From 2200854da02085576fd6128cdb7fbb7b5d33d2f1 Mon Sep 17 00:00:00 2001
From: Shreyas Vaidya <95679215+Shreyasvaidya@users.noreply.github.com>
Date: Fri, 1 Sep 2023 14:02:56 +0530
Subject: [PATCH] Siamese network according to the signet paper

---
 Shreyas_cnn.ipynb | 1566 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 1566 insertions(+)
 create mode 100644 Shreyas_cnn.ipynb

diff --git a/Shreyas_cnn.ipynb b/Shreyas_cnn.ipynb
new file mode 100644
index 0000000..864138c
--- /dev/null
+++ b/Shreyas_cnn.ipynb
@@ -0,0 +1,1566 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "zI6aeZ7qix8g"
+      },
+      "source": [
+        "# Loading Data"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "tNZaRpSFi6LT"
+      },
+      "source": [
+        "# Imports"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 1,
+      "metadata": {
+        "id": "PKeuXVNERsc-"
+      },
+      "outputs": [
+        {
+          "ename": "ModuleNotFoundError",
+          "evalue": "No module named 'otsu'",
+          "output_type": "error",
+          "traceback": [
+            "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
+            "\u001b[1;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
+            "\u001b[1;32md:\\Shreyas\\Shreyas_study\\vision_Challenges\\Shreyas_cnn.ipynb Cell 3\u001b[0m in \u001b[0;36m<cell line: 21>\u001b[1;34m()\u001b[0m\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#W2sZmlsZQ%3D%3D?line=18'>19</a>\u001b[0m \u001b[39mfrom\u001b[39;00m \u001b[39msklearn\u001b[39;00m\u001b[39m.\u001b[39;00m\u001b[39mmetrics\u001b[39;00m \u001b[39mimport\u001b[39;00m roc_curve, auc\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#W2sZmlsZQ%3D%3D?line=19'>20</a>\u001b[0m \u001b[39mimport\u001b[39;00m \u001b[39mrandom\u001b[39;00m\n\u001b[1;32m---> <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#W2sZmlsZQ%3D%3D?line=20'>21</a>\u001b[0m \u001b[39mfrom\u001b[39;00m \u001b[39motsu\u001b[39;00m \u001b[39mimport\u001b[39;00m otsu\n",
+            "\u001b[1;31mModuleNotFoundError\u001b[0m: No module named 'otsu'"
+          ]
+        }
+      ],
+      "source": [
+        "from torch.utils.data import Dataset\n",
+        "from torchvision import transforms\n",
+        "from PIL import Image\n",
+        "import pandas as pd\n",
+        "import os\n",
+        "import torch\n",
+        "\n",
+        "from tqdm import tqdm\n",
+        "import pickle\n",
+        "from torch import nn\n",
+        "from torch.optim import Adam,RMSprop\n",
+        "from torch.utils.data import DataLoader\n",
+        "import matplotlib.pyplot as plt\n",
+        "import math\n",
+        "from PIL import Image\n",
+        "import numpy as np\n",
+        "import pandas as pd\n",
+        "from sklearn.metrics import f1_score, roc_auc_score\n",
+        "from sklearn.metrics import roc_curve, auc\n",
+        "import random\n",
+        "from otsu import otsu\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "b8xBWB4FkjfQ"
+      },
+      "source": [
+        "# Creating Dataloader"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 2,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "bCU7jy-wT7hD",
+        "outputId": "a3ada429-d5c9-4aa9-82ef-bd36fb158100"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "0it [00:00, ?it/s]"
+          ]
+        },
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "1352it [00:08, 152.59it/s]\n"
+          ]
+        }
+      ],
+      "source": [
+        "#creating list of image pairs with combinations and labels of 1 or 0\n",
+        "path = r'dataset\\dataset\\train'\n",
+        "img_label_list = []\n",
+        "\n",
+        "count = 0\n",
+        "authors = os.listdir(path)\n",
+        "for i,author_name in tqdm(enumerate(authors)):\n",
+        "  other_list = authors[i:i+2] if i<len(authors)-1 else authors[i:]\n",
+        "  for author_name2 in other_list:\n",
+        "    label = 1 if author_name == author_name2 else 0\n",
+        "    for img1 in os.listdir(os.path.join(path,author_name)):\n",
+        "      for img2 in os.listdir(os.path.join(path,author_name2)):\n",
+        "        if(author_name != author_name2 or img1 !=img2):\n",
+        "\n",
+        "          \n",
+        "          img_label_list.append([os.path.join(path,author_name,img1),os.path.join(path,author_name2,img2),label])\n",
+        "         "
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 3,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "NXhWUavyBLok",
+        "outputId": "b4a037da-c2e7-41c2-ffe2-0e96e7ee4e17"
+      },
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "183859"
+            ]
+          },
+          "execution_count": 3,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "len(img_label_list)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": []
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 4,
+      "metadata": {
+        "id": "MlkF0lCtRGnw"
+      },
+      "outputs": [],
+      "source": [
+        "class SiameseDataset(Dataset):\n",
+        "  def __init__(self,array,path= None,transform = None):\n",
+        "    self.array =array\n",
+        "    self.transform = transform\n",
+        "    self.path = path\n",
+        "  def __getitem__(self,index):\n",
+        "    \n",
+        "    indexlist = self.array[index]\n",
+        "      \n",
+        "    if self.path is None:\n",
+        "      img0 = Image.open(indexlist[0])\n",
+        "      img1 = Image.open(indexlist[1])\n",
+        "    else:\n",
+        "      img0 = Image.open(os.path.join(self.path,indexlist[0]))\n",
+        "      img1 = Image.open(os.path.join(self.path,indexlist[1]))\n",
+        "    \n",
+        "    label = indexlist[2]\n",
+        "    img0 = img0.convert(\"L\")\n",
