residual_4(out)\n",
        "        out = self.residual_5(out)\n",
        "        out = self.residual_6(out)\n",
        "        out = self.residual_7(out)\n",
        "        out = self.residual_8(out)\n",
        "        out = self.residual_9(out)\n",
        "        out = self.dense_10(out)\n",
        "        out = nn.functional.normalize(out,p=2,dim=1)\n",
        "        if (not self.inference):\n",
        "          out = self.cosine_layer(out)\n",
        "        # out = self.cosine_layer(out)\n",
        "        return out"
      ],
      "execution_count": 213,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jX74w8aWLr6d",
        "outputId": "29b429ff-9d71-4e4f-8bc7-3bc82c997915"
      },
      "source": [
        "base_model = models.resnet50(pretrained = True, progress= False)\n",
        "base_model = torch.nn.Sequential(*(list(base_model.children())[:-1]))\n",
        "base_model.to(device)\n",
        "summary(base_model, (3, 128, 48))"
      ],
      "execution_count": 214,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "----------------------------------------------------------------\n",
            "        Layer (type)               Output Shape         Param #\n",
            "================================================================\n",
            "            Conv2d-1           [-1, 64, 64, 24]           9,408\n",
            "       BatchNorm2d-2           [-1, 64, 64, 24]             128\n",
            "              ReLU-3           [-1, 64, 64, 24]               0\n",
            "         MaxPool2d-4           [-1, 64, 32, 12]               0\n",
            "            Conv2d-5           [-1, 64, 32, 12]           4,096\n",
            "       BatchNorm2d-6           [-1, 64, 32, 12]             128\n",
            "              ReLU-7           [-1, 64, 32, 12]               0\n",
            "            Conv2d-8           [-1, 64, 32, 12]          36,864\n",
            "       BatchNorm2d-9           [-1, 64, 32, 12]             128\n",
            "             ReLU-10           [-1, 64, 32, 12]               0\n",
            "           Conv2d-11          [-1, 256, 32, 12]          16,384\n",
            "      BatchNorm2d-12          [-1, 256, 32, 12]             512\n",
            "           Conv2d-13          [-1, 256, 32, 12]          16,384\n",
            "      BatchNorm2d-14          [-1, 256, 32, 12]             512\n",
            "             ReLU-15          [-1, 256, 32, 12]               0\n",
            "       Bottleneck-16          [-1, 256, 32, 12]               0\n",
            "           Conv2d-17           [-1, 64, 32, 12]          16,384\n",
            "      BatchNorm2d-18           [-1, 64, 32, 12]             128\n",
            "             ReLU-19           [-1, 64, 32, 12]               0\n",
            "           Conv2d-20           [-1, 64, 32, 12]          36,864\n",
            "      BatchNorm2d-21           [-1, 64, 32, 12]             128\n",
            "             ReLU-22           [-1, 64, 32, 12]               0\n",
            "           Conv2d-23          [-1, 256, 32, 12]          16,384\n",
            "      BatchNorm2d-24          [-1, 256, 32, 12]             512\n",
            "             ReLU-25          [-1, 256, 32, 12]               0\n",
            "       Bottleneck-26          [-1, 256, 32, 12]               0\n",
            "           Conv2d-27           [-1, 64, 32, 12]          16,384\n",
            "      BatchNorm2d-28           [-1, 64, 32, 12]             128\n",
            "             ReLU-29           [-1, 64, 32, 12]               0\n",
            "           Conv2d-30           [-1, 64, 32, 12]          36,864\n",
            "      BatchNorm2d-31           [-1, 64, 32, 12]             128\n",
            "             ReLU-32           [-1, 64, 32, 12]               0\n",
            "           Conv2d-33          [-1, 256, 32, 12]          16,384\n",
            "      BatchNorm2d-34          [-1, 256, 32, 12]             512\n",
            "             ReLU-35          [-1, 256, 32, 12]               0\n",
            "       Bottleneck-36          [-1, 256, 32, 12]               0\n",
            "           Conv2d-37          [-1, 128, 32, 12]          32,768\n",
            "      BatchNorm2d-38          [-1, 128, 32, 12]             256\n",
            "             ReLU-39          [-1, 128, 32, 12]               0\n",
            "           Conv2d-40           [-1, 128, 16, 6]         147,456\n",
            "      BatchNorm2d-41           [-1, 128, 16, 6]             256\n",
            "             ReLU-42           [-1, 128, 16, 6]               0\n",
            "           Conv2d-43           [-1, 512, 16, 6]          65,536\n",
            "      BatchNorm2d-44           [-1, 512, 16, 6]           1,024\n",
            "           Conv2d-45           [-1, 512, 16, 6]         131,072\n",
            "      BatchNorm2d-46           [-1, 512, 16, 6]           1,024\n",
            "             ReLU-47           [-1, 512, 16, 6]               0\n",
            "       Bottleneck-48           [-1, 512, 16, 6]               0\n",
            "           Conv2d-49           [-1, 128, 16, 6]          65,536\n",
            "      BatchNorm2d-50           [-1, 128, 16, 6]             256\n",
            "             ReLU-51           [-1, 128, 16, 6]               0\n",
            "           Conv2d-52           [-1, 128, 16, 6]         147,456\n",
            "      BatchNorm2d-53           [-1, 128, 16, 6]             256\n",
            "             ReLU-54           [-1, 128, 16, 6]               0\n",
            "           Conv2d-55           [-1, 512, 16, 6]          65,536\n",
            "      BatchNorm2d-56           [-1, 512, 16, 6]           1,024\n",
            "             ReLU-57           [-1, 512, 16, 6]               0\n",
            "       Bottleneck-58           [-1, 512, 16, 6]               0\n",
            "           Conv2d-59           [-1, 128, 16, 6]          65,536\n",
            "      BatchNorm2d-60           [-1, 128, 16, 6]             256\n",
            "             ReLU-61           [-1, 128, 16, 6]               0\n",
            "           Conv2d-62           [-1, 128, 16, 6]         147,456\n",
            "      BatchNorm2d-63           [-1, 128, 16, 6]             256\n",
            "             ReLU-64           [-1, 128, 16, 6]               0\n",
            "           Conv2d-65           [-1, 512, 16, 6]          65,536\n",
            "      BatchNorm2d-66           [-1, 512, 16, 6]           1,024\n",
            "             ReLU-67           [-1, 512, 16, 6]               0\n",
            "       Bottleneck-68           [-1, 512, 16, 6]               0\n",
            "           Conv2d-69           [-1, 128, 16, 6]          65,536\n",
            "      BatchNorm2d-70           [-1, 128, 16, 6]             256\n",
            "             ReLU-71           [-1, 128, 16, 6]               0\n",
            "           Conv2d-72           [-1, 128, 16, 6]         147,456\n",
            "      BatchNorm2d-73           [-1, 128, 16, 6]             256\n",
            "             ReLU-74           [-1, 128, 16, 6]               0\n",
            "           Conv2d-75           [-1, 512, 16, 6]          65,536\n",
            "      BatchNorm2d-76           [-1, 512, 16, 6]           1,024\n",
            "             ReLU-77           [-1, 512, 16, 6]               0\n",
            "       Bottleneck-78           [-1, 512, 16, 6]               0\n",
            "           Conv2d-79           [-1, 256, 16, 6]         131,072\n",
            "      BatchNorm2d-80           [-1, 256, 16, 6]             512\n",
            "             ReLU-81           [-1, 256, 16, 6]               0\n",
            "           Conv2d-82            [-1, 256, 8, 3]         589,824\n",
            "      BatchNorm2d-83            [-1, 256, 8, 3]             512\n",
            "             ReLU-84            [-1, 256, 8, 3]               0\n",
            "           Conv2d-85           [-1, 1024, 8, 3]         262,144\n",
            "      BatchNorm2d-86           [-1, 1024, 8, 3]           2,048\n",
            "           Conv2d-87           [-1, 1024, 8, 3]         524,288\n",
            "      BatchNorm2d-88           [-1, 1024, 8, 3]           2,048\n",
            "             ReLU-89           [-1, 1024, 8, 3]               0\n",
            "       Bottleneck-90           [-1, 1024, 8, 3]               0\n",
            "           Conv2d-91            [-1, 256, 8, 3]         262,144\n",
            "      BatchNorm2d-92            [-1, 256, 8, 3]             512\n",
            "             ReLU-93            [-1, 256, 8, 3]               0\n",
            "           Conv2d-94            [-1, 256, 8, 3]         589,824\n",
            "      BatchNorm2d-95            [-1, 256, 8, 3]             512\n",
