Saliency.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "executionInfo": {
     "elapsed": 3301,
     "status": "ok",
     "timestamp": 1619696354713,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "zInRO0ULtxvQ"
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os\n",
    "from tqdm import tqdm\n",
    "from natsort import natsorted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "executionInfo": {
     "elapsed": 4742,
     "status": "ok",
     "timestamp": 1619696356410,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "Jdrg-wrQAtMT"
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "executionInfo": {
     "elapsed": 4537,
     "status": "ok",
     "timestamp": 1619696356420,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "l42ZyxytuOK_"
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 28241,
     "status": "ok",
     "timestamp": 1619696391090,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "vRVqhkTSV78h",
    "outputId": "91ef117b-7c3b-40cb-9b81-5a2ed3ca4963"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mounted at /content/drive\n"
     ]
    }
   ],
   "source": [
    "from google.colab import drive\n",
    "drive.mount('/content/drive')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "executionInfo": {
     "elapsed": 5607,
     "status": "ok",
     "timestamp": 1619696397626,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "JEoXi6869GTx"
   },
   "outputs": [],
   "source": [
    "from tensorflow import keras\n",
    "model = keras.models.load_model('/content/drive/MyDrive/best models crispcnn/model_seed10_epoch50.hdf5', compile=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "executionInfo": {
     "elapsed": 2507,
     "status": "ok",
     "timestamp": 1619696397630,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "BB6OCnc4U9Kd"
   },
   "outputs": [],
   "source": [
    "epoch = 50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 1569,
     "status": "ok",
     "timestamp": 1619696400838,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "NOjdaAgUALuK",
    "outputId": "0c21e608-884b-436d-a09d-d15a599a59c2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_1\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "i_1 (InputLayer)                [(None, 4, 28, 1)]   0                                            \n",
      "__________________________________________________________________________________________________\n",
      "i_2 (InputLayer)                [(None, 4, 28, 1)]   0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_1 (Conv2D)               (None, 2, 26, 128)   1280        i_1[0][0]                        \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_3 (Conv2D)               (None, 2, 26, 128)   1280        i_2[0][0]                        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_1 (BatchNor (None, 2, 26, 128)   512         conv2d_1[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_3 (BatchNor (None, 2, 26, 128)   512         conv2d_3[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_2 (Conv2D)               (None, 1, 25, 128)   65664       batch_normalization_1[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_4 (Conv2D)               (None, 1, 25, 128)   65664       batch_normalization_3[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_2 (BatchNor (None, 1, 25, 128)   512         conv2d_2[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_4 (BatchNor (None, 1, 25, 128)   512         conv2d_4[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "lambda_1 (Lambda)               (None, 25, 128)      0           batch_normalization_2[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "lambda_2 (Lambda)               (None, 25, 128)      0           batch_normalization_4[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "bidirectional_1 (Bidirectional) (None, 25, 256)      263168      lambda_1[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "bidirectional_2 (Bidirectional) (None, 25, 256)      263168      lambda_2[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "flatten_1 (Flatten)             (None, 6400)         0           bidirectional_1[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "flatten_2 (Flatten)             (None, 6400)         0           bidirectional_2[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "concatenate_1 (Concatenate)     (None, 12800)        0           flatten_1[0][0]                  \n",