+        "    img1 = img1.convert(\"L\")\n",
+        "    \n",
+        "    threshold = 128\n",
+        "    bw_image0 = img0.point(lambda x: 0 if x < threshold else 1, \"1\")\n",
+        "    bw_image1 = img1.point(lambda x: 0 if x < threshold else 1, \"1\")\n",
+        "    \n",
+        "    if self.transform is not None:\n",
+        "      bw_image0 = self.transform(bw_image0)\n",
+        "      bw_image1 = self.transform(bw_image1)\n",
+        "    #converting grayscale to binary\n",
+        "    \n",
+        "    \n",
+        "    return bw_image0,bw_image1,label\n",
+        "  def __len__(self):\n",
+        "    return len(self.array)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 5,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "class AddGaussianNoise(object):\n",
+        "    def __init__(self, mean=0., std=0.05):\n",
+        "        self.std = std\n",
+        "        self.mean = mean\n",
+        "        \n",
+        "    def __call__(self, tensor):\n",
+        "        return tensor + torch.randn(tensor.size()) * self.std + self.mean\n",
+        "    \n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 6,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "mytransform = transforms.Compose([transforms.Resize((250,25)),\n",
+        "                                  transforms.RandomRotation(22.5),\n",
+        "                                  \n",
+        "                                transforms.ToTensor(),\n",
+        "                                AddGaussianNoise()])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 7,
+      "metadata": {
+        "id": "HpqVn7aY1nUL"
+      },
+      "outputs": [],
+      "source": [
+        "random.shuffle(img_label_list)\n",
+        "siamese_dataset = SiameseDataset(img_label_list,transform=mytransform)\n",
+        "\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 15,
+      "metadata": {
+        "id": "2wOyO6V2lIcD"
+      },
+      "outputs": [],
+      "source": [
+        "train_dataloader = DataLoader(siamese_dataset,batch_size = 512)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": []
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 11,
+      "metadata": {
+        "id": "JbZbFUIsLmtn"
+      },
+      "outputs": [],
+      "source": [
+        "train_dataloader1 = DataLoader(siamese_dataset,batch_size = 1)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 9,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>img1_name</th>\n",
+              "      <th>img2_name</th>\n",
+              "      <th>label</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>70a4ba9a.jpg</td>\n",
+              "      <td>58f68a00.jpg</td>\n",
+              "      <td>1</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>e15b612e.jpg</td>\n",
+              "      <td>16ce5df2.jpg</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>6c64d988.jpg</td>\n",
+              "      <td>735d3636.jpg</td>\n",
+              "      <td>1</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>b74681a6.jpg</td>\n",
+              "      <td>cb50496d.jpg</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>c97aa428.jpg</td>\n",
+              "      <td>6f9dc747.jpg</td>\n",
+              "      <td>1</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>...</th>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5765</th>\n",
+              "      <td>4e7762f7.jpg</td>\n",
+              "      <td>2f3d2bce.jpg</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5766</th>\n",
+              "      <td>c1c49f87.jpg</td>\n",
+              "      <td>1072597f.jpg</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5767</th>\n",
+              "      <td>03e66099.jpg</td>\n",
+              "      <td>f877b1b6.jpg</td>\n",
+              "      <td>1</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5768</th>\n",
+              "      <td>8e056e51.jpg</td>\n",
+              "      <td>f68e79a1.jpg</td>\n",
+              "      <td>1</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5769</th>\n",
+              "      <td>acd11d73.jpg</td>\n",
+              "      <td>7eae060a.jpg</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "<p>5770 rows × 3 columns</p>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "         img1_name     img2_name  label\n",
+              "0     70a4ba9a.jpg  58f68a00.jpg      1\n",
+              "1     e15b612e.jpg  16ce5df2.jpg      0\n",
+              "2     6c64d988.jpg  735d3636.jpg      1\n",
+              "3     b74681a6.jpg  cb50496d.jpg      0\n",
+              "4     c97aa428.jpg  6f9dc747.jpg      1\n",
+              "...            ...           ...    ...\n",
+              "5765  4e7762f7.jpg  2f3d2bce.jpg      0\n",
+              "5766  c1c49f87.jpg  1072597f.jpg      0\n",
+              "5767  03e66099.jpg  f877b1b6.jpg      1\n",
+              "5768  8e056e51.jpg  f68e79a1.jpg      1\n",
+              "5769  acd11d73.jpg  7eae060a.jpg      0\n",
+              "\n",
+              "[5770 rows x 3 columns]"
+            ]
+          },
+          "execution_count": 9,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "df = pd.read_csv(r'D:\\Shreyas\\Shreyas_study\\vision_Challenges\\dataset\\dataset\\val.csv')\n",
+        "df"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 10,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "val_dataset = SiameseDataset(np.array(df),path = r'D:\\Shreyas\\Shreyas_study\\vision_Challenges\\dataset\\dataset\\val',transform=mytransform)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 11,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "val_dataloader = DataLoader(val_dataset)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "MysR98Wjk-Gp"