            "             ReLU-96            [-1, 256, 8, 3]               0\n",
            "           Conv2d-97           [-1, 1024, 8, 3]         262,144\n",
            "      BatchNorm2d-98           [-1, 1024, 8, 3]           2,048\n",
            "             ReLU-99           [-1, 1024, 8, 3]               0\n",
            "      Bottleneck-100           [-1, 1024, 8, 3]               0\n",
            "          Conv2d-101            [-1, 256, 8, 3]         262,144\n",
            "     BatchNorm2d-102            [-1, 256, 8, 3]             512\n",
            "            ReLU-103            [-1, 256, 8, 3]               0\n",
            "          Conv2d-104            [-1, 256, 8, 3]         589,824\n",
            "     BatchNorm2d-105            [-1, 256, 8, 3]             512\n",
            "            ReLU-106            [-1, 256, 8, 3]               0\n",
            "          Conv2d-107           [-1, 1024, 8, 3]         262,144\n",
            "     BatchNorm2d-108           [-1, 1024, 8, 3]           2,048\n",
            "            ReLU-109           [-1, 1024, 8, 3]               0\n",
            "      Bottleneck-110           [-1, 1024, 8, 3]               0\n",
            "          Conv2d-111            [-1, 256, 8, 3]         262,144\n",
            "     BatchNorm2d-112            [-1, 256, 8, 3]             512\n",
            "            ReLU-113            [-1, 256, 8, 3]               0\n",
            "          Conv2d-114            [-1, 256, 8, 3]         589,824\n",
            "     BatchNorm2d-115            [-1, 256, 8, 3]             512\n",
            "            ReLU-116            [-1, 256, 8, 3]               0\n",
            "          Conv2d-117           [-1, 1024, 8, 3]         262,144\n",
            "     BatchNorm2d-118           [-1, 1024, 8, 3]           2,048\n",
            "            ReLU-119           [-1, 1024, 8, 3]               0\n",
            "      Bottleneck-120           [-1, 1024, 8, 3]               0\n",
            "          Conv2d-121            [-1, 256, 8, 3]         262,144\n",
            "     BatchNorm2d-122            [-1, 256, 8, 3]             512\n",
            "            ReLU-123            [-1, 256, 8, 3]               0\n",
            "          Conv2d-124            [-1, 256, 8, 3]         589,824\n",
            "     BatchNorm2d-125            [-1, 256, 8, 3]             512\n",
            "            ReLU-126            [-1, 256, 8, 3]               0\n",
            "          Conv2d-127           [-1, 1024, 8, 3]         262,144\n",
            "     BatchNorm2d-128           [-1, 1024, 8, 3]           2,048\n",
            "            ReLU-129           [-1, 1024, 8, 3]               0\n",
            "      Bottleneck-130           [-1, 1024, 8, 3]               0\n",
            "          Conv2d-131            [-1, 256, 8, 3]         262,144\n",
            "     BatchNorm2d-132            [-1, 256, 8, 3]             512\n",
            "            ReLU-133            [-1, 256, 8, 3]               0\n",
            "          Conv2d-134            [-1, 256, 8, 3]         589,824\n",
            "     BatchNorm2d-135            [-1, 256, 8, 3]             512\n",
            "            ReLU-136            [-1, 256, 8, 3]               0\n",
            "          Conv2d-137           [-1, 1024, 8, 3]         262,144\n",
            "     BatchNorm2d-138           [-1, 1024, 8, 3]           2,048\n",
            "            ReLU-139           [-1, 1024, 8, 3]               0\n",
            "      Bottleneck-140           [-1, 1024, 8, 3]               0\n",
            "          Conv2d-141            [-1, 512, 8, 3]         524,288\n",
            "     BatchNorm2d-142            [-1, 512, 8, 3]           1,024\n",
            "            ReLU-143            [-1, 512, 8, 3]               0\n",
            "          Conv2d-144            [-1, 512, 4, 2]       2,359,296\n",
            "     BatchNorm2d-145            [-1, 512, 4, 2]           1,024\n",
            "            ReLU-146            [-1, 512, 4, 2]               0\n",
            "          Conv2d-147           [-1, 2048, 4, 2]       1,048,576\n",
            "     BatchNorm2d-148           [-1, 2048, 4, 2]           4,096\n",
            "          Conv2d-149           [-1, 2048, 4, 2]       2,097,152\n",
            "     BatchNorm2d-150           [-1, 2048, 4, 2]           4,096\n",
            "            ReLU-151           [-1, 2048, 4, 2]               0\n",
            "      Bottleneck-152           [-1, 2048, 4, 2]               0\n",
            "          Conv2d-153            [-1, 512, 4, 2]       1,048,576\n",
            "     BatchNorm2d-154            [-1, 512, 4, 2]           1,024\n",
            "            ReLU-155            [-1, 512, 4, 2]               0\n",
            "          Conv2d-156            [-1, 512, 4, 2]       2,359,296\n",
            "     BatchNorm2d-157            [-1, 512, 4, 2]           1,024\n",
            "            ReLU-158            [-1, 512, 4, 2]               0\n",
            "          Conv2d-159           [-1, 2048, 4, 2]       1,048,576\n",
            "     BatchNorm2d-160           [-1, 2048, 4, 2]           4,096\n",
            "            ReLU-161           [-1, 2048, 4, 2]               0\n",
            "      Bottleneck-162           [-1, 2048, 4, 2]               0\n",
            "          Conv2d-163            [-1, 512, 4, 2]       1,048,576\n",
            "     BatchNorm2d-164            [-1, 512, 4, 2]           1,024\n",
            "            ReLU-165            [-1, 512, 4, 2]               0\n",
            "          Conv2d-166            [-1, 512, 4, 2]       2,359,296\n",
            "     BatchNorm2d-167            [-1, 512, 4, 2]           1,024\n",
            "            ReLU-168            [-1, 512, 4, 2]               0\n",
            "          Conv2d-169           [-1, 2048, 4, 2]       1,048,576\n",
            "     BatchNorm2d-170           [-1, 2048, 4, 2]           4,096\n",
            "            ReLU-171           [-1, 2048, 4, 2]               0\n",
            "      Bottleneck-172           [-1, 2048, 4, 2]               0\n",
            "AdaptiveAvgPool2d-173           [-1, 2048, 1, 1]               0\n",
            "================================================================\n",
            "Total params: 23,508,032\n",
            "Trainable params: 23,508,032\n",
            "Non-trainable params: 0\n",
            "----------------------------------------------------------------\n",
            "Input size (MB): 0.07\n",
            "Forward/backward pass size (MB): 35.66\n",
            "Params size (MB): 89.68\n",
            "Estimated Total Size (MB): 125.40\n",
            "----------------------------------------------------------------\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hIRW2rFOKOwG"
      },
      "source": [
        "class ReidModelResnet50(Module):\n",
        "  def __init__(self, numClasses = 2, inference = False):\n",
        "    super(ReidModelResnet50, self).__init__()\n",
        "    self.inference = inference\n",
        "    self.numClasses = numClasses\n",
        "    # load pretrained resnet50\n",
        "    base_model = models.resnet50(pretrained = True, progress= False)\n",
        "    base_model = torch.nn.Sequential(*(list(base_model.children())[:-1]))\n",
        "    # freeze weights\n",
        "    num_layers = len(list(base_model.children()))\n",
        "    ct = 0\n",
        "    for child in base_model.children():\n",
        "      ct += 1\n",
        "      if ct < (num_layers-1):\n",
        "          for param in child.parameters():\n",
        "              param.requires_grad = False\n",
        "    self.base_model = base_model\n",
        "    self.final_layer = nn.Sequential(\n",
        "            nn.Flatten(),\n",
        "            nn.Dropout(p=0.4),\n",
        "            nn.Linear(in_features=2048,out_features=128),\n",
        "            nn.BatchNorm1d(128),\n",
        "            nn.ELU()\n",
        "        )\n",
        "    self.cosine_layer = nn.utils.weight_norm(nn.Linear(in_features=128,out_features=numClasses,bias=False))\n",
        "  def forward(self, x):\n",
        "    out = self.base_model(x)\n",
        "    # out = torch.squeeze(out, dim = 3)\n",
        "    # out = torch.squeeze(out, dim = 2)\n",
        "    out = self.final_layer(out)\n",
        "    out = nn.functional.normalize(out,p=2,dim=1)\n",
        "    if (not self.inference):\n",
        "      out = self.cosine_layer(out)\n",
        "    return out\n",
        "\n"
      ],
      "execution_count": 215,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C24i_9npgVm0",
        "outputId": "6ecaa039-78e3-4805-abbf-4110c1c83963"
      },
      "source": [
        "model = ReidModelResnet50()\n",
        "model.to(device)\n",
        "summary(model, (3, 128, 48))"
