14.Pytorch Deep Learning Model Tracking using Tensorboard.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "source": [
    "**TensorBoard Model Summary using Pytorch**\r\n",
    "---\r\n",
    "1. How to use PyTorch with TensorBoard for tracking your deep learning project?\r\n",
    "1. We will use colored CIFAR10 images.\r\n",
    "1. How to correctly visualize colored images in TensorBoard?\r\n",
    "1. How to track training loss and accuracy of your deep learning project using TensorBoard?\r\n",
    "\r\n",
    "#### Installing Tensorboard\r\n",
    "\r\n",
    "```bash\r\n",
    "!pip install tensorboard\r\n",
    "```\r\n",
    "\r\n"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "**1.Importing Modules**\r\n",
    "---"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "source": [
    "# torch loaded...!!\r\n",
    "import torch\r\n",
    "import torch.nn as nn\r\n",
    "import torch.nn.functional as F\r\n",
    "import torch.optim as optim\r\n",
    "from torch.utils.tensorboard import SummaryWriter\r\n",
    "from torchsummary import summary\r\n",
    "\r\n",
    "# torchvision loaded...!!\r\n",
    "import torchvision\r\n",
    "from torchvision import datasets, transforms\r\n",
    "\r\n",
    "# other module loaded...!!\r\n",
    "import numpy as np\r\n",
    "import matplotlib.pyplot as plt\r\n",
    "%matplotlib inline\r\n",
    "from tqdm import tqdm\r\n"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "Some of the important modules include:\r\n",
    "\r\n",
    "* `SummaryWriter`: this is the TensorBoard module that PyTorch provides. We will be able to access all its functionalities after creating an object of SummaryWriter.\r\n",
    "* `torchvision`: we can download PyTorch datasets and access the transforms using this module.\r\n",
    "* `torch.nn`: provides all the neural network layers that we will need.\r\n",
    "\r\n",
    "**2.Define Data Transformatiopn and prepare the Dataset**\r\n",
    "---"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "source": [
    "transform = transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "source": [
    "# dataset downloaded...!!\r\n",
    "trainset = datasets.CIFAR10(root='../data_cifar10', train=True, download=True, transform=transform)\r\n",
    "testset = datasets.CIFAR10(root='../data_cifar10', train=False, download=True, transform=transform)\r\n",
    "\r\n",
    "# Data loader...!!\r\n",
    "trainLoader = torch.utils.data.DataLoader(trainset, batch_size=4, shuffle=True, num_workers=2)\r\n",
    "testLoader = torch.utils.data.DataLoader(testset, batch_size=4, shuffle=False, num_workers=2)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "**3.CIFAR 10 Classes**\r\n",
    "---"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "source": [
    "classes = ['plane','car','bird','cat','deer','dog','frog','horse','ship','truck']"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "**4. Setting Up Tensorboard**\r\n",
    "---"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "source": [
    "writer = SummaryWriter('runs/cifar10')"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "**5. Adding Images to tensorboard**\r\n",
    "---"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "source": [
    "def show_img(image):\r\n",
    "    image = image / 2 + 0.5\r\n",
    "    npimage = image.numpy()\r\n",
    "    plt.imshow(np.transpose(npimage, (1, 2, 0)))\r\n",
    "    return npimage"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "* In the above show_img() function (takes an image tensor as a parameter), first, we unnormalize the images. \r\n",
    "* This is a very important step for TensorBoard visualization. If we do not unnormalize the images, then the images will be a lot noisy with random pixels. Then we convert the images to NumPy array and plot the images. \r\n",
    "* Finally, we return the unnormalized images. We return the image here, otherwise, we will again have to unnormalize them before passing them to add_img(). That would be a repetition of code and we want to avoid that."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "source": [
    "# get images\r\n",
    "dataiter = iter(trainLoader)\r\n",
    "images, labels = dataiter.next()\r\n",
    "\r\n",
    "# create grid of images\r\n",
    "img_grid = torchvision.utils.make_grid(images)\r\n",
    "\r\n",
    "# get and show the unnormalized images\r\n",
    "img_grid = show_img(img_grid)\r\n",
    "\r\n",
    "# write to tensorboard\r\n",
    "writer.add_image('Cifar10 Image', img_grid)"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Open the tensorboard in cmd\r\n",
    "\r\n",
    "```bash\r\n",
    "tensorboard --logdir=runs\r\n",
    "```"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "**6.Define the Deep Neural Network Graph to Tensorboard**\r\n",
    "---"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "source": [
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\r\n",
    "class Net(nn.Module):\r\n",
    "    def __init__(self):\r\n",
    "        super(Net, self).__init__()\r\n",
    "        self.conv1 = nn.Conv2d(3, 6, 5)\r\n",
    "        self.pool = nn.MaxPool2d(2, 2)\r\n",
    "        self.conv2 = nn.Conv2d(6, 16, 5)\r\n",
    "        self.fc1 = nn.Linear(16 * 5 * 5, 120)\r\n",
    "        self.fc2 = nn.Linear(120, 84)\r\n",
    "        self.fc3 = nn.Linear(84, 10)\r\n",
    "    def forward(self, x):\r\n",
    "        x = self.pool(F.relu(self.conv1(x)))\r\n",
    "        x = self.pool(F.relu(self.conv2(x)))\r\n",
    "        x = x.view(-1, 16 * 5 * 5)\r\n",
    "        x = F.relu(self.fc1(x))\r\n",
    "        x = F.relu(self.fc2(x))\r\n",
    "        x = self.fc3(x)\r\n",
    "        return x\r\n",
    "net = Net().to(device)\r\n",
    "criterion = nn.CrossEntropyLoss()\r\n",
    "optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "source": [
    "summary(net, (3, 32, 32))"
   ],
   "outputs": [
    {
     "output_type": "error",
     "ename": "NameError",