+      },
+      "source": [
+        "# Training the model"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 12,
+      "metadata": {
+        "id": "DhCdYUSr1xaR"
+      },
+      "outputs": [],
+      "source": [
+        "#create a siamese network\n",
+        "class SiameseNetwork(nn.Module):\n",
+        "    def __init__(self):\n",
+        "        super(SiameseNetwork, self).__init__()\n",
+        "        # Defining the fully connected layers\n",
+        "        self.cnn1 = nn.Sequential(\n",
+        "            nn.Conv2d(1, 1, kernel_size=3,stride=1),\n",
+        "            nn.MaxPool2d(3, stride=1),\n",
+        "            \n",
+        "           \n",
+        "            \n",
+        "\n",
+        "           \n",
+        "        )\n",
+        "\n",
+        "\n",
+        "\n",
+        "        # Defining the fully connected layers\n",
+        "        self.fc1 = nn.Sequential(\n",
+        "            \n",
+        "           \n",
+        "            \n",
+        "            nn.Linear(5166, 4096),\n",
+        "        \n",
+        "            nn.ReLU(inplace=True),\n",
+        "            \n",
+        "            nn.Linear(4096, 1024),\n",
+        "        \n",
+        "            nn.ReLU(inplace=True),\n",
+        "           \n",
+        "            nn.Linear(1024, 128),\n",
+        "            nn.ReLU(inplace=True),\n",
+        "\n",
+        "            nn.Linear(128,2)\n",
+        "           \n",
+        "            )\n",
+        "\n",
+        "            \n",
+        "        \n",
+        "        \n",
+        "    def forward_once(self, x):\n",
+        "        # Forward pass \n",
+        "        output = self.cnn1(x)\n",
+        "        output = output.view(output.size()[0], -1)\n",
+        "        output = self.fc1(output)\n",
+        "        return output\n",
+        "\n",
+        "        \n",
+        "   \n",
+        "\n",
+        "        \n",
+        "\n",
+        "    def forward(self, input1, input2):\n",
+        "        # forward pass of input 1\n",
+        "        output1 = self.forward_once(input1)\n",
+        "        # forward pass of input 2\n",
+        "        output2 = self.forward_once(input2)\n",
+        "        return output1, output2\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 13,
+      "metadata": {
+        "id": "TGgK_zrC189r"
+      },
+      "outputs": [],
+      "source": [
+        "class ContrastiveLoss(torch.nn.Module):\n",
+        "    \"\"\"\n",
+        "    Contrastive loss function.\n",
+        "    Based on:\n",
+        "    \"\"\"\n",
+        "\n",
+        "    def __init__(self, margin=1.0):\n",
+        "        super(ContrastiveLoss, self).__init__()\n",
+        "        self.margin = margin\n",
+        "\n",
+        "    def forward(self, x0, x1, y):\n",
+        "        # euclidian distance\n",
+        "        diff = x0 - x1\n",
+        "        dist_sq = torch.sum(torch.pow(diff, 2), 1)\n",
+        "        dist = torch.sqrt(dist_sq)\n",
+        "        \n",
+        "        mdist = self.margin - dist\n",
+        "        dist = torch.clamp(mdist, min=0.0)\n",
+        "        loss = y * dist_sq + (1 - y) * torch.pow(dist, 2) \n",
+        "        \n",
+        "        loss = torch.sum(loss) / 2.0 / x0.size()[0]\n",
+        "        return loss"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 17,
+      "metadata": {
+        "colab": {
+          "background_save": true,
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "_Ut0mo3j2EVk",
+        "outputId": "d961b095-ed17-4e5a-8c87-1da4f50c432f"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "  4%|▍         | 15/360 [03:14<1:14:27, 12.95s/it]\n"
+          ]
+        },
+        {
+          "ename": "KeyboardInterrupt",
+          "evalue": "",
+          "output_type": "error",
+          "traceback": [
+            "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
+            "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
+            "\u001b[1;32md:\\Shreyas\\Shreyas_study\\vision_Challenges\\Shreyas_cnn.ipynb Cell 21\u001b[0m in \u001b[0;36m<cell line: 58>\u001b[1;34m()\u001b[0m\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=56'>57</a>\u001b[0m device \u001b[39m=\u001b[39m torch\u001b[39m.\u001b[39mdevice(\u001b[39m'\u001b[39m\u001b[39mcuda\u001b[39m\u001b[39m'\u001b[39m \u001b[39mif\u001b[39;00m torch\u001b[39m.\u001b[39mcuda\u001b[39m.\u001b[39mis_available() \u001b[39melse\u001b[39;00m \u001b[39m'\u001b[39m\u001b[39mcpu\u001b[39m\u001b[39m'\u001b[39m)\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=57'>58</a>\u001b[0m \u001b[39mfor\u001b[39;00m i \u001b[39min\u001b[39;00m \u001b[39mrange\u001b[39m(\u001b[39m1\u001b[39m,n_epochs\u001b[39m+\u001b[39m\u001b[39m1\u001b[39m):\n\u001b[1;32m---> <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=58'>59</a>\u001b[0m     train(i)\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=59'>60</a>\u001b[0m     test(i)\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=60'>61</a>\u001b[0m     counter\u001b[39m.\u001b[39mappend(i)\n",
+            "\u001b[1;32md:\\Shreyas\\Shreyas_study\\vision_Challenges\\Shreyas_cnn.ipynb Cell 21\u001b[0m in \u001b[0;36mtrain\u001b[1;34m(epoch)\u001b[0m\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=12'>13</a>\u001b[0m net\u001b[39m.\u001b[39mtrain()\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=13'>14</a>\u001b[0m tloss \u001b[39m=\u001b[39m \u001b[39m0\u001b[39m\n\u001b[1;32m---> <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=14'>15</a>\u001b[0m \u001b[39mfor\u001b[39;00m data \u001b[39min\u001b[39;00m tqdm(train_dataloader):\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=15'>16</a>\u001b[0m     img0, img1 , label \u001b[39m=\u001b[39m data\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=16'>17</a>\u001b[0m         \u001b[39m#img0, img1 , label = img0.cuda(), img1.cuda() , label.cuda()\u001b[39;00m\n",
