ViT Scratch Train.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e999f3b2-6030-4067-9e45-65dcee87dab7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch \n",
    "import os\n",
    "import numpy as np \n",
    "import cv2\n",
    "from torchvision.models.vision_transformer import VisionTransformer\n",
    "from ViT import ViT\n",
    "from torch import Tensor\n",
    "from torch import nn,optim\n",
    "from torch.utils.data import DataLoader,TensorDataset\n",
    "import torch\n",
    "from torch.utils.data import DataLoader\n",
    "from tqdm import tqdm\n",
    "\n",
    "if torch.cuda.is_available():\n",
    "    device = \"cuda\"\n",
    "else:\n",
    "    device = \"cpu\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "212f5ed7-2f6b-4620-a7cc-bc6264432402",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open(\"Features/img_data.pkl\", \"rb\") as f:\n",
    "    image_data = pickle.load(f)\n",
    "with open (\"Features/labels.pkl\",\"rb\") as f:\n",
    "    labels = pickle.load(f)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "1a105062-816a-4a5b-a5f2-2bb1c89a1152",
   "metadata": {},
   "outputs": [],
   "source": [
    "IMG_SIZE = 224\n",
    "BATCH_SIZE = 32\n",
    "EPOCHS = 50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "496cea97-90bd-473a-b6bf-ccb7e3b7c8d0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1000, 224, 224, 3), (1000,))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "first_500_images = image_data[:500]\n",
    "last_500_images = image_data[-500:]\n",
    "images_subset = np.concatenate([first_500_images,last_500_images],axis = 0)\n",
    "\n",
    "first_500_labels = labels[:500]\n",
    "last_500_labels = labels[-500:]\n",
    "labels_subset = np.concatenate([first_500_labels,last_500_labels],axis = 0)\n",
    "\n",
    "images_subset.shape, labels_subset.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "9285fb5a-5e5a-4964-af18-9cc9918cbaf7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1000, 224, 224, 3), (1000,))"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val_first_500_images = image_data[500:1000]\n",
    "val_last_500_images = image_data[-1000:-500]\n",
    "val_images_subset = np.concatenate([val_first_500_images,val_last_500_images],axis = 0)\n",
    "\n",
    "val_first_500_labels = labels[500:1000]\n",
    "val_last_500_labels = labels[-1000:-500]\n",
    "val_labels_subset = np.concatenate([val_first_500_labels, val_last_500_labels], axis=0)\n",
    "\n",
    "val_images_subset.shape, val_labels_subset.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "619c9a88-fcf4-4eae-9efe-69ac8fad8459",
   "metadata": {},
   "outputs": [],
   "source": [
    "image_data_tensors = Tensor(images_subset)\n",
    "labels_tensors = Tensor(labels_subset)\n",
    "val_image_data_tensors = Tensor(val_images_subset)\n",
    "val_labels_tensors = Tensor(val_labels_subset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "470f8909-c00e-4863-a6a3-3568b867b073",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch.nn.functional as F\n",
    "labels_tensors = labels_tensors.to(torch.int64)\n",
    "labels_tensors = F.one_hot(labels_tensors, num_classes=2).float()\n",
    "val_labels_tensors = val_labels_tensors.to(torch.int64)\n",
    "val_labels_tensors = F.one_hot(val_labels_tensors,num_classes=2).float()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3203987f-fce9-404e-9d0b-a04f44bb8f53",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([[1., 0.],\n",
       "         [1., 0.],\n",
       "         [1., 0.],\n",
       "         ...,\n",
       "         [0., 1.],\n",
       "         [0., 1.],\n",
       "         [0., 1.]]),\n",
       " tensor([[1., 0.],\n",
       "         [1., 0.],\n",
       "         [1., 0.],\n",
       "         ...,\n",
       "         [0., 1.],\n",
       "         [0., 1.],\n",
       "         [0., 1.]]))"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels_tensors, val_labels_tensors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "b9474ceb-b2de-49f9-a4f4-cd0fee074fd3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of the Image Data : torch.Size([1000, 224, 224, 3])\n",
      "Shape of the Labels : torch.Size([1000, 2])\n",
      "--------------------------------------------------------\n",
      "Shape of the Validation Image Data : torch.Size([1000, 224, 224, 3])\n",
      "Shape of the Validation Labels : torch.Size([1000, 2])\n"
     ]
    }
   ],
   "source": [
    "print(f\"Shape of the Image Data : {image_data_tensors.shape}\")\n",
    "print(f\"Shape of the Labels : {labels_tensors.shape}\")\n",
    "print(\"--------------------------------------------------------\")\n",
    "print(f\"Shape of the Validation Image Data : {val_image_data_tensors.shape}\")\n",
    "print(f\"Shape of the Validation Labels : {val_labels_tensors.shape}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "1483144a-9549-4722-9ab7-9b1f0c05d5d5",
   "metadata": {},
   "outputs": [],
   "source": [
    "from ViT import ViT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "d8caa857-f5b1-4e22-b595-d7ffa3931948",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ViT(\n",
       "  (patch_embedding): patch_Embedding(\n",
       "    (projection): Sequential(\n",
       "      (0): Rearrange('b c (h p1) (w p2) -> b (h w) (p1 p2 c)', p1=16, p2=16)\n",
       "      (1): Linear(in_features=768, out_features=32, bias=True)\n",
       "    )\n",
       "  )\n",
       "  (layers): ModuleList(\n",
       "    (0-6): 7 x Sequential(\n",