      ],
      "execution_count": 216,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "----------------------------------------------------------------\n",
            "        Layer (type)               Output Shape         Param #\n",
            "================================================================\n",
            "            Conv2d-1           [-1, 64, 64, 24]           9,408\n",
            "       BatchNorm2d-2           [-1, 64, 64, 24]             128\n",
            "              ReLU-3           [-1, 64, 64, 24]               0\n",
            "         MaxPool2d-4           [-1, 64, 32, 12]               0\n",
            "            Conv2d-5           [-1, 64, 32, 12]           4,096\n",
            "       BatchNorm2d-6           [-1, 64, 32, 12]             128\n",
            "              ReLU-7           [-1, 64, 32, 12]               0\n",
            "            Conv2d-8           [-1, 64, 32, 12]          36,864\n",
            "       BatchNorm2d-9           [-1, 64, 32, 12]             128\n",
            "             ReLU-10           [-1, 64, 32, 12]               0\n",
            "           Conv2d-11          [-1, 256, 32, 12]          16,384\n",
            "      BatchNorm2d-12          [-1, 256, 32, 12]             512\n",
            "           Conv2d-13          [-1, 256, 32, 12]          16,384\n",
            "      BatchNorm2d-14          [-1, 256, 32, 12]             512\n",
            "             ReLU-15          [-1, 256, 32, 12]               0\n",
            "       Bottleneck-16          [-1, 256, 32, 12]               0\n",
            "           Conv2d-17           [-1, 64, 32, 12]          16,384\n",
            "      BatchNorm2d-18           [-1, 64, 32, 12]             128\n",
            "             ReLU-19           [-1, 64, 32, 12]               0\n",
            "           Conv2d-20           [-1, 64, 32, 12]          36,864\n",
            "      BatchNorm2d-21           [-1, 64, 32, 12]             128\n",
            "             ReLU-22           [-1, 64, 32, 12]               0\n",
            "           Conv2d-23          [-1, 256, 32, 12]          16,384\n",
            "      BatchNorm2d-24          [-1, 256, 32, 12]             512\n",
            "             ReLU-25          [-1, 256, 32, 12]               0\n",
            "       Bottleneck-26          [-1, 256, 32, 12]               0\n",
            "           Conv2d-27           [-1, 64, 32, 12]          16,384\n",
            "      BatchNorm2d-28           [-1, 64, 32, 12]             128\n",
            "             ReLU-29           [-1, 64, 32, 12]               0\n",
            "           Conv2d-30           [-1, 64, 32, 12]          36,864\n",
            "      BatchNorm2d-31           [-1, 64, 32, 12]             128\n",
            "             ReLU-32           [-1, 64, 32, 12]               0\n",
            "           Conv2d-33          [-1, 256, 32, 12]          16,384\n",
            "      BatchNorm2d-34          [-1, 256, 32, 12]             512\n",
            "             ReLU-35          [-1, 256, 32, 12]               0\n",
            "       Bottleneck-36          [-1, 256, 32, 12]               0\n",
            "           Conv2d-37          [-1, 128, 32, 12]          32,768\n",
            "      BatchNorm2d-38          [-1, 128, 32, 12]             256\n",
            "             ReLU-39          [-1, 128, 32, 12]               0\n",
            "           Conv2d-40           [-1, 128, 16, 6]         147,456\n",
            "      BatchNorm2d-41           [-1, 128, 16, 6]             256\n",
            "             ReLU-42           [-1, 128, 16, 6]               0\n",
            "           Conv2d-43           [-1, 512, 16, 6]          65,536\n",
            "      BatchNorm2d-44           [-1, 512, 16, 6]           1,024\n",
            "           Conv2d-45           [-1, 512, 16, 6]         131,072\n",
            "      BatchNorm2d-46           [-1, 512, 16, 6]           1,024\n",
            "             ReLU-47           [-1, 512, 16, 6]               0\n",
            "       Bottleneck-48           [-1, 512, 16, 6]               0\n",
            "           Conv2d-49           [-1, 128, 16, 6]          65,536\n",
            "      BatchNorm2d-50           [-1, 128, 16, 6]             256\n",
            "             ReLU-51           [-1, 128, 16, 6]               0\n",
            "           Conv2d-52           [-1, 128, 16, 6]         147,456\n",
            "      BatchNorm2d-53           [-1, 128, 16, 6]             256\n",
            "             ReLU-54           [-1, 128, 16, 6]               0\n",
            "           Conv2d-55           [-1, 512, 16, 6]          65,536\n",
            "      BatchNorm2d-56           [-1, 512, 16, 6]           1,024\n",
            "             ReLU-57           [-1, 512, 16, 6]               0\n",
            "       Bottleneck-58           [-1, 512, 16, 6]               0\n",
            "           Conv2d-59           [-1, 128, 16, 6]          65,536\n",
            "      BatchNorm2d-60           [-1, 128, 16, 6]             256\n",
            "             ReLU-61           [-1, 128, 16, 6]               0\n",
            "           Conv2d-62           [-1, 128, 16, 6]         147,456\n",
            "      BatchNorm2d-63           [-1, 128, 16, 6]             256\n",
            "             ReLU-64           [-1, 128, 16, 6]               0\n",
            "           Conv2d-65           [-1, 512, 16, 6]          65,536\n",
            "      BatchNorm2d-66           [-1, 512, 16, 6]           1,024\n",
            "             ReLU-67           [-1, 512, 16, 6]               0\n",
            "       Bottleneck-68           [-1, 512, 16, 6]               0\n",
            "           Conv2d-69           [-1, 128, 16, 6]          65,536\n",
            "      BatchNorm2d-70           [-1, 128, 16, 6]             256\n",
            "             ReLU-71           [-1, 128, 16, 6]               0\n",
            "           Conv2d-72           [-1, 128, 16, 6]         147,456\n",
            "      BatchNorm2d-73           [-1, 128, 16, 6]             256\n",
            "             ReLU-74           [-1, 128, 16, 6]               0\n",
            "           Conv2d-75           [-1, 512, 16, 6]          65,536\n",
            "      BatchNorm2d-76           [-1, 512, 16, 6]           1,024\n",
            "             ReLU-77           [-1, 512, 16, 6]               0\n",
            "       Bottleneck-78           [-1, 512, 16, 6]               0\n",
            "           Conv2d-79           [-1, 256, 16, 6]         131,072\n",
            "      BatchNorm2d-80           [-1, 256, 16, 6]             512\n",
            "             ReLU-81           [-1, 256, 16, 6]               0\n",
            "           Conv2d-82            [-1, 256, 8, 3]         589,824\n",
            "      BatchNorm2d-83            [-1, 256, 8, 3]             512\n",
            "             ReLU-84            [-1, 256, 8, 3]               0\n",
            "           Conv2d-85           [-1, 1024, 8, 3]         262,144\n",
            "      BatchNorm2d-86           [-1, 1024, 8, 3]           2,048\n",
            "           Conv2d-87           [-1, 1024, 8, 3]         524,288\n",
            "      BatchNorm2d-88           [-1, 1024, 8, 3]           2,048\n",
            "             ReLU-89           [-1, 1024, 8, 3]               0\n",
            "       Bottleneck-90           [-1, 1024, 8, 3]               0\n",
            "           Conv2d-91            [-1, 256, 8, 3]         262,144\n",
            "      BatchNorm2d-92            [-1, 256, 8, 3]             512\n",
            "             ReLU-93            [-1, 256, 8, 3]               0\n",
            "           Conv2d-94            [-1, 256, 8, 3]         589,824\n",
            "      BatchNorm2d-95            [-1, 256, 8, 3]             512\n",
            "             ReLU-96            [-1, 256, 8, 3]               0\n",
            "           Conv2d-97           [-1, 1024, 8, 3]         262,144\n",
            "      BatchNorm2d-98           [-1, 1024, 8, 3]           2,048\n",
            "             ReLU-99           [-1, 1024, 8, 3]               0\n",
            "      Bottleneck-100           [-1, 1024, 8, 3]               0\n",
            "          Conv2d-101            [-1, 256, 8, 3]         262,144\n",
            "     BatchNorm2d-102            [-1, 256, 8, 3]             512\n",
            "            ReLU-103            [-1, 256, 8, 3]               0\n",
            "          Conv2d-104            [-1, 256, 8, 3]         589,824\n",
            "     BatchNorm2d-105            [-1, 256, 8, 3]             512\n",
            "            ReLU-106            [-1, 256, 8, 3]               0\n",
            "          Conv2d-107           [-1, 1024, 8, 3]         262,144\n",