+            "File \u001b[1;32mc:\\ProgramData\\miniconda3\\lib\\site-packages\\tqdm\\std.py:1195\u001b[0m, in \u001b[0;36mtqdm.__iter__\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m   1192\u001b[0m time \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_time\n\u001b[0;32m   1194\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[1;32m-> 1195\u001b[0m     \u001b[39mfor\u001b[39;00m obj \u001b[39min\u001b[39;00m iterable:\n\u001b[0;32m   1196\u001b[0m         \u001b[39myield\u001b[39;00m obj\n\u001b[0;32m   1197\u001b[0m         \u001b[39m# Update and possibly print the progressbar.\u001b[39;00m\n\u001b[0;32m   1198\u001b[0m         \u001b[39m# Note: does not call self.update(1) for speed optimisation.\u001b[39;00m\n",
+            "File \u001b[1;32mc:\\ProgramData\\miniconda3\\lib\\site-packages\\torch\\utils\\data\\dataloader.py:633\u001b[0m, in \u001b[0;36m_BaseDataLoaderIter.__next__\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    630\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_sampler_iter \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[0;32m    631\u001b[0m     \u001b[39m# TODO(https://github.com/pytorch/pytorch/issues/76750)\u001b[39;00m\n\u001b[0;32m    632\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_reset()  \u001b[39m# type: ignore[call-arg]\u001b[39;00m\n\u001b[1;32m--> 633\u001b[0m data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_next_data()\n\u001b[0;32m    634\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_num_yielded \u001b[39m+\u001b[39m\u001b[39m=\u001b[39m \u001b[39m1\u001b[39m\n\u001b[0;32m    635\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_dataset_kind \u001b[39m==\u001b[39m _DatasetKind\u001b[39m.\u001b[39mIterable \u001b[39mand\u001b[39;00m \\\n\u001b[0;32m    636\u001b[0m         \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_IterableDataset_len_called \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39mand\u001b[39;00m \\\n\u001b[0;32m    637\u001b[0m         \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_num_yielded \u001b[39m>\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_IterableDataset_len_called:\n",
+            "File \u001b[1;32mc:\\ProgramData\\miniconda3\\lib\\site-packages\\torch\\utils\\data\\dataloader.py:677\u001b[0m, in \u001b[0;36m_SingleProcessDataLoaderIter._next_data\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    675\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_next_data\u001b[39m(\u001b[39mself\u001b[39m):\n\u001b[0;32m    676\u001b[0m     index \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_next_index()  \u001b[39m# may raise StopIteration\u001b[39;00m\n\u001b[1;32m--> 677\u001b[0m     data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_dataset_fetcher\u001b[39m.\u001b[39;49mfetch(index)  \u001b[39m# may raise StopIteration\u001b[39;00m\n\u001b[0;32m    678\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_pin_memory:\n\u001b[0;32m    679\u001b[0m         data \u001b[39m=\u001b[39m _utils\u001b[39m.\u001b[39mpin_memory\u001b[39m.\u001b[39mpin_memory(data, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_pin_memory_device)\n",
+            "File \u001b[1;32mc:\\ProgramData\\miniconda3\\lib\\site-packages\\torch\\utils\\data\\_utils\\fetch.py:51\u001b[0m, in \u001b[0;36m_MapDatasetFetcher.fetch\u001b[1;34m(self, possibly_batched_index)\u001b[0m\n\u001b[0;32m     49\u001b[0m         data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset\u001b[39m.\u001b[39m__getitems__(possibly_batched_index)\n\u001b[0;32m     50\u001b[0m     \u001b[39melse\u001b[39;00m:\n\u001b[1;32m---> 51\u001b[0m         data \u001b[39m=\u001b[39m [\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset[idx] \u001b[39mfor\u001b[39;00m idx \u001b[39min\u001b[39;00m possibly_batched_index]\n\u001b[0;32m     52\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m     53\u001b[0m     data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset[possibly_batched_index]\n",
+            "File \u001b[1;32mc:\\ProgramData\\miniconda3\\lib\\site-packages\\torch\\utils\\data\\_utils\\fetch.py:51\u001b[0m, in \u001b[0;36m<listcomp>\u001b[1;34m(.0)\u001b[0m\n\u001b[0;32m     49\u001b[0m         data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset\u001b[39m.\u001b[39m__getitems__(possibly_batched_index)\n\u001b[0;32m     50\u001b[0m     \u001b[39melse\u001b[39;00m:\n\u001b[1;32m---> 51\u001b[0m         data \u001b[39m=\u001b[39m [\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mdataset[idx] \u001b[39mfor\u001b[39;00m idx \u001b[39min\u001b[39;00m possibly_batched_index]\n\u001b[0;32m     52\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m     53\u001b[0m     data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset[possibly_batched_index]\n",