       "      (0): ResidualAdd(\n",
       "        (fn): PreNorm(\n",
       "          (fn): Attension(\n",
       "            (attension): MultiheadAttention(\n",
       "              (out_proj): NonDynamicallyQuantizableLinear(in_features=32, out_features=32, bias=True)\n",
       "            )\n",
       "            (q): Linear(in_features=32, out_features=32, bias=True)\n",
       "            (k): Linear(in_features=32, out_features=32, bias=True)\n",
       "            (v): Linear(in_features=32, out_features=32, bias=True)\n",
       "          )\n",
       "          (norm): LayerNorm((32,), eps=1e-05, elementwise_affine=True)\n",
       "        )\n",
       "      )\n",
       "      (1): ResidualAdd(\n",
       "        (fn): PreNorm(\n",
       "          (fn): FeedForward(\n",
       "            (fc1): Linear(in_features=32, out_features=32, bias=True)\n",
       "            (gelu): GELU(approximate='none')\n",
       "            (dropout1): Dropout(p=0.1, inplace=False)\n",
       "            (fc2): Linear(in_features=32, out_features=32, bias=True)\n",
       "            (dropout2): Dropout(p=0.1, inplace=False)\n",
       "          )\n",
       "          (norm): LayerNorm((32,), eps=1e-05, elementwise_affine=True)\n",
       "        )\n",
       "      )\n",
       "    )\n",
       "  )\n",
       "  (head): Sequential(\n",
       "    (0): LayerNorm((32,), eps=1e-05, elementwise_affine=True)\n",
       "    (1): Linear(in_features=32, out_features=2, bias=True)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vit_model = ViT(output_dim=2)\n",
    "vit_model.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "5f43d65a-7a38-4c78-b565-52229bf959e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "LOSS_FUNCTION = torch.nn.CrossEntropyLoss()\n",
    "OPTIMIZER = torch.optim.Adam(vit_model.parameters(), lr=1e-5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "019f5910-2673-4025-93b0-66a4b093cdae",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = torch.utils.data.TensorDataset(image_data_tensors, labels_tensors)\n",
    "data_loader = torch.utils.data.DataLoader(dataset, batch_size=BATCH_SIZE, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0268f85f-d732-4622-8a31-3beffcb06ac1",
   "metadata": {},
   "outputs": [],
   "source": [
    "val_dataset = torch.utils.data.TensorDataset(val_image_data_tensors, val_labels_tensors)\n",
    "val_data_loader = torch.utils.data.DataLoader(val_dataset, batch_size = BATCH_SIZE, shuffle = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "91bce04b-623d-447b-8d56-3b7df7e2afb6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 1/50 [Train]: 100%|██████████| 32/32 [00:01<00:00, 23.60it/s]\n",
      "Epoch 1/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 98.91it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50, Train Loss: 0.7015, Train Acc: 0.4880, Val Loss: 0.6936, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 2/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 42.90it/s]\n",
      "Epoch 2/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 110.90it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2/50, Train Loss: 0.6981, Train Acc: 0.5120, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 3/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 42.69it/s]\n",
      "Epoch 3/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 112.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3/50, Train Loss: 0.6941, Train Acc: 0.5060, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 4/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 42.60it/s]\n",
      "Epoch 4/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 111.83it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4/50, Train Loss: 0.7008, Train Acc: 0.4990, Val Loss: 0.6933, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 5/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 43.40it/s]\n",
      "Epoch 5/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 111.78it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5/50, Train Loss: 0.6930, Train Acc: 0.5020, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 6/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 43.61it/s]\n",
      "Epoch 6/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 120.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6/50, Train Loss: 0.6986, Train Acc: 0.5040, Val Loss: 0.6933, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 7/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.26it/s]\n",
      "Epoch 7/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 116.90it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7/50, Train Loss: 0.7050, Train Acc: 0.4690, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 8/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.40it/s]\n",
      "Epoch 8/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 113.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8/50, Train Loss: 0.6963, Train Acc: 0.5080, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 9/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.53it/s]\n",
      "Epoch 9/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 118.59it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9/50, Train Loss: 0.6943, Train Acc: 0.5220, Val Loss: 0.6936, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 10/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.41it/s]\n",