            "     BatchNorm2d-108           [-1, 1024, 8, 3]           2,048\n",
            "            ReLU-109           [-1, 1024, 8, 3]               0\n",
            "      Bottleneck-110           [-1, 1024, 8, 3]               0\n",
            "          Conv2d-111            [-1, 256, 8, 3]         262,144\n",
            "     BatchNorm2d-112            [-1, 256, 8, 3]             512\n",
            "            ReLU-113            [-1, 256, 8, 3]               0\n",
            "          Conv2d-114            [-1, 256, 8, 3]         589,824\n",
            "     BatchNorm2d-115            [-1, 256, 8, 3]             512\n",
            "            ReLU-116            [-1, 256, 8, 3]               0\n",
            "          Conv2d-117           [-1, 1024, 8, 3]         262,144\n",
            "     BatchNorm2d-118           [-1, 1024, 8, 3]           2,048\n",
            "            ReLU-119           [-1, 1024, 8, 3]               0\n",
            "      Bottleneck-120           [-1, 1024, 8, 3]               0\n",
            "          Conv2d-121            [-1, 256, 8, 3]         262,144\n",
            "     BatchNorm2d-122            [-1, 256, 8, 3]             512\n",
            "            ReLU-123            [-1, 256, 8, 3]               0\n",
            "          Conv2d-124            [-1, 256, 8, 3]         589,824\n",
            "     BatchNorm2d-125            [-1, 256, 8, 3]             512\n",
            "            ReLU-126            [-1, 256, 8, 3]               0\n",
            "          Conv2d-127           [-1, 1024, 8, 3]         262,144\n",
            "     BatchNorm2d-128           [-1, 1024, 8, 3]           2,048\n",
            "            ReLU-129           [-1, 1024, 8, 3]               0\n",
            "      Bottleneck-130           [-1, 1024, 8, 3]               0\n",
            "          Conv2d-131            [-1, 256, 8, 3]         262,144\n",
            "     BatchNorm2d-132            [-1, 256, 8, 3]             512\n",
            "            ReLU-133            [-1, 256, 8, 3]               0\n",
            "          Conv2d-134            [-1, 256, 8, 3]         589,824\n",
            "     BatchNorm2d-135            [-1, 256, 8, 3]             512\n",
            "            ReLU-136            [-1, 256, 8, 3]               0\n",
            "          Conv2d-137           [-1, 1024, 8, 3]         262,144\n",
            "     BatchNorm2d-138           [-1, 1024, 8, 3]           2,048\n",
            "            ReLU-139           [-1, 1024, 8, 3]               0\n",
            "      Bottleneck-140           [-1, 1024, 8, 3]               0\n",
            "          Conv2d-141            [-1, 512, 8, 3]         524,288\n",
            "     BatchNorm2d-142            [-1, 512, 8, 3]           1,024\n",
            "            ReLU-143            [-1, 512, 8, 3]               0\n",
            "          Conv2d-144            [-1, 512, 4, 2]       2,359,296\n",
            "     BatchNorm2d-145            [-1, 512, 4, 2]           1,024\n",
            "            ReLU-146            [-1, 512, 4, 2]               0\n",
            "          Conv2d-147           [-1, 2048, 4, 2]       1,048,576\n",
            "     BatchNorm2d-148           [-1, 2048, 4, 2]           4,096\n",
            "          Conv2d-149           [-1, 2048, 4, 2]       2,097,152\n",
            "     BatchNorm2d-150           [-1, 2048, 4, 2]           4,096\n",
            "            ReLU-151           [-1, 2048, 4, 2]               0\n",
            "      Bottleneck-152           [-1, 2048, 4, 2]               0\n",
            "          Conv2d-153            [-1, 512, 4, 2]       1,048,576\n",
            "     BatchNorm2d-154            [-1, 512, 4, 2]           1,024\n",
            "            ReLU-155            [-1, 512, 4, 2]               0\n",
            "          Conv2d-156            [-1, 512, 4, 2]       2,359,296\n",
            "     BatchNorm2d-157            [-1, 512, 4, 2]           1,024\n",
            "            ReLU-158            [-1, 512, 4, 2]               0\n",
            "          Conv2d-159           [-1, 2048, 4, 2]       1,048,576\n",
            "     BatchNorm2d-160           [-1, 2048, 4, 2]           4,096\n",
            "            ReLU-161           [-1, 2048, 4, 2]               0\n",
            "      Bottleneck-162           [-1, 2048, 4, 2]               0\n",
            "          Conv2d-163            [-1, 512, 4, 2]       1,048,576\n",
            "     BatchNorm2d-164            [-1, 512, 4, 2]           1,024\n",
            "            ReLU-165            [-1, 512, 4, 2]               0\n",
            "          Conv2d-166            [-1, 512, 4, 2]       2,359,296\n",
            "     BatchNorm2d-167            [-1, 512, 4, 2]           1,024\n",
            "            ReLU-168            [-1, 512, 4, 2]               0\n",
            "          Conv2d-169           [-1, 2048, 4, 2]       1,048,576\n",
            "     BatchNorm2d-170           [-1, 2048, 4, 2]           4,096\n",
            "            ReLU-171           [-1, 2048, 4, 2]               0\n",
            "      Bottleneck-172           [-1, 2048, 4, 2]               0\n",
            "AdaptiveAvgPool2d-173           [-1, 2048, 1, 1]               0\n",
            "         Flatten-174                 [-1, 2048]               0\n",
            "         Dropout-175                 [-1, 2048]               0\n",
            "          Linear-176                  [-1, 128]         262,272\n",
            "     BatchNorm1d-177                  [-1, 128]             256\n",
            "             ELU-178                  [-1, 128]               0\n",
            "          Linear-179                    [-1, 2]             256\n",
            "================================================================\n",
            "Total params: 23,770,816\n",
            "Trainable params: 15,227,520\n",
            "Non-trainable params: 8,543,296\n",
            "----------------------------------------------------------------\n",
            "Input size (MB): 0.07\n",
            "Forward/backward pass size (MB): 35.69\n",
            "Params size (MB): 90.68\n",
            "Estimated Total Size (MB): 126.44\n",
            "----------------------------------------------------------------\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vq5FlpXnoLXH"
      },
      "source": [
        "### Training"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2R31ZIOvoLXH"
      },
      "source": [
        "from torch.utils.data import DataLoader, ConcatDataset\n",
        "from torch.optim import Adam\n",
        "from sklearn.model_selection import KFold"
      ],
      "execution_count": 217,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ac9DlN9noLXH",
        "outputId": "ce853efb-0ece-471a-9e7e-6c7d96b757e9"
      },
      "source": [
        "k_folds = 5\n",
        "num_epochs = 200\n",
        "batch_size = 16\n",
        "torch.manual_seed(42)"
      ],
      "execution_count": 238,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<torch._C.Generator at 0x7fcf74384650>"
            ]
          },
          "metadata": {},
          "execution_count": 238
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k0I-tc4-oLXI"
      },
      "source": [
        "def reset_weights(m):\n",
        "  '''\n",
        "    Try resetting model weights to avoid\n",
        "    weight leakage.\n",
        "  '''\n",
        "  for layer in m.children():\n",
        "    if hasattr(layer, 'reset_parameters'):\n",
        "      # print(f'Reset trainable parameters of layer = {layer}')\n",
        "      layer.reset_parameters()"
      ],
      "execution_count": 219,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0JoTRDGLoLXI"
      },
      "source": [
        "Use GPU if available"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fGvJqOf1oLXI"
      },
      "source": [
        "def reset_net():\n",
        "  global net\n",
        "  net = ReidModel1(numClasses = 62, inference = False)\n",
        "  net.to(device)\n",
        "reset_net()"
      ],
      "execution_count": 220,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cozg-A9moLXI"
      },
      "source": [
        "def train_epoch(trainDataset,criterion,optimizer):\n",
        "\n",
        "    num_minibatches = len(trainDataset)//batch_size\n",
        "    running_loss = 0.0\n",
        "    for i, data in enumerate(trainDataset, 0):\n",
        "        # get the inputs; data is a list of [inputs, labels]\n",
        "        inputs, labels = data[0].to(device), data[1].to(device)\n",
        "\n",
        "        # zero the parameter gradients\n",
        "        optimizer.zero_grad()\n",
        "\n",
        "        # forward + backward + optimize\n",
        "        outputs = net(inputs)\n",