+            "\u001b[1;32md:\\Shreyas\\Shreyas_study\\vision_Challenges\\Shreyas_cnn.ipynb Cell 21\u001b[0m in \u001b[0;36mSiameseDataset.__getitem__\u001b[1;34m(self, index)\u001b[0m\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=20'>21</a>\u001b[0m threshold \u001b[39m=\u001b[39m \u001b[39m128\u001b[39m\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=21'>22</a>\u001b[0m bw_image0 \u001b[39m=\u001b[39m img0\u001b[39m.\u001b[39mpoint(\u001b[39mlambda\u001b[39;00m x: \u001b[39m0\u001b[39m \u001b[39mif\u001b[39;00m x \u001b[39m<\u001b[39m threshold \u001b[39melse\u001b[39;00m \u001b[39m1\u001b[39m, \u001b[39m\"\u001b[39m\u001b[39m1\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[1;32m---> <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=22'>23</a>\u001b[0m bw_image1 \u001b[39m=\u001b[39m img1\u001b[39m.\u001b[39;49mpoint(\u001b[39mlambda\u001b[39;49;00m x: \u001b[39m0\u001b[39;49m \u001b[39mif\u001b[39;49;00m x \u001b[39m<\u001b[39;49m threshold \u001b[39melse\u001b[39;49;00m \u001b[39m1\u001b[39;49m, \u001b[39m\"\u001b[39;49m\u001b[39m1\u001b[39;49m\u001b[39m\"\u001b[39;49m)\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=24'>25</a>\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mtransform \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[0;32m     <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#X26sZmlsZQ%3D%3D?line=25'>26</a>\u001b[0m   bw_image0 \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mtransform(bw_image0)\n",
+            "File \u001b[1;32mc:\\ProgramData\\miniconda3\\lib\\site-packages\\PIL\\Image.py:1833\u001b[0m, in \u001b[0;36mImage.point\u001b[1;34m(self, lut, mode)\u001b[0m\n\u001b[0;32m   1831\u001b[0m \u001b[39mif\u001b[39;00m mode \u001b[39m!=\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mF\u001b[39m\u001b[39m\"\u001b[39m:\n\u001b[0;32m   1832\u001b[0m     lut \u001b[39m=\u001b[39m [\u001b[39mround\u001b[39m(i) \u001b[39mfor\u001b[39;00m i \u001b[39min\u001b[39;00m lut]\n\u001b[1;32m-> 1833\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_new(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mim\u001b[39m.\u001b[39;49mpoint(lut, mode))\n",
+            "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
+          ]
+        }
+      ],
+      "source": [
+        "# Declare Siamese Network\n",
+        "n_epochs = 1\n",
+        "net = SiameseNetwork()#.cuda()\n",
+        "# Decalre Loss Function\n",
+        "criterion = ContrastiveLoss()\n",
+        "# Declare Optimizer\n",
+        "optimizer =RMSprop(net.parameters(), lr=1e-4, weight_decay=0.0005,momentum =0.9,eps = 1e-8)\n",
+        "#train the model\n",
+        "loss=[] \n",
+        "counter=[]\n",
+        "test_loss = []\n",
+        "def train(epoch):\n",
+        "    net.train()\n",
+        "    tloss = 0\n",
+        "    for data in tqdm(train_dataloader):\n",
+        "        img0, img1 , label = data\n",
+        "            #img0, img1 , label = img0.cuda(), img1.cuda() , label.cuda()\n",
+        "        \n",
+        "        optimizer.zero_grad()\n",
+        "        output1,output2 = net(img0,img1)\n",
+        "        \n",
+        "        loss_contrastive = criterion(output1,output2,label)\n",
+        "        \n",
+        "        if(math.isnan(loss_contrastive.item())):\n",
+        "            print(\"Nan\")\n",
+        "            break\n",
+        "        tloss+=(loss_contrastive.item())\n",
+        "        loss_contrastive.backward()\n",
+        "        optimizer.step()\n",
+        "        \n",
+        "        \n",
+        "        \n",
+        "    print(\"Epoch {}\\n Current loss {}\\n\".format(epoch,tloss/1437))\n",
+        "    loss.append(tloss/1437)\n",
+        "def test(epoch):\n",
+        "    net.eval()\n",
+        "    with torch.no_grad():\n",
+        "        \n",
+        "        loss = 0\n",
+        "        for i in tqdm(val_dataloader):\n",
+        "\n",
+        "            img0,img1,label = i\n",
+        "            #img0,img1 = img0.cuda(),img1.cuda()\n",
+        "            output1,output2 = net(img0,img1)\n",
+        "            sample_loss = criterion(output1,output2,label)\n",
+        "            loss+=sample_loss.item()\n",
+        "        \n",
+        "        test_loss.append(loss/5770)\n",
+        "        print(\"Epoch {}\\n Current test loss {}\\n\".format(epoch,loss/5700))\n",
+        "    \n",
+        "\n",
+        "\n",
+        "\n",
+        "        \n",
+        "#set the device to cuda\n",
+        "\n",
+        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
+        "for i in range(1,n_epochs+1):\n",
+        "    train(i)\n",
+        "    test(i)\n",
+        "    counter.append(i)\n",
+        "plt.plot(range(n_epochs),loss,color = 'blue',label = 'Train Loss')\n",
+        "plt.plot(range(n_epochs),test_loss,color = 'red',label = 'Test Loss')\n",
+        "plt.legend()\n",
+        "plt.show()\n"
+      ]
+    },
+    {
+      "attachments": {},
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "REMBG !"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "WnTgUk7PlTiC"
+      },
+      "source": [
+        "# Testing"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 81
+        },
+        "id": "84lJ7wF3U6Kw",
+        "outputId": "2849a0bc-d4f6-4e1d-ea54-7d682c58be62"
+      },
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<PIL.JpegImagePlugin.JpegImageFile image mode=L size=2722x172>"
+            ]
+          },
+          "execution_count": 22,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "img0 = Image.open(img_label_list[405][0])\n",
+        "img0"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "wwn9RvWikYiR",
+        "outputId": "dabd56fc-3b26-4bf6-e71b-2e3d266db1ef"
+      },
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "(2722, 172)"
+            ]
+          },
+          "execution_count": 23,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "img0.size"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 73
+        },
+        "id": "w4jKLMoEVYs0",
+        "outputId": "052c7b9a-77f5-4032-8ce1-ae912a6d13b1"
+      },
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<PIL.JpegImagePlugin.JpegImageFile image mode=L size=2934x182>"
+            ]
+          },
+          "execution_count": 24,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "img1 = Image.open(img_label_list[405][1])\n",
+        "img1"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "pAfdhopfVarx",