      "Epoch 10/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 115.98it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10/50, Train Loss: 0.6943, Train Acc: 0.4960, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 11/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.52it/s]\n",
      "Epoch 11/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 118.18it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11/50, Train Loss: 0.7036, Train Acc: 0.4760, Val Loss: 0.6933, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 12/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.64it/s]\n",
      "Epoch 12/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 118.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12/50, Train Loss: 0.7023, Train Acc: 0.4890, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 13/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 43.33it/s]\n",
      "Epoch 13/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 117.12it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13/50, Train Loss: 0.6980, Train Acc: 0.5000, Val Loss: 0.6933, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 14/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.70it/s]\n",
      "Epoch 14/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 119.37it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14/50, Train Loss: 0.7045, Train Acc: 0.4670, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 15/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 42.35it/s]\n",
      "Epoch 15/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 119.03it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15/50, Train Loss: 0.6977, Train Acc: 0.4890, Val Loss: 0.6934, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 16/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 43.08it/s]\n",
      "Epoch 16/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 120.09it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16/50, Train Loss: 0.6929, Train Acc: 0.5110, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 17/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.44it/s]\n",
      "Epoch 17/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 118.11it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17/50, Train Loss: 0.7019, Train Acc: 0.4810, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 18/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.42it/s]\n",
      "Epoch 18/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 115.83it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18/50, Train Loss: 0.7014, Train Acc: 0.4890, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 19/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.66it/s]\n",
      "Epoch 19/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 118.25it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19/50, Train Loss: 0.6934, Train Acc: 0.5220, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 20/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.30it/s]\n",
      "Epoch 20/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 114.62it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20/50, Train Loss: 0.6989, Train Acc: 0.4990, Val Loss: 0.6936, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 21/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 43.21it/s]\n",
      "Epoch 21/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 115.22it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21/50, Train Loss: 0.7035, Train Acc: 0.4910, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 22/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.20it/s]\n",
      "Epoch 22/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 117.54it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22/50, Train Loss: 0.6953, Train Acc: 0.5190, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 23/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.98it/s]\n",
      "Epoch 23/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 115.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23/50, Train Loss: 0.7007, Train Acc: 0.4840, Val Loss: 0.6934, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 24/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.85it/s]\n",
      "Epoch 24/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 120.42it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24/50, Train Loss: 0.6949, Train Acc: 0.4990, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 25/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.54it/s]\n",
      "Epoch 25/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 118.38it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 25/50, Train Loss: 0.6984, Train Acc: 0.5010, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 26/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.01it/s]\n",
      "Epoch 26/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 116.43it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 26/50, Train Loss: 0.6993, Train Acc: 0.4900, Val Loss: 0.6934, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 27/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.55it/s]\n",
      "Epoch 27/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 116.71it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 27/50, Train Loss: 0.7032, Train Acc: 0.4890, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 28/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.32it/s]\n",