        "        loss = criterion(outputs, labels)\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "\n",
        "        # print statistics\n",
        "        running_loss += loss.item()\n",
        "        # if i % (num_minibatches//5) == (num_minibatches//5-1):    # print every 100 mini-batches\n",
        "        #     print('[%d, %5d] loss: %.3f' %\n",
        "        #           (epoch + 1, i + 1, running_loss / (num_minibatches//5)))\n",
        "        #     running_loss = 0.0\n",
        "    return running_loss"
      ],
      "execution_count": 221,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KIN7ep2DoLXI"
      },
      "source": [
        "#### Test Set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DbSoZ40ioLXI"
      },
      "source": [
        "def test(testDataset,criterion):\n",
        "    correct = 0\n",
        "    loss = 0\n",
        "    total = 0\n",
        "    # since we're not training, we don't need to calculate the gradients for our outputs\n",
        "    with torch.no_grad():\n",
        "        for data in testDataset:\n",
        "            images, labels = data[0].to(device), data[1].to(device)\n",
        "            # calculate outputs by running images through the network\n",
        "            y_pred = net(images)\n",
        "\n",
        "            # cross entropy loss\n",
        "            loss += criterion(y_pred,labels).item()\n",
        "\n",
        "            # get labels\n",
        "            y_pred_softmax = torch.log_softmax(y_pred, dim = 1)\n",
        "            # the class with the highest energy is what we choose as prediction\n",
        "            _, predicted = torch.max(y_pred_softmax, dim = 1) \n",
        "            _, labels = torch.max(labels, dim = 1) \n",
        "\n",
        "            # get accuracy\n",
        "            correct += (predicted == labels).sum().item()\n",
        "\n",
        "            # total items\n",
        "            total += labels.size(0)\n",
        "\n",
        "    return loss,100 * correct / total"
      ],
      "execution_count": 222,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ch0rKW-toLXI"
      },
      "source": [
        "#### KFold"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S8m1wmG3dWbB"
      },
      "source": [
        "from sklearn.model_selection import KFold"
      ],
      "execution_count": 223,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Q8y60F3nddlH"
      },
      "source": [
        "kfold = KFold(n_splits=k_folds,shuffle=True,random_state=42)"
      ],
      "execution_count": 134,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 907
        },
        "id": "LMO8_bbCoLXJ",
        "outputId": "54e4d960-173a-4477-a621-f9f5f6503c00"
      },
      "source": [
        "results = {}\n",
        "# K-fold Cross Validation model evaluation\n",
        "for fold, (train_ids, test_ids) in enumerate(kfold.split(dataset)):\n",
        "  \n",
        "  # Print\n",
        "  print(f'FOLD {fold}')\n",
        "  print('--------------------------------')\n",
        "  \n",
        "  # Sample elements randomly from a given list of ids, no replacement.\n",
        "  train_subsampler = torch.utils.data.SubsetRandomSampler(train_ids)\n",
        "  test_subsampler = torch.utils.data.SubsetRandomSampler(test_ids)\n",
        "  \n",
        "  # Define data loaders for training and testing data in this fold\n",
        "  trainloader = torch.utils.data.DataLoader(\n",
        "                    dataset, \n",
        "                    batch_size=batch_size, sampler=train_subsampler)\n",
        "  testloader = torch.utils.data.DataLoader(\n",
        "                    dataset,\n",
        "                    batch_size=batch_size, sampler=test_subsampler)\n",
        "  \n",
        "  # reset\n",
        "  reset_net()\n",
        "  criterion = nn.CrossEntropyLoss()\n",
        "  optimizer = Adam(net.parameters())\n",
        "  # train\n",
        "  for epoch in range(num_epochs):\n",
        "    loss = train_epoch(trainloader,criterion,optimizer)\n",
        "    val_loss,val_acc = test(testloader,criterion)\n",
        "    print(f'Epoch\\t{epoch+1}\\tloss:{loss:.4f}\\tval_loss:{val_loss:.4f}\\tval_acc:{val_acc:.4f}')\n",
        "  # test\n",
        "  results[fold] = test(testloader,criterion)[1]\n",
        "  print('--------------------------------')\n",
        "\n",
        "print('--------------------------------')\n",
        "sum = 0.0\n",
        "for key, value in results.items():\n",
        "  print(f'Fold {key}: {value} %')\n",
        "  sum += value\n",
        "print(f'Average: {sum/len(results.keys())} %')"
      ],
      "execution_count": 226,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "FOLD 0\n",
            "--------------------------------\n",
            "Epoch\t1\tloss:200.4371\tval_loss:51.1931\tval_acc:35.6784\n",
            "Epoch\t2\tloss:191.2545\tval_loss:49.7381\tval_acc:50.2513\n",
            "Epoch\t3\tloss:185.8997\tval_loss:48.8883\tval_acc:58.7940\n",
            "Epoch\t4\tloss:182.4201\tval_loss:48.1127\tval_acc:66.3317\n",
            "Epoch\t5\tloss:179.6453\tval_loss:47.5720\tval_acc:66.8342\n",
            "Epoch\t6\tloss:177.6343\tval_loss:47.1196\tval_acc:67.8392\n",
            "Epoch\t7\tloss:175.7819\tval_loss:46.6306\tval_acc:73.8693\n",
            "Epoch\t8\tloss:174.3770\tval_loss:46.4068\tval_acc:72.3618\n",
            "Epoch\t9\tloss:172.8434\tval_loss:46.2313\tval_acc:73.3668\n",
            "Epoch\t10\tloss:172.0626\tval_loss:45.8350\tval_acc:74.3719\n",
            "Epoch\t11\tloss:170.3403\tval_loss:45.5539\tval_acc:74.8744\n",
            "Epoch\t12\tloss:169.0253\tval_loss:45.3911\tval_acc:72.8643\n",
            "Epoch\t13\tloss:167.9299\tval_loss:45.1245\tval_acc:77.3869\n",
            "Epoch\t14\tloss:167.0179\tval_loss:45.1373\tval_acc:71.8593\n",
            "Epoch\t15\tloss:165.7286\tval_loss:44.7386\tval_acc:76.8844\n",
            "Epoch\t16\tloss:164.5903\tval_loss:44.5874\tval_acc:80.4020\n",
            "Epoch\t17\tloss:163.6358\tval_loss:44.5098\tval_acc:72.3618\n",
            "Epoch\t18\tloss:162.8259\tval_loss:44.2666\tval_acc:78.3920\n",
            "Epoch\t19\tloss:161.6108\tval_loss:43.9993\tval_acc:76.8844\n",
            "Epoch\t20\tloss:160.4158\tval_loss:43.7506\tval_acc:80.4020\n",
            "Epoch\t21\tloss:159.3412\tval_loss:43.8017\tval_acc:76.3819\n",
            "Epoch\t22\tloss:158.4537\tval_loss:43.4136\tval_acc:80.4020\n",
            "Epoch\t23\tloss:157.3882\tval_loss:43.6878\tval_acc:77.3869\n"
          ]
        },
        {
          "output_type": "error",
          "ename": "KeyboardInterrupt",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-226-533c43a014a6>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     26\u001b[0m   \u001b[0;32mfor\u001b[0m \u001b[0mepoch\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnum_epochs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     27\u001b[0m     \u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain_epoch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrainloader\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mcriterion\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0moptimizer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 28\u001b[0;31m     \u001b[0mval_loss\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mval_acc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtest\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtestloader\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mcriterion\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     29\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'Epoch\\t{epoch+1}\\tloss:{loss:.4f}\\tval_loss:{val_loss:.4f}\\tval_acc:{val_acc:.4f}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     30\u001b[0m   \u001b[0;31m# test\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<ipython-input-222-14b89e39a470>\u001b[0m in \u001b[0;36mtest\u001b[0;34m(testDataset, criterion)\u001b[0m\n\u001b[1;32m      8\u001b[0m             \u001b[0mimages\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabels\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      9\u001b[0m             \u001b[0;31m# calculate outputs by running images through the network\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 10\u001b[0;31m             \u001b[0my_pred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnet\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimages\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     11\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     12\u001b[0m             \u001b[0;31m# cross entropy loss\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1100\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1101\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1102\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1103\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1104\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<ipython-input-213-cc499a358857>\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m     36\u001b[0m     \u001b[0;31m# Defining the forward pass\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     37\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 38\u001b[0;31m         \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconv_1\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     39\u001b[0m         \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconv_2\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     40\u001b[0m         \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1100\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1101\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1102\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1103\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1104\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/container.