+        "outputId": "8ed2a56f-a3fe-4678-e3a8-789cf0181b75"
+      },
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "0"
+            ]
+          },
+          "execution_count": 25,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "label = img_label_list[405][2]\n",
+        "label"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "iVMToqKyBwBY",
+        "outputId": "a5f496d0-8abd-4373-e720-2df1271f5238"
+      },
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "(2722, 172)"
+            ]
+          },
+          "execution_count": 26,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "img0.size"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "RIvV874XfY2-",
+        "outputId": "8300d50c-357f-4071-f6da-ab616fed9ca0"
+      },
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "(2934, 182)"
+            ]
+          },
+          "execution_count": 27,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "img1.size"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "<All keys matched successfully>"
+            ]
+          },
+          "execution_count": 16,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "net = SiameseNetwork()\n",
+        "net.load_state_dict(torch.load(r'D:\\Shreyas\\Shreyas_study\\vision_Challenges\\optim.pt'))"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "sXRh7jqUZyFy",
+        "outputId": "5eb575a0-2a7f-40db-fe76-37b9ac658dc5"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "100%|██████████| 183859/183859 [59:01<00:00, 51.92it/s] \n"
+          ]
+        }
+      ],
+      "source": [
+        "distances=np.zeros(183859)\n",
+        "net.eval()\n",
+        "with torch.no_grad():\n",
+        "  sumdist = 0\n",
+        "  counter = 0\n",
+        "  batch_size = 1\n",
+        "\n",
+        "  for i in tqdm(train_dataloader1):\n",
+        "      \n",
+        "      img0,img1,label = i\n",
+        "      \n",
+        "      #img0,img1 = img0.cuda(),img1.cuda()\n",
+        "      \n",
+        "      output1,output2 = net(img0,img1)\n",
+        "      diff = output1 - output2\n",
+        "      dist_sq = torch.sum(torch.pow(diff, 2), 1)\n",
+        "      \n",
+        "      dist = torch.sqrt(dist_sq)\n",
+        "      \n",
+        "      # if(counter+batch_size-1<len(img_label_list)):\n",
+        "      #   distances[counter:counter+batch_size] = dist.cpu()\n",
+        "      #   counter+=batch_size\n",
+        "      # else:\n",
+        "      #   distances[counter:] = dist.cpu()\n",
+        "# for i in tqdm(enumerate(val_dataloader)):\n",
+        "#     img0,img1,_ = i\n",
+        "#     #img0,img1 = img0.cuda(),img1.cuda()\n",
+        "#     output1,output2 = model(img0,img1)\n",
+        "#     diff = output1 - output2\n",
+        "#     dist_sq = torch.sum(torch.pow(diff, 2), 1)\n",
+        "#     dist = torch.sqrt(dist_sq)\n",
+        "      \n",
+        "\n",
+        "      \n",
+        "      \n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "94065"
+            ]
+          },
+          "execution_count": 26,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "len(distances0)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "0.0"
+            ]
+          },
+          "execution_count": 30,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "min(distances0)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {
+            "needs_background": "light"
+          },
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "import seaborn as sns\n",
+        "sns.kdeplot(distances1,color = 'blue')\n",
+        "sns.kdeplot(distances0,color= 'red')\n",
+        "plt.show()\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [
+        {
+          "ename": "TypeError",
+          "evalue": "object of type 'NoneType' has no len()",
+          "output_type": "error",
+          "traceback": [
+            "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
+            "\u001b[1;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+            "\u001b[1;32md:\\Shreyas\\Shreyas_study\\vision_Challenges\\Shreyas_cnn.ipynb Cell 37\u001b[0m in \u001b[0;36m<cell line: 2>\u001b[1;34m()\u001b[0m\n\u001b[0;32m      <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#Y125sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m distances \u001b[39m=\u001b[39m distances1\u001b[39m.\u001b[39mextend(distances0)\n\u001b[1;32m----> <a href='vscode-notebook-cell:/d%3A/Shreyas/Shreyas_study/vision_Challenges/Shreyas_cnn.ipynb#Y125sZmlsZQ%3D%3D?line=1'>2</a>\u001b[0m random\u001b[39m.\u001b[39;49mshuffle(distances)\n",
+            "File \u001b[1;32mc:\\ProgramData\\miniconda3\\lib\\random.py:391\u001b[0m, in \u001b[0;36mRandom.shuffle\u001b[1;34m(self, x, random)\u001b[0m\n\u001b[0;32m    389\u001b[0m \u001b[39mif\u001b[39;00m random \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[0;32m    390\u001b[0m     randbelow \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_randbelow\n\u001b[1;32m--> 391\u001b[0m     \u001b[39mfor\u001b[39;00m i \u001b[39min\u001b[39;00m \u001b[39mreversed\u001b[39m(\u001b[39mrange\u001b[39m(\u001b[39m1\u001b[39m, \u001b[39mlen\u001b[39;49m(x))):\n\u001b[0;32m    392\u001b[0m         \u001b[39m# pick an element in x[:i+1] with which to exchange x[i]\u001b[39;00m\n\u001b[0;32m    393\u001b[0m         j \u001b[39m=\u001b[39m randbelow(i \u001b[39m+\u001b[39m \u001b[39m1\u001b[39m)\n\u001b[0;32m    394\u001b[0m         x[i], x[j] \u001b[39m=\u001b[39m x[j], x[i]\n",
+            "\u001b[1;31mTypeError\u001b[0m: object of type 'NoneType' has no len()"
+          ]
+        }
+      ],
+      "source": [
+        "distances1.extend(distances0)\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "distances = distances1\n",