      "                                                                 flatten_2[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "dense_1 (Dense)                 (None, 256)          3277056     concatenate_1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "dense_2 (Dense)                 (None, 256)          65792       dense_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dense_3 (Dense)                 (None, 256)          3277056     concatenate_1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "r (Dense)                       (None, 1)            257         dense_2[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "c (Dense)                       (None, 1)            257         dense_3[0][0]                    \n",
      "==================================================================================================\n",
      "Total params: 7,282,690\n",
      "Trainable params: 7,281,666\n",
      "Non-trainable params: 1,024\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "executionInfo": {
     "elapsed": 1015,
     "status": "ok",
     "timestamp": 1619696407045,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "tEOyT_T3ebkB"
   },
   "outputs": [],
   "source": [
    "#CAMS Code below"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "executionInfo": {
     "elapsed": 4490,
     "status": "ok",
     "timestamp": 1619696410785,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "3rhZGuSU-TeM"
   },
   "outputs": [],
   "source": [
    "#load directly from npy files #this step skips the data loading from folders as images\n",
    "off_dataset = np.load('/content/drive/MyDrive/best models crispcnn/off_dataset.npy')\n",
    "target_dataset = np.load('/content/drive/MyDrive/best models crispcnn/target_dataset.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3110,
     "status": "ok",
     "timestamp": 1619696410785,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "16c9VuF9U9Kh",
    "outputId": "645832ae-65de-488b-9ab0-63156bd4fbd8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((153924, 4, 28, 1), (153924, 4, 28, 1))"
      ]
     },
     "execution_count": 11,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "off_dataset.shape, target_dataset.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "executionInfo": {
     "elapsed": 2546,
     "status": "ok",
     "timestamp": 1619696410786,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "2QSZbFmLU9Ki"
   },
   "outputs": [],
   "source": [
    "sample_id = 9399"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "executionInfo": {
     "elapsed": 1950,
     "status": "ok",
     "timestamp": 1619696410787,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "Xrnn16p1U9Kj"
   },
   "outputs": [],
   "source": [
    "x1 = off_dataset[sample_id]\n",
    "x2 = target_dataset[sample_id]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 114
    },
    "executionInfo": {
     "elapsed": 2114,
     "status": "ok",
     "timestamp": 1619696411364,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "JMxVgQIH_eSq",
    "outputId": "24c5a920-2c06-4a28-aacf-201d009aeae0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fd5e8669690>"
      ]
     },
     "execution_count": 14,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(x1.squeeze(), cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 114
    },
    "executionInfo": {
     "elapsed": 1383,
     "status": "ok",
     "timestamp": 1619696411366,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "f1ehoinY_el1",
    "outputId": "05ce4bde-6d08-440f-c248-09490b6c4c42"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fd5e8139d50>"
      ]
     },
     "execution_count": 15,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(x2.squeeze(), cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "executionInfo": {
     "elapsed": 710,
     "status": "ok",
     "timestamp": 1619696411367,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "-Zv9tcts_ekH"
   },
   "outputs": [],
   "source": [
    "# Rendering\n",
    "subplot_args = { 'nrows': 1, 'ncols': 2, 'figsize': (3, 3),\n",
    "                 'subplot_kw': {'xticks': [], 'yticks': []} }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "executionInfo": {
     "elapsed": 1416,
     "status": "ok",
     "timestamp": 1619696412649,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "-Sv9QAiK_ei3"
   },
   "outputs": [],
   "source": [
    "def loss(output):\n",
    "    return (output[0][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "executionInfo": {
     "elapsed": 833,
     "status": "ok",
     "timestamp": 1619696412651,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "lh2krlpZ_ehS"
   },
   "outputs": [],
   "source": [
    "def model_modifier(m):\n",
    "    m.layers[-1].activation = tf.keras.activations.linear\n",