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    139\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    140\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 141\u001b[0;31m             \u001b[0minput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodule\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    142\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    143\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1100\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1101\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1102\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1103\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1104\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/batchnorm.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    177\u001b[0m             \u001b[0mbn_training\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    178\u001b[0m             \u001b[0mexponential_average_factor\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 179\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0meps\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    180\u001b[0m         )\n\u001b[1;32m    181\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py\u001b[0m in \u001b[0;36mbatch_norm\u001b[0;34m(input, running_mean, running_var, weight, bias, training, momentum, eps)\u001b[0m\n\u001b[1;32m   2281\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2282\u001b[0m     return torch.batch_norm(\n\u001b[0;32m-> 2283\u001b[0;31m         \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbias\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrunning_mean\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrunning_var\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtraining\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmomentum\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0meps\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackends\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcudnn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menabled\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2284\u001b[0m     )\n\u001b[1;32m   2285\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YZrjC94HoLXJ"
      },
      "source": [
        "Final Training on complete dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v_mrqNzPoLXJ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6df92e77-8efb-464d-ee0b-8614dabd7075"
      },
      "source": [
        "datasetLoader = torch.utils.data.DataLoader(dataset, batch_size=batch_size, shuffle = True)\n",
        "reset_net()\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer = Adam(net.parameters(), lr=0.001)\n",
        "for epoch in range(num_epochs):\n",
        "  loss = train_epoch(datasetLoader,criterion,optimizer)\n",
        "  print(f'Epoch\\t{epoch+1}\\tloss:{loss:.4f}')"
      ],
      "execution_count": 239,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch\t1\tloss:247.2648\n",
            "Epoch\t2\tloss:235.3156\n",
            "Epoch\t3\tloss:228.5250\n",
            "Epoch\t4\tloss:223.6824\n",
            "Epoch\t5\tloss:220.3557\n",
            "Epoch\t6\tloss:217.9892\n",
            "Epoch\t7\tloss:215.6699\n",
            "Epoch\t8\tloss:213.9618\n",
            "Epoch\t9\tloss:211.9828\n",
            "Epoch\t10\tloss:210.5367\n",
            "Epoch\t11\tloss:208.5740\n",
            "Epoch\t12\tloss:207.0578\n",
            "Epoch\t13\tloss:205.0313\n",
            "Epoch\t14\tloss:203.8956\n",
            "Epoch\t15\tloss:202.5827\n",
            "Epoch\t16\tloss:200.4906\n",
            "Epoch\t17\tloss:199.1673\n",
            "Epoch\t18\tloss:197.7532\n",
            "Epoch\t19\tloss:196.1889\n",
            "Epoch\t20\tloss:194.7737\n",
            "Epoch\t21\tloss:192.8270\n",
            "Epoch\t22\tloss:191.3840\n",
            "Epoch\t23\tloss:189.7968\n",
            "Epoch\t24\tloss:188.3414\n",
            "Epoch\t25\tloss:187.0779\n",
            "Epoch\t26\tloss:185.4001\n",
            "Epoch\t27\tloss:183.9859\n",
            "Epoch\t28\tloss:182.7346\n",
            "Epoch\t29\tloss:181.2585\n",
            "Epoch\t30\tloss:179.9532\n",
            "Epoch\t31\tloss:178.1110\n",
            "Epoch\t32\tloss:177.0186\n",
            "Epoch\t33\tloss:175.3506\n",
            "Epoch\t34\tloss:173.9112\n",
            "Epoch\t35\tloss:172.4497\n",
            "Epoch\t36\tloss:170.7974\n",
            "Epoch\t37\tloss:169.3408\n",
            "Epoch\t38\tloss:168.2335\n",
            "Epoch\t39\tloss:166.9082\n",
            "Epoch\t40\tloss:165.2790\n",
            "Epoch\t41\tloss:163.8974\n",
            "Epoch\t42\tloss:162.6295\n",
            "Epoch\t43\tloss:160.7169\n",
            "Epoch\t44\tloss:159.2948\n",
            "Epoch\t45\tloss:158.0128\n",
            "Epoch\t46\tloss:156.9035\n",
            "Epoch\t47\tloss:155.3620\n",
            "Epoch\t48\tloss:153.7903\n",
            "Epoch\t49\tloss:152.5809\n",
            "Epoch\t50\tloss:151.2143\n",
            "Epoch\t51\tloss:149.7799\n",
            "Epoch\t52\tloss:148.2200\n",
            "Epoch\t53\tloss:147.1504\n",
            "Epoch\t54\tloss:145.5656\n",
            "Epoch\t55\tloss:144.3415\n",
            "Epoch\t56\tloss:142.6774\n",
            "Epoch\t57\tloss:141.6715\n",
            "Epoch\t58\tloss:140.3614\n",
            "Epoch\t59\tloss:139.0377\n",
            "Epoch\t60\tloss:137.4460\n",
            "Epoch\t61\tloss:136.2031\n",
            "Epoch\t62\tloss:134.9140\n",
            "Epoch\t63\tloss:133.4608\n",
            "Epoch\t64\tloss:131.8032\n",
            "Epoch\t65\tloss:130.3108\n",
            "Epoch\t66\tloss:129.0039\n",
            "Epoch\t67\tloss:128.0669\n",
            "Epoch\t68\tloss:126.6908\n",
            "Epoch\t69\tloss:125.4031\n",
            "Epoch\t70\tloss:123.7563\n",
            "Epoch\t71\tloss:122.9955\n",
            "Epoch\t72\tloss:121.3818\n",
            "Epoch\t73\tloss:119.7337\n",
            "Epoch\t74\tloss:119.0667\n",
            "Epoch\t75\tloss:117.6083\n",
            "Epoch\t76\tloss:116.1113\n",
            "Epoch\t77\tloss:114.6517\n",
            "Epoch\t78\tloss:113.9259\n",
            "Epoch\t79\tloss:112.5459\n",
            "Epoch\t80\tloss:110.9353\n",
            "Epoch\t81\tloss:109.9883\n",
            "Epoch\t82\tloss:108.5436\n",
            "Epoch\t83\tloss:107.3570\n",
            "Epoch\t84\tloss:106.1349\n",
            "Epoch\t85\tloss:105.2088\n",
            "Epoch\t86\tloss:103.8396\n",
            "Epoch\t87\tloss:102.4645\n",
            "Epoch\t88\tloss:101.4250\n",
            "Epoch\t89\tloss:99.9345\n",
            "Epoch\t90\tloss:98.9502\n",
            "Epoch\t91\tloss:97.3991\n",
            "Epoch\t92\tloss:96.1742\n",
            "Epoch\t93\tloss:95.3375\n",
            "Epoch\t94\tloss:93.7254\n",
            "Epoch\t95\tloss:92.9834\n",
            "Epoch\t96\tloss:91.5074\n",
            "Epoch\t97\tloss:90.5965\n",
            "Epoch\t98\tloss:89.3127\n",
            "Epoch\t99\tloss:88.1183\n",
            "Epoch\t100\tloss:87.8673\n",
            "Epoch\t101\tloss:86.3920\n",
            "Epoch\t102\tloss:85.1555\n",
            "Epoch\t103\tloss:84.0413\n",
            "Epoch\t104\tloss:82.6530\n",
            "Epoch\t105\tloss:81.8814\n",
            "Epoch\t106\tloss:80.3326\n",
            "Epoch\t107\tloss:79.7278\n",
            "Epoch\t108\tloss:78.7023\n",
            "Epoch\t109\tloss:77.6342\n",
            "Epoch\t110\tloss:76.6109\n",
            "Epoch\t111\tloss:75.4095\n",
            "Epoch\t112\tloss:74.0818\n",
            "Epoch\t113\tloss:73.3112\n",
            "Epoch\t114\tloss:71.8396\n",
            "Epoch\t115\tloss:71.3292\n",
            "Epoch\t116\tloss:70.2561\n",
            "Epoch\t117\tloss:69.0892\n",
            "Epoch\t118\tloss:68.0609\n",
            "Epoch\t119\tloss:66.9679\n",
            "Epoch\t120\tloss:66.4598\n",
            "Epoch\t121\tloss:65.2318\n",