+        "random.shuffle(distances)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "94065"
+            ]
+          },
+          "execution_count": 30,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": []
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 435
+        },
+        "id": "0T7sT69tkPbz",
+        "outputId": "52d4dedf-8800-4c26-ecad-326db3a442f4"
+      },
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {
+            "needs_background": "light"
+          },
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "_,probabilities = sns.ecdfplot(distances).get_lines()[0].get_data()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "hvJX5Qnekd25"
+      },
+      "outputs": [],
+      "source": [
+        "def get_nearest_index(test_distance,array = sorted(distances)):\n",
+        "    array = np.asarray(array)\n",
+        "    \n",
+        "    return ((np.abs(array-test_distance)).argmin() if test_distance<max(array) else len(array)-1)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "xKSBRCjYkkdZ"
+      },
+      "outputs": [],
+      "source": [
+        "def get_label_and_probability(test_distance,prob_array = probabilities[:-1]):\n",
+        "  idx = get_nearest_index(test_distance)\n",
+        "  prob = prob_array[idx]\n",
+        "  return (1,1-prob) if prob<0.5 else (0,1-prob)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "3IqBWDP2k6C5"
+      },
+      "outputs": [],
+      "source": [
+        "transform=transforms.Compose([transforms.Resize((250,25)),\n",
+        "                                                              transforms.ToTensor()\n",
+        "                                                                      ])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "9aZOh1RgBpbl"
+      },
+      "outputs": [],
+      "source": [
+        "val_dataset = SiameseDataset(np.array(df),path = r'D:\\Shreyas\\Shreyas_study\\vision_Challenges\\dataset\\dataset\\val',transform=mytransform)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "LU_Id_3QeKGv"
+      },
+      "outputs": [],
+      "source": [
+        "val_dataloader = DataLoader(val_dataset)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "qqYL_Di_knb7",
+        "outputId": "26dff04f-341a-4426-d270-ffc248c0ac18"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "5770it [06:16, 15.31it/s]\n"
+          ]
+        }
+      ],
+      "source": [
+        "with torch.no_grad():\n",
+        "  \n",
+        "  y_pred = np.zeros(5770)\n",
+        "  y_prob  = np.zeros(5770)\n",
+        "\n",
+        "  for count,i in tqdm(enumerate(val_dataloader)):\n",
+        "    img0,img1,_ = i\n",
+        "    #img0,img1 = img0.cuda(),img1.cuda()\n",
+        "    output1,output2 = net(img0,img1)\n",
+        "    diff = output1 - output2\n",
+        "    dist_sq = torch.sum(torch.pow(diff, 2), 1)\n",
+        "    dist = torch.sqrt(dist_sq)\n",
+        "    \n",
+        "    (label,prob) = get_label_and_probability(dist.item())\n",
+        "    \n",
+        "    y_pred[count] = label\n",
+        "    y_prob[count] = prob\n",
+        "    "
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "vpxVSErmMOw3"
+      },
+      "outputs": [],
+      "source": [
+        "y_true = df['label'].values"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "tkgZLTmQMQev",
+        "outputId": "2ef4e0fa-fbd3-4240-93dc-0c9a366b11a4"
+      },
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "0.5125348189415042"
+            ]
+          },
+          "execution_count": 42,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "f1_score(y_true,y_pred)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "ei6rYr0LMeNl",
+        "outputId": "ccbe5902-9f5d-4a3a-cbe4-e66b3a857602"
+      },
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "0.5177352528617212"
+            ]
+          },
+          "execution_count": 43,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "roc_auc_score(y_true,y_prob)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "rqBmMrFv_mmy"
+      },
+      "outputs": [],
+      "source": [
+        "df['label'] = y_pred\n",
+        "df['proba'] = y_prob"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "8Fcw_x4yd1Yk",
+        "outputId": "d26b6280-4c77-477f-ab96-665862928dd7"
+      },
+      "outputs": [],
+      "source": [
+        "df2 = pd.DataFrame()\n",
+        "df2['img1_name']  = df['img1_name']\n",
+        "df2['img2_name']  = df['img2_name']\n",
+        "df2['label'] = y_pred\n",
+        "df2['proba'] = y_prob"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>img1_name</th>\n",
+              "      <th>img2_name</th>\n",
+              "      <th>label</th>\n",
+              "      <th>proba</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>70a4ba9a.jpg</td>\n",
+              "      <td>58f68a00.jpg</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.786837</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>e15b612e.jpg</td>\n",
+              "      <td>16ce5df2.jpg</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.000044</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>6c64d988.jpg</td>\n",
+              "      <td>735d3636.jpg</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.537205</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>b74681a6.jpg</td>\n",
+              "      <td>cb50496d.jpg</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.000174</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>c97aa428.jpg</td>\n",
+              "      <td>6f9dc747.jpg</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.116252</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>...</th>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5765</th>\n",