    "    return m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 5119,
     "status": "ok",
     "timestamp": 1619696419920,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "r4gwOgUeZUcy",
    "outputId": "fa0207ae-710d-4ad2-8a11-fd1ade26b9d4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting tf_keras_vis\n",
      "  Downloading https://files.pythonhosted.org/packages/82/b3/692323c96f54c379ce0f2d2dec5ec3d1abbad076b962dd5f3974fb5d4858/tf_keras_vis-0.5.5-py3-none-any.whl\n",
      "Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from tf_keras_vis) (1.19.5)\n",
      "Requirement already satisfied: pillow in /usr/local/lib/python3.7/dist-packages (from tf_keras_vis) (7.1.2)\n",
      "Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from tf_keras_vis) (1.4.1)\n",
      "Requirement already satisfied: imageio in /usr/local/lib/python3.7/dist-packages (from tf_keras_vis) (2.4.1)\n",
      "Installing collected packages: tf-keras-vis\n",
      "Successfully installed tf-keras-vis-0.5.5\n"
     ]
    }
   ],
   "source": [
    "!pip install tf_keras_vis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "executionInfo": {
     "elapsed": 1257,
     "status": "ok",
     "timestamp": 1619696422376,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "c_9Bn-Kf_edk"
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras import backend as K\n",
    "from tf_keras_vis.saliency import Saliency\n",
    "from tf_keras_vis.utils import normalize\n",
    "from tf_keras_vis.gradcam import Gradcam,GradcamPlusPlus\n",
    "from tf_keras_vis.scorecam import ScoreCAM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 1073,
     "status": "ok",
     "timestamp": 1619696426743,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "3891jeomeKNH",
    "outputId": "b410ccb0-91d7-4459-8e69-823c754384c7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 <tensorflow.python.keras.engine.input_layer.InputLayer object at 0x7fd627f648d0>\n",
      "1 <tensorflow.python.keras.engine.input_layer.InputLayer object at 0x7fd5ec29b4d0>\n",
      "2 <tensorflow.python.keras.layers.convolutional.Conv2D object at 0x7fd5ec29b9d0>\n",
      "3 <tensorflow.python.keras.layers.convolutional.Conv2D object at 0x7fd5ec2c7610>\n",
      "4 <tensorflow.python.keras.layers.normalization_v2.BatchNormalization object at 0x7fd5ec2c7d50>\n",
      "5 <tensorflow.python.keras.layers.normalization_v2.BatchNormalization object at 0x7fd5ec300dd0>\n",
      "6 <tensorflow.python.keras.layers.convolutional.Conv2D object at 0x7fd627f64550>\n",
      "7 <tensorflow.python.keras.layers.convolutional.Conv2D object at 0x7fd5ec297f10>\n",
      "8 <tensorflow.python.keras.layers.normalization_v2.BatchNormalization object at 0x7fd5ec223690>\n",
      "9 <tensorflow.python.keras.layers.normalization_v2.BatchNormalization object at 0x7fd5ec223bd0>\n",
      "10 <tensorflow.python.keras.layers.core.Lambda object at 0x7fd5ec223f50>\n",
      "11 <tensorflow.python.keras.layers.core.Lambda object at 0x7fd5ec29b3d0>\n",
      "12 <tensorflow.python.keras.layers.wrappers.Bidirectional object at 0x7fd5f767f550>\n",
      "13 <tensorflow.python.keras.layers.wrappers.Bidirectional object at 0x7fd5ecbadbd0>\n",
      "14 <tensorflow.python.keras.layers.core.Flatten object at 0x7fd5f32aa7d0>\n",
      "15 <tensorflow.python.keras.layers.core.Flatten object at 0x7fd5f32aa4d0>\n",
      "16 <tensorflow.python.keras.layers.merge.Concatenate object at 0x7fd5f32aa210>\n",
      "17 <tensorflow.python.keras.layers.core.Dense object at 0x7fd5f32aa0d0>\n",
      "18 <tensorflow.python.keras.layers.core.Dense object at 0x7fd5f331d410>\n",
      "19 <tensorflow.python.keras.layers.core.Dense object at 0x7fd5f331ae90>\n",
      "20 <tensorflow.python.keras.layers.core.Dense object at 0x7fd627f44c10>\n",
      "21 <tensorflow.python.keras.layers.core.Dense object at 0x7fd5f3317290>\n"
     ]
    }
   ],
   "source": [
    "for ind in range(len(model.layers)):\n",
    "    print (ind, model.layers[ind])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "0aIyaNTFaZeG"
   },
   "outputs": [],
   "source": [
    "#starts below"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "t03u-5IG_eaf"
   },
   "outputs": [],
   "source": [
    "saliency = Saliency(model,model_modifier=model_modifier,clone=True)\n",
    "saliency_map = saliency(loss, [x1,x2])\n",
    "saliency_map_1,saliency_map_2 = normalize(saliency_map[0]),normalize(saliency_map[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 297
    },
    "executionInfo": {
     "elapsed": 21840,
     "status": "ok",