            "Epoch\t122\tloss:64.3795\n",
            "Epoch\t123\tloss:63.3340\n",
            "Epoch\t124\tloss:62.6400\n",
            "Epoch\t125\tloss:61.4256\n",
            "Epoch\t126\tloss:61.0096\n",
            "Epoch\t127\tloss:59.6087\n",
            "Epoch\t128\tloss:59.4219\n",
            "Epoch\t129\tloss:58.3283\n",
            "Epoch\t130\tloss:57.1711\n",
            "Epoch\t131\tloss:56.3932\n",
            "Epoch\t132\tloss:55.4790\n",
            "Epoch\t133\tloss:55.0496\n",
            "Epoch\t134\tloss:53.9310\n",
            "Epoch\t135\tloss:52.9582\n",
            "Epoch\t136\tloss:52.3371\n",
            "Epoch\t137\tloss:51.1039\n",
            "Epoch\t138\tloss:50.0624\n",
            "Epoch\t139\tloss:50.0148\n",
            "Epoch\t140\tloss:48.9883\n",
            "Epoch\t141\tloss:48.5088\n",
            "Epoch\t142\tloss:47.3530\n",
            "Epoch\t143\tloss:46.6468\n",
            "Epoch\t144\tloss:45.8617\n",
            "Epoch\t145\tloss:45.1125\n",
            "Epoch\t146\tloss:44.6194\n",
            "Epoch\t147\tloss:43.7812\n",
            "Epoch\t148\tloss:43.0454\n",
            "Epoch\t149\tloss:42.2550\n",
            "Epoch\t150\tloss:41.6287\n",
            "Epoch\t151\tloss:40.8267\n",
            "Epoch\t152\tloss:40.3095\n",
            "Epoch\t153\tloss:39.8348\n",
            "Epoch\t154\tloss:38.9514\n",
            "Epoch\t155\tloss:37.9457\n",
            "Epoch\t156\tloss:37.8607\n",
            "Epoch\t157\tloss:37.0776\n",
            "Epoch\t158\tloss:36.4051\n",
            "Epoch\t159\tloss:35.8196\n",
            "Epoch\t160\tloss:35.4615\n",
            "Epoch\t161\tloss:34.6690\n",
            "Epoch\t162\tloss:34.2666\n",
            "Epoch\t163\tloss:33.6905\n",
            "Epoch\t164\tloss:32.9728\n",
            "Epoch\t165\tloss:32.1956\n",
            "Epoch\t166\tloss:31.8926\n",
            "Epoch\t167\tloss:31.3889\n",
            "Epoch\t168\tloss:31.0082\n",
            "Epoch\t169\tloss:30.5031\n",
            "Epoch\t170\tloss:30.0923\n",
            "Epoch\t171\tloss:29.4278\n",
            "Epoch\t172\tloss:28.9691\n",
            "Epoch\t173\tloss:28.4719\n",
            "Epoch\t174\tloss:27.8183\n",
            "Epoch\t175\tloss:27.3932\n",
            "Epoch\t176\tloss:26.9374\n",
            "Epoch\t177\tloss:26.5207\n",
            "Epoch\t178\tloss:25.8541\n",
            "Epoch\t179\tloss:25.8977\n",
            "Epoch\t180\tloss:24.9569\n",
            "Epoch\t181\tloss:24.7569\n",
            "Epoch\t182\tloss:24.3667\n",
            "Epoch\t183\tloss:23.9381\n",
            "Epoch\t184\tloss:23.5158\n",
            "Epoch\t185\tloss:22.8221\n",
            "Epoch\t186\tloss:22.7375\n",
            "Epoch\t187\tloss:22.3349\n",
            "Epoch\t188\tloss:21.5464\n",
            "Epoch\t189\tloss:21.5345\n",
            "Epoch\t190\tloss:21.2046\n",
            "Epoch\t191\tloss:20.7647\n",
            "Epoch\t192\tloss:20.2716\n",
            "Epoch\t193\tloss:19.9628\n",
            "Epoch\t194\tloss:19.6404\n",
            "Epoch\t195\tloss:19.3189\n",
            "Epoch\t196\tloss:19.0670\n",
            "Epoch\t197\tloss:18.6845\n",
            "Epoch\t198\tloss:18.2874\n",
            "Epoch\t199\tloss:17.9733\n",
            "Epoch\t200\tloss:17.7579\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "H1UpQGREoLXJ"
      },
      "source": [
        "PATH = './final_model.pth'\n",
        "torch.save(net.state_dict(), PATH)"
      ],
      "execution_count": 240,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2sszD5hhOI5F"
      },
      "source": [
        "### Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jTtlg0QMSJ5S",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "41375803-7799-4a79-e1dd-cbe34ecbda62"
      },
      "source": [
        "!apt install libomp-dev\n",
        "!pip install faiss\n",
        "!pip install faiss-gpu"
      ],
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Reading package lists... Done\n",
            "Building dependency tree       \n",
            "Reading state information... Done\n",
            "The following additional packages will be installed:\n",
            "  libomp5\n",
            "Suggested packages:\n",
            "  libomp-doc\n",
            "The following NEW packages will be installed:\n",
            "  libomp-dev libomp5\n",
            "0 upgraded, 2 newly installed, 0 to remove and 37 not upgraded.\n",
            "Need to get 239 kB of archives.\n",
            "After this operation, 804 kB of additional disk space will be used.\n",
            "Get:1 http://archive.ubuntu.com/ubuntu bionic/universe amd64 libomp5 amd64 5.0.1-1 [234 kB]\n",
            "Get:2 http://archive.ubuntu.com/ubuntu bionic/universe amd64 libomp-dev amd64 5.0.1-1 [5,088 B]\n",
            "Fetched 239 kB in 1s (361 kB/s)\n",
            "Selecting previously unselected package libomp5:amd64.\n",
            "(Reading database ... 155222 files and directories currently installed.)\n",
            "Preparing to unpack .../libomp5_5.0.1-1_amd64.deb ...\n",
            "Unpacking libomp5:amd64 (5.0.1-1) ...\n",
            "Selecting previously unselected package libomp-dev.\n",
            "Preparing to unpack .../libomp-dev_5.0.1-1_amd64.deb ...\n",
            "Unpacking libomp-dev (5.0.1-1) ...\n",
            "Setting up libomp5:amd64 (5.0.1-1) ...\n",
            "Setting up libomp-dev (5.0.1-1) ...\n",
            "Processing triggers for libc-bin (2.27-3ubuntu1.3) ...\n",
            "/sbin/ldconfig.real: /usr/local/lib/python3.7/dist-packages/ideep4py/lib/libmkldnn.so.0 is not a symbolic link\n",
            "\n",
            "Collecting faiss\n",
            "  Downloading faiss-1.5.3-cp37-cp37m-manylinux1_x86_64.whl (4.7 MB)\n",
            "\u001b[K     |████████████████████████████████| 4.7 MB 4.9 MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from faiss) (1.19.5)\n",
            "Installing collected packages: faiss\n",
            "Successfully installed faiss-1.5.3\n",
            "Collecting faiss-gpu\n",
            "  Downloading faiss_gpu-1.7.1.post2-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (89.7 MB)\n",
            "\u001b[K     |████████████████████████████████| 89.7 MB 1.1 MB/s \n",
            "\u001b[?25hInstalling collected packages: faiss-gpu\n",
            "Successfully installed faiss-gpu-1.7.1.post2\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iwSBlkdRODsk",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 163,
          "referenced_widgets": [
            "0cd1884758d943e4a7cbcee44c5d88ad",
            "bc23887ba3d5441baedf58e14efdb64d",
            "99fe32270325402d892fafe5fd9e7048",
            "17e2590b35d04f6c9a689921fbbc28b9",
            "60853b977f1948dfb764960c623b153a",
            "9bd23908225a4502bbf864bf6761bbe7",
            "c088f1d254534036a2235f629d4bf7a6",
            "c3d7fa5cfd5e423983dea5d3e6cce18c",
            "9d1566e06a764d97a30df558124a1262",
            "8ac39e397aff436a85a4ed3a4a028b53",
            "c6bf132b57754fecb817195548262e22",
            "bce6852264b64ee599e23bce2aea316e",
            "ccd77c70a8fd48c498f87bc43ec312e7",
            "18b8484209dc49418aa6333e8944f248",
            "f88dff762ed549ffba1f894eb61ac4d4",
            "502bd8ba34a949f88775fa07300031b3",
            "c2d4bba57b5f4c38ab64661df0d27c5e",
            "70c6202ed76142788a55b1911ece23cd",
            "045b787c9b3b44f8858255cbffad9fe5",
            "c5c7053f3be546e78f32ab9eda61d173",
            "5dda229fdcb84adfb845966c02128e8a",
            "3ac5aba209a342ce91ddc5369a205a3f",
            "44a792a0d89d438bb8d7e13098f5563d",
            "1cf90ccdb0a249ea906da9bc4e645a19",
            "ce7ad7ccf2154cc3bbca8edb6ef3cdf7",
            "8ddc904e7b474adcb803c66c319572d0",
            "3a2953e5e8f74e139d117bf343aa9dc5",
            "f7a8eb6d27c84977bb50b2d6a4876baf",
            "13a2160214f44257a97503c647c821a1",
            "7f5688001e4d4180970a4d8a4930d77a",
            "89cfeb4e98e24353a114d1e400d6b232",
            "532d981b923c4323a1409b6840bcefc6",
            "657eb0dd50164caea73671544e78d4f1",
            "a96508424b81468594e66d2bdb7eef2c",
            "e5c0b960e93d457dbfd3fc1afec2add4",
            "ac0c92b66d814930989d242661c1c41a",
            "55ba1aa8f97940f9b569f7cf24eb4ba3",
            "3c8534f2dc3d4889be8e53ae8e34a102",
            "a3475060715743c9b1a2086b9d933d79",
            "0901239051704a9aa6435a63ef3eb6a7",
            "2af63ec015ab4990a06c685202955d2e",
            "0929a6b4f40b4ed19f2f0897fef40b4c",
            "a7d6f429edc94dd399ba7dce98e7fbc6",
            "3045bc3462cb4a729759f43b0c8471c0"
          ]
        },
        "outputId": "baf934ef-3a28-4c53-8f87-8ae14ffff3a4"
      },
      "source": [
        "# Evaluation \n",
        "# Acknowledgement: the code is based on Siddhant Kapil's repo on LA-Transformer\n",
        "\n",
        "from __future__ import print_function\n",