      "Epoch 28/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 120.39it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 28/50, Train Loss: 0.6982, Train Acc: 0.4720, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 29/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.59it/s]\n",
      "Epoch 29/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 106.93it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 29/50, Train Loss: 0.6948, Train Acc: 0.5190, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 30/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.36it/s]\n",
      "Epoch 30/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 118.39it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 30/50, Train Loss: 0.6939, Train Acc: 0.4970, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 31/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.42it/s]\n",
      "Epoch 31/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 117.43it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 31/50, Train Loss: 0.6985, Train Acc: 0.4920, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 32/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.43it/s]\n",
      "Epoch 32/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 120.89it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 32/50, Train Loss: 0.6949, Train Acc: 0.5100, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 33/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.23it/s]\n",
      "Epoch 33/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 118.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 33/50, Train Loss: 0.7000, Train Acc: 0.4850, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 34/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.43it/s]\n",
      "Epoch 34/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 121.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 34/50, Train Loss: 0.6970, Train Acc: 0.4940, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 35/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.26it/s]\n",
      "Epoch 35/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 120.97it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 35/50, Train Loss: 0.6947, Train Acc: 0.4990, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 36/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.15it/s]\n",
      "Epoch 36/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 121.21it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 36/50, Train Loss: 0.6960, Train Acc: 0.5040, Val Loss: 0.6933, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 37/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 43.56it/s]\n",
      "Epoch 37/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 110.73it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 37/50, Train Loss: 0.6932, Train Acc: 0.5110, Val Loss: 0.6934, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 38/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 42.93it/s]\n",
      "Epoch 38/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 120.66it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 38/50, Train Loss: 0.7010, Train Acc: 0.4840, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 39/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.84it/s]\n",
      "Epoch 39/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 119.84it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 39/50, Train Loss: 0.6984, Train Acc: 0.4980, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 40/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.18it/s]\n",
      "Epoch 40/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 120.13it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 40/50, Train Loss: 0.6974, Train Acc: 0.5110, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 41/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.47it/s]\n",
      "Epoch 41/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 119.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 41/50, Train Loss: 0.6972, Train Acc: 0.5020, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 42/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.95it/s]\n",
      "Epoch 42/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 118.18it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 42/50, Train Loss: 0.6990, Train Acc: 0.4630, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 43/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.68it/s]\n",
      "Epoch 43/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 117.24it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 43/50, Train Loss: 0.6987, Train Acc: 0.4850, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 44/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.50it/s]\n",
      "Epoch 44/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 119.60it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 44/50, Train Loss: 0.6914, Train Acc: 0.5480, Val Loss: 0.6936, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 45/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.17it/s]\n",
      "Epoch 45/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 119.82it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 45/50, Train Loss: 0.6950, Train Acc: 0.5000, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 46/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.87it/s]\n",
      "Epoch 46/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 119.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 46/50, Train Loss: 0.6981, Train Acc: 0.4760, Val Loss: 0.6933, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 47/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.57it/s]\n",