+              "      <td>4e7762f7.jpg</td>\n",
+              "      <td>2f3d2bce.jpg</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.500144</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5766</th>\n",
+              "      <td>c1c49f87.jpg</td>\n",
+              "      <td>1072597f.jpg</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.018819</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5767</th>\n",
+              "      <td>03e66099.jpg</td>\n",
+              "      <td>f877b1b6.jpg</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.698329</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5768</th>\n",
+              "      <td>8e056e51.jpg</td>\n",
+              "      <td>f68e79a1.jpg</td>\n",
+              "      <td>1.0</td>\n",
+              "      <td>0.891705</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5769</th>\n",
+              "      <td>acd11d73.jpg</td>\n",
+              "      <td>7eae060a.jpg</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.434393</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "<p>5770 rows × 4 columns</p>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "         img1_name     img2_name  label     proba\n",
+              "0     70a4ba9a.jpg  58f68a00.jpg    1.0  0.786837\n",
+              "1     e15b612e.jpg  16ce5df2.jpg    0.0  0.000044\n",
+              "2     6c64d988.jpg  735d3636.jpg    1.0  0.537205\n",
+              "3     b74681a6.jpg  cb50496d.jpg    0.0  0.000174\n",
+              "4     c97aa428.jpg  6f9dc747.jpg    0.0  0.116252\n",
+              "...            ...           ...    ...       ...\n",
+              "5765  4e7762f7.jpg  2f3d2bce.jpg    1.0  0.500144\n",
+              "5766  c1c49f87.jpg  1072597f.jpg    0.0  0.018819\n",
+              "5767  03e66099.jpg  f877b1b6.jpg    1.0  0.698329\n",
+              "5768  8e056e51.jpg  f68e79a1.jpg    1.0  0.891705\n",
+              "5769  acd11d73.jpg  7eae060a.jpg    0.0  0.434393\n",
+              "\n",
+              "[5770 rows x 4 columns]"
+            ]
+          },
+          "execution_count": 95,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "df2"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "df2.to_csv(\"Open CV_02.csv\")"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "mDFQZ4yM_wLi",
+        "outputId": "1fb47ee9-c728-47a3-a3da-ebc4f5be62ee"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "0it [00:00, ?it/s]/usr/local/lib/python3.10/dist-packages/torch/nn/functional.py:1331: UserWarning: dropout2d: Received a 2-D input to dropout2d, which is deprecated and will result in an error in a future release. To retain the behavior and silence this warning, please use dropout instead. Note that dropout2d exists to provide channel-wise dropout on inputs with 2 spatial dimensions, a channel dimension, and an optional batch dimension (i.e. 3D or 4D inputs).\n",
+            "  warnings.warn(warn_msg)\n",
+            "5770it [01:26, 66.72it/s]\n"
+          ]
+        }
+      ],
+      "source": [
+        "threshold_dist = 9.2683e-06\n",
+        "with torch.no_grad():\n",
+        "  y_pred = np.zeros(5770)\n",
+        "  for count,i in tqdm(enumerate(val_dataloader)):\n",
+        "    path = '/content/dataset/val'\n",
+        "    img0,img1,label = i\n",
+        "    \n",
+        "    img0,img1 = img0.cuda(),img1.cuda()\n",
+        "    output1,output2 = net(img0,img1)\n",
+        "    diff = output1 - output2\n",
+        "    dist_sq = torch.sum(torch.pow(diff, 2), 1)\n",
+        "    dist = torch.sqrt(dist_sq)\n",
+        "    \n",
+        "    y_pred[count] = 1 if dist[0] <= threshold_dist else 0\n",
+        "    \n",
+        "\n",
+        "\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "t5KrUJ60BcuV"
+      },
+      "outputs": [],
+      "source": [
+        "y_true = df['label'].values"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "3ob-XEweC0La",
+        "outputId": "88255077-f697-4029-ab83-c521fdf710b2"
+      },
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "0.598336210955894"
+            ]
+          },
+          "execution_count": 37,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "f1_score(y_true, y_pred)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "0.5882648853058761"
+            ]
+          },
+          "execution_count": 38,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "roc_auc_score(y_true,y_prob)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "-XtNlBAdEzxF",
+        "outputId": "82b8ff88-89be-4bad-df33-3c1af7204e63"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "torch.Size([128, 1, 250, 25])\n"
+          ]
+        }
+      ],
+      "source": [
+        "for i in train_dataloader:\n",
+        "  img0,img1,label = i\n",
+        "  print(img0.size())\n",
+        "  break"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "sMJFC4ZoFaL8",
+        "outputId": "6aba761a-9609-4394-e35c-3f4165a32586"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "torch.Size([1, 1, 250, 25])\n"
+          ]
+        }
+      ],
+      "source": [
+        "for i in img_label_list:\n",
+        "  img0,img1,label = i\n",
+        "  mlist= transform(Image.open(img0))\n",
+        "  mlist = mlist[None,:,:,:]\n",
+        "  print(mlist.size())\n",
+        "  break"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "def train():\n",
+        "    "
+      ]
+    }
+  ],
+  "metadata": {
+    "accelerator": "GPU",
+    "colab": {
+      "provenance": []
+    },
+    "kernelspec": {
+      "display_name": "Python 3",
+      "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": "3.10.9"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}