{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Cnn Cifar10"
   ],
   "id": "site-title"
  },
  {
   "cell_type": "markdown",
   "id": "dde401c3-5b74-4146-8733-2ba0db4293d0",
   "metadata": {},
   "source": "## CNN on Cifar10\n\nWhat would be a deep learning example / tutorial series without an example of image classification eh? In this notebook, we're going to define a small sized CNN and train it on CIFAR10. Actually, this notebook is mostly taken from the pytorch example on the same topic, which you can find [here](https://pytorch.org/tutorials/beginner/blitz/cifar10_tutorial.html). \n\nSo let me give you a brief overview of what we're going to do. \n\n1. Load the `cifar10` dataset from `torchvision`\n2. Define a simple CNN and see how Convolution module works in Flax"
  },
  {
   "cell_type": "markdown",
   "id": "38a2dda6-7189-40d9-9d35-7bdc4416ffb9",
   "metadata": {},
   "source": "### Dataset and Dataloaders"
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "b6c9d099-e9f6-4d6c-ba3a-ca01073d48bd",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/shawon/Projects/jax_examples/venv/lib64/python3.12/site-packages/transformers/utils/generic.py:441: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.\n",
      "  _torch_pytree._register_pytree_node(\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import jax_dataloader as jdl\n",
    "import torchvision\n",
    "import torchvision.transforms as transforms"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "de2c8af5-f623-49ad-9ada-b391e16eeb8c",
   "metadata": {},
   "source": "Now one thing to note here. Torchvision datasets will return images as Torch tensors which are not compatible with Jax. Workaround? Write a transform to convert them to numpy."
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5a779179-6c55-4c1a-b0a6-4e174e46ef4c",
   "metadata": {},
   "outputs": [],
   "source": [
    "class ToNumpy:\n",
    "    def __call__(self, x: torch.Tensor):\n",
    "        return x.numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a20cfe89-8d06-451c-ab97-47b44abb49ff",
   "metadata": {},
   "outputs": [],
   "source": [
    "transform = transforms.Compose(\n",
    "    [transforms.ToTensor(),\n",
    "     transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)), ToNumpy()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2e4d9d64-05e9-4d03-95ac-ecf349ac8903",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 170498071/170498071 [00:08<00:00, 20540710.29it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting ./data/cifar-10-python.tar.gz to ./data\n",
      "Files already downloaded and verified\n"
     ]
    }
   ],
   "source": [
    "trainset = torchvision.datasets.CIFAR10(root='./data', train=True,\n",
    "                                        download=True, transform=transform)\n",
    "\n",
    "testset = torchvision.datasets.CIFAR10(root='./data', train=False,\n",
    "                                       download=True, transform=transform)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8d49cb61-e5ed-4879-ae63-7bd883848d65",
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_size = 128\n",
    "\n",
    "trainloader = jdl.DataLoader(trainset, backend=\"pytorch\", batch_size=batch_size,\n",
    "                                          shuffle=True)\n",
    "testloader = jdl.DataLoader(testset, backend=\"pytorch\", batch_size=batch_size,\n",
    "                                         shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4e07ba73-7aa8-48e7-beae-50ce6bf5ba41",
   "metadata": {},
   "outputs": [],
   "source": [
    "# classes in cifar10\n",
    "classes = ('plane', 'car', 'bird', 'cat',\n",
    "           'deer', 'dog', 'frog', 'horse', 'ship', 'truck')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e8cb5bc-1a8c-45a8-a2b6-25121f1e46ca",
   "metadata": {},
   "source": "### Model"
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "23dfd451-6bb7-4882-88c3-401854fbdc78",
   "metadata": {},
   "outputs": [],
   "source": [
    "import jax\n",
    "import jax.numpy as jnp\n",
    "import numpy as np\n",
    "import flax\n",
    "import flax.linen as nn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "162b01a0-03b9-4bcc-bda2-76d1a7684c8e",
   "metadata": {},
   "outputs": [],
   "source": [
    "from einops import rearrange\n",
    "\n",
    "class ConvNet(nn.Module):\n",
    "    @nn.compact\n",
    "    def __call__(self, x):\n",
    "        # convs\n",
    "        out = nn.Conv(features=6, kernel_size=(5, 5))(x)\n",
    "        out = nn.max_pool(out, window_shape=(2, 2))\n",
    "        out = nn.Conv(features=16, kernel_size=(5, 5))(out)\n",
    "        out = nn.max_pool(out, window_shape=(2, 2))\n",
    "\n",
    "\n",
    "        # flatten into a vector \n",
    "        # skip the batch dim\n",
    "        if len(x.shape) > 3:\n",
    "            out = rearrange(x, \"batch c h w -> batch (c h w)\")\n",
    "        else:\n",
    "            out = out.flatten()\n",
    "\n",
    "        # dense\n",
    "        out = nn.Dense(features=120)(out)\n",
    "        out = nn.Dense(features=84)(out)\n",
    "        out = nn.Dense(features=10)(out)\n",
    "\n",
    "        return out"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf6ed774-4f6d-49b9-9972-e5b8d6595c84",
   "metadata": {},