     "evalue": "name 'summary' is not defined",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp/ipykernel_2024/2635738260.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0msummary\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mnet\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m(\u001b[0m\u001b[1;36m3\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;36m32\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;36m32\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m: name 'summary' is not defined"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "source": [
    "# load the model into the device\r\n",
    "writer.add_graph(net, images.to(device))\r\n",
    "writer.close()"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "**7.Track Model training using tensorboard**\r\n",
    "---"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "source": [
    "for epoch in range(10):\r\n",
    "    running_loss = 0.0\r\n",
    "    running_correct = 0\r\n",
    "    with tqdm(trainLoader, unit=\"batch\") as tepoch:\r\n",
    "        for data in tepoch:\r\n",
    "            tepoch.set_description(f\"Epoch {epoch+1}\")\r\n",
    "            inputs, labels = data[0].to(device), data[1].to(device)\r\n",
    "            optimizer.zero_grad()\r\n",
    "            outputs = net(inputs)\r\n",
    "            loss = criterion(outputs, labels)\r\n",
    "            _, preds = torch.max(outputs.data, 1)\r\n",
    "            loss.backward()\r\n",
    "            optimizer.step()\r\n",
    "            running_loss += loss.item()\r\n",
    "            running_correct += torch.sum(preds == labels.data)\r\n",
    "            tepoch.set_postfix(loss=loss)  \r\n",
    "            \r\n",
    "        writer.add_scalar('loss', running_loss / len(trainLoader), epoch)\r\n",
    "        writer.add_scalar('accuracy', running_correct.double() / len(trainLoader.dataset), epoch)\r\n",
    "        print('epoch: {}, loss: {}, accuracy: {}'.format(epoch, running_loss / len(trainLoader), running_correct.double() / len(trainLoader.dataset)))\r\n",
    "    print('Finished Training')"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "Epoch 1: 100%|██████████| 12500/12500 [03:27<00:00, 60.12batch/s, loss=tensor(0.5355, device='cuda:0', grad_fn=<NllLossBackward>)]\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "epoch: 0, loss: 1.3957625457572937, accuracy: 0.49612000000000006\n",
      "Finished Training\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "Epoch 2: 100%|██████████| 12500/12500 [04:43<00:00, 44.12batch/s, loss=tensor(2.0570, device='cuda:0', grad_fn=<NllLossBackward>)]\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "epoch: 1, loss: 1.2268571854040027, accuracy: 0.5634\n",
      "Finished Training\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "Epoch 3: 100%|██████████| 12500/12500 [04:19<00:00, 48.22batch/s, loss=tensor(1.9112, device='cuda:0', grad_fn=<NllLossBackward>)]\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "epoch: 2, loss: 1.1254645399177075, accuracy: 0.60074\n",
      "Finished Training\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "Epoch 4: 100%|██████████| 12500/12500 [03:16<00:00, 63.74batch/s, loss=tensor(0.9148, device='cuda:0', grad_fn=<NllLossBackward>)] \n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "epoch: 3, loss: 1.0458213572911919, accuracy: 0.63166\n",
      "Finished Training\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "Epoch 5: 100%|██████████| 12500/12500 [02:08<00:00, 97.57batch/s, loss=tensor(0.8473, device='cuda:0', grad_fn=<NllLossBackward>)] \n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "epoch: 4, loss: 0.9887139956446737, accuracy: 0.65242\n",
      "Finished Training\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "Epoch 6: 100%|██████████| 12500/12500 [02:07<00:00, 97.87batch/s, loss=tensor(1.1014, device='cuda:0', grad_fn=<NllLossBackward>)]\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "epoch: 5, loss: 0.9419117592780665, accuracy: 0.6694000000000001\n",
      "Finished Training\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "Epoch 7: 100%|██████████| 12500/12500 [02:24<00:00, 86.69batch/s, loss=tensor(1.2462, device='cuda:0', grad_fn=<NllLossBackward>)]\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "epoch: 6, loss: 0.9060661251644977, accuracy: 0.68042\n",
      "Finished Training\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "Epoch 8: 100%|██████████| 12500/12500 [02:58<00:00, 70.09batch/s, loss=tensor(0.8456, device='cuda:0', grad_fn=<NllLossBackward>)]\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "epoch: 7, loss: 0.86699169231073, accuracy: 0.6929200000000001\n",
      "Finished Training\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "Epoch 9: 100%|██████████| 12500/12500 [02:15<00:00, 92.03batch/s, loss=tensor(0.4044, device='cuda:0', grad_fn=<NllLossBackward>)] \n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "epoch: 8, loss: 0.8372248044571001, accuracy: 0.70494\n",
      "Finished Training\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "Epoch 10: 100%|██████████| 12500/12500 [02:18<00:00, 90.12batch/s, loss=tensor(1.3713, device='cuda:0', grad_fn=<NllLossBackward>)] "
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "epoch: 9, loss: 0.8136869410972949, accuracy: 0.7133\n",
      "Finished Training\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "**8.Tensorboard Results**\r\n",
    "---"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "source": [
    "from IPython.display import Image\r\n",
    "Image('Noise_image\\Screenshot_2.png')"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "execution_count": 58
    }
   ],
   "metadata": {}
  }
 ],
 "metadata": {
  "orig_nbformat": 4,
  "language_info": {
   "name": "python",
   "version": "3.8.0",
   "mimetype": "text/x-python",
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "pygments_lexer": "ipython3",
   "nbconvert_exporter": "python",
   "file_extension": ".py"
  },
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3.8.0 64-bit ('pytorch19': conda)"
  },
  "interpreter": {
   "hash": "884a8f81666a19c0851426c83cd6eaa7b212468ad852fb3caa21591c98d4369f"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}