     "timestamp": 1619611723708,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "hrhh3-JaCIBZ",
    "outputId": "de5f52d1-a1d5-4f18-e4a6-012045c6d89e"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 6 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig  = plt.figure(figsize=(15,5),facecolor=\"#ffffff\",linewidth=6, edgecolor=\"#000000\")\n",
    "ax = fig.gca()\n",
    "plt.subplots_adjust(hspace = .1, wspace = .1)\n",
    "plt.subplot(2,3,1)\n",
    "plt.axis('off')\n",
    "plt.imshow(x1.squeeze(), cmap='gray')\n",
    "plt.title('off_target', color='k')\n",
    "plt.subplot(2,3,4)\n",
    "plt.axis('off')\n",
    "plt.imshow(x2.squeeze(), cmap='gray')\n",
    "plt.title('target', color='k')\n",
    "plt.subplot(2,3,2)\n",
    "plt.axis('off')\n",
    "plt.imshow(saliency_map_1.squeeze(), cmap='jet', interpolation='bilinear')\n",
    "plt.title('off_target saliency', color='k')\n",
    "plt.subplot(2,3,5)\n",
    "plt.axis('off')\n",
    "plt.imshow(saliency_map_2.squeeze(), cmap='jet', interpolation='bilinear')\n",
    "plt.title('target saliency', color='k')\n",
    "plt.subplot(2,3,3)\n",
    "plt.axis('off')\n",
    "plt.imshow(x1.squeeze(), cmap='gray')\n",
    "plt.imshow(saliency_map_1.squeeze(), cmap='jet', interpolation='bilinear', alpha=0.7)\n",
    "plt.title('off_target saliency superimposed', color='k')\n",
    "plt.subplot(2,3,6)\n",
    "plt.axis('off')\n",
    "plt.imshow(x2.squeeze(), cmap='gray')\n",
    "plt.imshow(saliency_map_2.squeeze(), cmap='jet', interpolation='bilinear', alpha=0.7)\n",
    "plt.title('target saliency superimposed', color='k')\n",
    "plt.tight_layout()\n",
    "fig.suptitle('Saliency CAMS: seq_no_' + str(sample_id) , fontsize=20, color='k')\n",
    "# plt.show()\n",
    "plt.savefig('/content/drive/MyDrive/best models crispcnn/cams/saliency_epoch'+'best_model_seed10_epoch_'+str(epoch)+'_sample'+str(sample_id)+'.png', dpi=1200)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "IgnaHUZCESOb"
   },
   "outputs": [],
   "source": [
    "#smoothed\n",
    "smooth_saliency = Saliency(model,model_modifier=model_modifier,clone=True)\n",
    "saliency_map = smooth_saliency(loss, [x1,x2], smooth_samples=30,smooth_noise=0.10)\n",
    "saliency_map_1,saliency_map_2 = normalize(saliency_map[0]),normalize(saliency_map[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 297
    },
    "executionInfo": {
     "elapsed": 47306,
     "status": "ok",
     "timestamp": 1619611749680,
     "user": {
      "displayName": "Dhvani Vora",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gh1WcloBC66e-N8Q6m-DpNIM04jeU4uqEUpcEao=s64",
      "userId": "10048243182801573586"
     },
     "user_tz": -330
    },
    "id": "m_2r08DiMVzw",
    "outputId": "8378db37-2818-4189-c554-ca614a168930"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 6 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig  = plt.figure(figsize=(15,5),facecolor=\"#ffffff\",linewidth=6, edgecolor=\"#000000\")\n",
    "ax = fig.gca()\n",
    "plt.subplots_adjust(hspace = .1, wspace = .1)\n",
    "plt.subplot(2,3,1)\n",
    "plt.axis('off')\n",
    "plt.imshow(x1.squeeze(), cmap='gray')\n",
    "plt.title('off_target', color='k')\n",
    "plt.subplot(2,3,4)\n",
    "plt.axis('off')\n",
    "plt.imshow(x2.squeeze(), cmap='gray')\n",
    "plt.title('target', color='k')\n",
    "plt.subplot(2,3,2)\n",
    "plt.axis('off')\n",
    "plt.imshow(saliency_map_1.squeeze(), cmap='jet', interpolation='bilinear')\n",
    "plt.title('off_target saliency', color='k')\n",
    "plt.subplot(2,3,5)\n",
    "plt.axis('off')\n",
    "plt.imshow(saliency_map_2.squeeze(), cmap='jet', interpolation='bilinear')\n",
    "plt.title('target saliency', color='k')\n",
    "plt.subplot(2,3,3)\n",
    "plt.axis('off')\n",
    "plt.imshow(x1.squeeze(), cmap='gray')\n",
    "plt.imshow(saliency_map_1.squeeze(), cmap='jet', interpolation='bilinear', alpha=0.7)\n",
    "plt.title('off_target saliency superimposed', color='k')\n",
    "plt.subplot(2,3,6)\n",
    "plt.axis('off')\n",
    "plt.imshow(x2.squeeze(), cmap='gray')\n",
    "plt.imshow(saliency_map_2.squeeze(), cmap='jet', interpolation='bilinear', alpha=0.7)\n",
    "plt.title('target saliency superimposed', color='k')\n",
    "plt.tight_layout()\n",
    "fig.suptitle('Smoothed Saliency CAMS: seq_no_' + str(sample_id) , fontsize=20, color='k')\n",
    "# plt.show()\n",
    "plt.savefig('/content/drive/MyDrive/best models crispcnn/cams/smoothsaliency_epoch'+str(epoch)+'_sample'+str(sample_id)+'.png', dpi=1200)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "name": "crisprcam.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}