      "Epoch 47/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 121.38it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 47/50, Train Loss: 0.6962, Train Acc: 0.5090, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 48/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 44.60it/s]\n",
      "Epoch 48/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 121.12it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 48/50, Train Loss: 0.6963, Train Acc: 0.5120, Val Loss: 0.6931, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 49/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.48it/s]\n",
      "Epoch 49/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 120.25it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 49/50, Train Loss: 0.6953, Train Acc: 0.5110, Val Loss: 0.6933, Val Acc: 0.5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 50/50 [Train]: 100%|██████████| 32/32 [00:00<00:00, 45.18it/s]\n",
      "Epoch 50/50 [Val]: 100%|██████████| 32/32 [00:00<00:00, 107.92it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 50/50, Train Loss: 0.6965, Train Acc: 0.4970, Val Loss: 0.6932, Val Acc: 0.5000\n"
     ]
    }
   ],
   "source": [
    "history = {\"train_loss\": [], \"val_loss\": [], \"train_acc\": [], \"val_acc\": []}\n",
    "\n",
    "for epoch in range(EPOCHS):\n",
    "    # Training phase\n",
    "    vit_model.train()\n",
    "    train_loss = 0.0\n",
    "    correct_train = 0\n",
    "    total_train = 0\n",
    "\n",
    "    for images, labels in tqdm(data_loader, desc=f\"Epoch {epoch+1}/{EPOCHS} [Train]\"):\n",
    "        images, labels = images.to(device), labels.to(device)\n",
    "        images = images.permute(0, 3, 1, 2)  # Ensure correct shape for model input\n",
    "\n",
    "        OPTIMIZER.zero_grad()  # Clear gradients\n",
    "        outputs = vit_model(images)  # Forward pass\n",
    "\n",
    "        # Compute loss\n",
    "        loss = LOSS_FUNCTION(outputs, labels)\n",
    "        loss.backward()  # Backpropagation\n",
    "        OPTIMIZER.step()  # Update weights\n",
    "\n",
    "        # Update training metrics\n",
    "        train_loss += loss.item()\n",
    "        correct_train += (outputs.argmax(dim=1) == labels.argmax(dim=1)).sum().item()\n",
    "        total_train += labels.size(0)\n",
    "\n",
    "    train_loss /= len(data_loader)\n",
    "    train_acc = correct_train / total_train\n",
    "    history[\"train_loss\"].append(train_loss)\n",
    "    history[\"train_acc\"].append(train_acc)\n",
    "\n",
    "    # Validation phase\n",
    "    vit_model.eval()\n",
    "    val_loss = 0.0\n",
    "    correct_val = 0\n",
    "    total_val = 0\n",
    "\n",
    "    with torch.no_grad():\n",
    "        for images, labels in tqdm(val_data_loader, desc=f\"Epoch {epoch+1}/{EPOCHS} [Val]\"):\n",
    "            images, labels = images.to(device), labels.to(device)\n",
    "            images = images.permute(0, 3, 1, 2)\n",
    "\n",
    "            outputs = vit_model(images)\n",
    "\n",
    "            # Compute loss\n",
    "            loss = LOSS_FUNCTION(outputs, labels)\n",
    "            val_loss += loss.item()\n",
    "\n",
    "            # Update validation metrics\n",
    "            correct_val += (outputs.argmax(dim=1) == labels.argmax(dim=1)).sum().item()\n",
    "            total_val += labels.size(0)\n",
    "\n",
    "    val_loss /= len(val_data_loader)\n",
    "    val_acc = correct_val / total_val\n",
    "    history[\"val_loss\"].append(val_loss)\n",
    "    history[\"val_acc\"].append(val_acc)\n",
    "\n",
    "    # Print epoch summary\n",
    "    print(f\"Epoch {epoch+1}/{EPOCHS}, Train Loss: {train_loss:.4f}, Train Acc: {train_acc:.4f}, \"\n",
    "          f\"Val Loss: {val_loss:.4f}, Val Acc: {val_acc:.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62b8da8e-be4a-4542-8489-6b3a0d16ce9c",
   "metadata": {},
   "source": [
    "accuarcy = 54.8\n",
    "loss = 69.14"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "999cae9a-69ca-4be0-b634-9506d1f5c417",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "3ad8d680-939f-4104-9765-e2a52797da8e",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(history)\n",
    "df[\"Epochs\"] = np.arange(1,51)\n",
    "df.to_csv(\"History/scratch_vit_history.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "3bef1b05-9170-47ad-b74b-e2f641b6850a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.548"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"train_acc\"].max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "32355ec2-e3ff-4632-a605-0ac81d84463e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6914036143571138"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"train_loss\"].min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "ee9eaa1e-287b-4af7-8cce-b469c7203247",
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.save(vit_model, \"ViT Model Pytorch.pth\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "a049e155-05ea-4f66-b696-f380efc515e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(x = df[\"Epochs\"],y = df[\"train_acc\"])\n",
    "plt.ylabel(\"Accuacy\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "708f1de8-fefe-4071-b361-d3d66c528460",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(x = df[\"Epochs\"],y = df[\"train_loss\"])\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}