        "\n",
        "import os, sys\n",
        "import faiss\n",
        "import numpy as np\n",
        "\n",
        "from PIL import Image\n",
        "from tqdm.notebook import tqdm\n",
        "\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "from torch.nn import init\n",
        "import torch.optim as optim\n",
        "from torchvision import models\n",
        "import torch.nn.functional as F\n",
        "from torchvision import datasets, transforms\n",
        "from torch.utils.data import DataLoader, Dataset\n",
        "\n",
        "from utils import get_id\n",
        "from metrics import rank1, rank5, calc_ap\n",
        "\n",
        "\n",
        "\n",
        "# ### Set feature volume sizes (height, width, depth) \n",
        "# TODO: update with your model's feature length\n",
        "\n",
        "batch_size = 1\n",
        "H, W, D = 1, 1, 128 # for dummymodel we have feature volume 7x7x2048\n",
        "\n",
        "# ### Load Model\n",
        "\n",
        "# TODO: Uncomment the following lines to load the Implemented and trained Model\n",
        "\n",
        "save_path = \"final_model.pth\"\n",
        "model = ReidModel1(numClasses = numClasses, inference=True)\n",
        "model.load_state_dict(torch.load(save_path), strict=False)\n",
        "model.eval()\n",
        "\n",
        "# TODO: Comment out the dummy model\n",
        "# model = DummyModel(batch_size, H, W, D)\n",
        "\n",
        "# ### Data Loader for query and gallery\n",
        "\n",
        "# TODO: For demo, we have resized to 224x224 during data augmentation\n",
        "# You are free to use augmentations of your own choice\n",
        "transform = [transforms.Resize((128,48)),\n",
        "        transforms.ToTensor()]\n",
        "transformCam1 = [\n",
        "        transforms.Normalize(meanCam1, stdCam1)\n",
        "    ]\n",
        "transformCam2 = [\n",
        "        transforms.Normalize(meanCam2, stdCam2)\n",
        "    ]\n",
        "\n",
        "data_transforms = {\n",
        "        'basic': transforms.Compose( transform ),\n",
        "        'Cam1': transforms.Compose( transformCam1 ),\n",
        "        'Cam2': transforms.Compose(transformCam2)\n",
        "    }\n",
        "\n",
        "\n",
        "image_datasets = {}\n",
        "data_dir = starter_path+\"/data/val\"\n",
        "\n",
        "image_datasets['query'] = datasets.ImageFolder(os.path.join(data_dir, 'query'), data_transforms['basic'])\n",
        "image_datasets['gallery'] = datasets.ImageFolder(os.path.join(data_dir, 'gallery'), data_transforms['basic'])\n",
        "query_loader = DataLoader(dataset = image_datasets['query'], batch_size=batch_size, shuffle=False )\n",
        "gallery_loader = DataLoader(dataset = image_datasets['gallery'], batch_size=batch_size, shuffle=False)\n",
        "\n",
        "class_names = image_datasets['query'].classes\n",
        "\n",
        "\n",
        "# ###  Extract Features\n",
        "def extract_feature_alt(dataset):\n",
        "  features =  torch.FloatTensor()\n",
        "  cam_labels, _ = get_id(dataset.imgs)\n",
        "  for i in tqdm(range(len(dataset))):\n",
        "    if (cam_labels[i]=='1'):\n",
        "      img = data_transforms['Cam1'](dataset[i][0])\n",
        "    else:\n",
        "      img = data_transforms['Cam2'](dataset[i][0])\n",
        "    img = img[None, :]\n",
        "    output = model(img)\n",
        "    output = output[None, None, :]\n",
        "    features = torch.cat((features, output.detach().cpu()), 0)\n",
        "  return features\n",
        "\n",
        "def extract_feature(dataloaders):\n",
        "    \n",
        "    features =  torch.FloatTensor()\n",
        "    count = 0\n",
        "    idx = 0\n",
        "    for data in tqdm(dataloaders):\n",
        "        img, label = data\n",
        "        # print(label)\n",
        "        # Uncomment if using GPU for inference\n",
        "        #img, label = img.cuda(), label.cuda()\n",
        "\n",
        "        output = model(img) # (B, D, H, W) --> B: batch size, HxWxD: feature volume size\n",
        "        n, c, h, w = img.size()\n",
        "        \n",
        "        count += n\n",
        "        features = torch.cat((features, output.detach().cpu()), 0)\n",
        "        idx += 1\n",
        "    return features\n",
        "\n",
        "# Extract Query Features\n",
        "\n",
        "query_feature= extract_feature_alt(image_datasets['query'])\n",
        "\n",
        "# Extract Gallery Features\n",
        "\n",
        "gallery_feature = extract_feature_alt(image_datasets['gallery'])\n",
        "\n",
        "# Retrieve labels\n",
        "\n",
        "gallery_path = image_datasets['gallery'].imgs\n",
        "query_path = image_datasets['query'].imgs\n",
        "\n",
        "gallery_cam,gallery_label = get_id(gallery_path)\n",
        "query_cam,query_label = get_id(query_path)\n",
        "\n",
        "\n",
        "# ## Concat Averaged GELTs\n",
        "concatenated_query_vectors = []\n",
        "for query in tqdm(query_feature):\n",
        "    fnorm = torch.norm(query, p=2, dim=0, keepdim=True)#*np.sqrt(H*W)\n",
        "    query_norm = query.div(fnorm.expand_as(query))\n",
        "    concatenated_query_vectors.append(query_norm.view((-1)))\n",
        "\n",
        "concatenated_gallery_vectors = []\n",
        "for gallery in tqdm(gallery_feature):\n",
        "    fnorm = torch.norm(gallery, p=2, dim=0, keepdim=True)#*np.sqrt(H*W)\n",
        "    gallery_norm = gallery.div(fnorm.expand_as(gallery))\n",
        "    concatenated_gallery_vectors.append(gallery_norm.view((-1)))\n",
        "  \n",
        "\n",
        "# ## Calculate Similarity using FAISS\n",
        "\n",
        "index = faiss.IndexIDMap(faiss.IndexFlatIP(H*W*D))\n",
        "\n",
        "index.add_with_ids(np.array([t.numpy() for t in concatenated_gallery_vectors]),np.array(gallery_label))\n",
        "\n",
        "def search(query: str, k=1):\n",
        "    encoded_query = query.unsqueeze(dim=0).numpy()\n",
        "    top_k = index.search(encoded_query, k)\n",
        "    return top_k\n",
        "\n",
        "\n",
        "# ### Evaluate \n",
        "\n",
        "rank1_score = 0\n",
        "rank5_score = 0\n",
        "ap = 0\n",
        "count = 0\n",
        "for query, label in zip(concatenated_query_vectors, query_label):\n",
        "    count += 1\n",
        "    label = label\n",
        "    output = search(query, k=10)\n",
        "    rank1_score += rank1(label, output) \n",
        "    rank5_score += rank5(label, output) \n",
        "    print(\"Correct: {}, Total: {}, Incorrect: {}\".format(rank1_score, count, count-rank1_score), end=\"\\r\")\n",
        "    ap += calc_ap(label, output)\n",
        "\n",
        "print(\"Rank1: %.3f, Rank5: %.3f, mAP: %.3f\"%(rank1_score/len(query_feature), \n",
        "                                             rank5_score/len(query_feature), \n",
        "                                             ap/len(query_feature)))    "
      ],
      "execution_count": 241,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "0cd1884758d943e4a7cbcee44c5d88ad",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "  0%|          | 0/28 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "bce6852264b64ee599e23bce2aea316e",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "  0%|          | 0/181 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "44a792a0d89d438bb8d7e13098f5563d",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "  0%|          | 0/28 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "a96508424b81468594e66d2bdb7eef2c",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "  0%|          | 0/181 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Correct: 1, Total: 1, Incorrect: 0\rCorrect: 2, Total: 2, Incorrect: 0\rCorrect: 3, Total: 3, Incorrect: 0\rCorrect: 4, Total: 4, Incorrect: 0\rCorrect: 5, Total: 5, Incorrect: 0\rCorrect: 6, Total: 6, Incorrect: 0\rCorrect: 7, Total: 7, Incorrect: 0\rCorrect: 8, Total: 8, Incorrect: 0\rCorrect: 9, Total: 9, Incorrect: 0\rCorrect: 10, Total: 10, Incorrect: 0\rCorrect: 11, Total: 11, Incorrect: 0\rCorrect: 12, Total: 12, Incorrect: 0\rCorrect: 13, Total: 13, Incorrect: 0\rCorrect: 14, Total: 14, Incorrect: 0\rCorrect: 15, Total: 15, Incorrect: 0\rCorrect: 16, Total: 16, Incorrect: 0\rCorrect: 17, Total: 17, Incorrect: 0\rCorrect: 18, Total: 18, Incorrect: 0\rCorrect: 19, Total: 19, Incorrect: 0\rCorrect: 20, Total: 20, Incorrect: 0\rCorrect: 21, Total: 21, Incorrect: 0\rCorrect: 22, Total: 22, Incorrect: 0\rCorrect: 22, Total: 23, Incorrect: 1\rCorrect: 22, Total: 24, Incorrect: 2\rCorrect: 22, Total: 25, Incorrect: 3\rCorrect: 23, Total: 26, Incorrect: 3\rCorrect: 24, Total: 27, Incorrect: 3\rCorrect: 24, Total: 28, Incorrect: 4\rRank1: 0.857, Rank5: 0.964, mAP: 0.792\n"
          ]
        }
      ]
    }
  ]
}