   "source": "Okay, let's have a look at the `Conv` module in Flax. The `features` parameter is, in my opinion, wrongly named here, and can create confusion but I suppose they had to keep the API consisten in between the Linen modules. \n\nAnyways, the `features` parameter designates the number of output channels in the output of the Convolution layer. Now is it a 2D conv or 1D or 3D? Flax doesn't have separate Conv classes for multi-dim conv, unlike pytorch. So you can just pass in multi dim arrays as long as you define your network properly. If you're wondering what the number of channels in a conv output means, let's say that you want to convolve using `n` filters of `(a, b)` size. The output size of your conv operation will then be `n x a x b`. If you wish to read more on the math behind it, I strongly recommend reading the related chapter from the [Deep Learning](https://www.deeplearningbook.org/contents/convnets.html) book.\n\nNext is the `kernel_size` parameter, which is quite self explanatory. I've used a `5 x 5` size filter. You can fiddle around with different filter sizes and see which one fits your needs. \n\nAs for other parameters, such as stride, dilation, padding, they're set to default as mentioned in the [docs](https://flax.readthedocs.io/en/latest/api_reference/flax.linen/_autosummary/flax.linen.Conv.html). \n\nIf you wish to read more on padding strategies, read [here](https://www.knowledgehut.com/blog/data-science/padding-in-cnn). The default padding strategy in Flax is \"VALID\". \n\nApart from the two conv layers, there are two `max_pool` layers which do max pooling. It is a function, unlike the Conv module, so you have to just pass in the array and the size of the pooling window. Next comes a flatten operation which skips the batch dimension, followed by 3 dense layers to project the model outputs to the number of classes. \n\nOkay, now to do the usual thing, init model and train. "
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e6286835-15a1-473e-b38e-df2b0a7b74b5",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = ConvNet()\n",
    "params = model.init(jax.random.key(0), jnp.empty((3, 32, 32)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "265182e7-c4ff-4634-96d0-bf78e5baa14c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10,)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# run a sample forward pass\n",
    "logits = model.apply(params, jnp.empty((3, 32, 32)))\n",
    "logits.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "50af15aa-e21b-4796-8f4f-f9478031a03a",
   "metadata": {},
   "source": "##### Loss function"
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "5cb762bb-c928-4d95-b785-4b657b84ccaf",
   "metadata": {},
   "outputs": [],
   "source": [
    "import optax\n",
    "\n",
    "@jax.jit\n",
    "def calculate_loss(params, x, y):\n",
    "    logits = model.apply(params, x)\n",
    "    loss = optax.softmax_cross_entropy_with_integer_labels(logits, y)\n",
    "    return loss\n",
    "\n",
    "@jax.jit\n",
    "def batched_loss(params, xs, ys):\n",
    "    batch_loss = jax.vmap(calculate_loss, in_axes=(None, 0, 0))(params, xs, ys)\n",
    "    return batch_loss.mean(axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "bc3391d8-a736-4cec-b365-e5c686e2ef46",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Array(2.3025851, dtype=float32)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "calculate_loss(params, jnp.empty((3, 32, 32)), 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "45bf3c30-075e-4d3b-85e8-34855c3178b7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.3177273\n"
     ]
    }
   ],
   "source": [
    "# check the batched loss thing\n",
    "for batch in trainloader:\n",
    "    loss = batched_loss(params, *batch)\n",
    "    print(loss)\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d12c0b66-ca0e-48f6-9b33-fadc9b21c46b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm.auto import trange, tqdm\n",
    "from flax.training import train_state\n",
    "from functools import partial\n",
    "\n",
    "\n",
    "optimiser = optax.adam(learning_rate=0.001)\n",
    "state = train_state.TrainState.create(\n",
    "    apply_fn=model.apply,\n",
    "    params=params,\n",
    "    tx=optimiser\n",
    ")\n",
    "criterion = jax.value_and_grad(batched_loss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a4e9a9ab-052f-4f95-ac37-db0c6507b071",
   "metadata": {},
   "outputs": [],
   "source": [
    "@jax.jit\n",
    "def train_step(state, batch):\n",
    "    loss_value, grads = criterion(state.params, *batch)\n",
    "    updated_state = state.apply_gradients(grads=grads)\n",
    "    return loss_value, updated_state"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6d433b4-ce90-4561-9f9e-aeb4b969f2c8",
   "metadata": {},
   "source": "##### Evaluation\n\nIt's a classification task, so let's just check on the f1 score for now. "
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "e831bea1-6ca9-4fed-b41d-eaa03d586db3",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import f1_score\n",
    "\n",
    "@jax.jit\n",
    "def test_step(state, xs):\n",
    "    def infer(params, x):\n",
    "        logits = model.apply(params, x)\n",
    "        return jax.nn.softmax(logits, axis=-1) \n",
    "\n",
    "    preds = jax.vmap(jax.jit(infer), in_axes=(None, 0))(state.params, xs)\n",
    "    return preds\n",
    "\n",
    "\n",
    "\n",
    "def evaluate(state, test_loader):\n",
    "    scores = list()\n",
    "    for batch in tqdm(test_loader):\n",
    "        xs, ys = batch\n",
    "        preds = test_step(state, xs)\n",
    "        preds = jnp.argmax(preds, axis=-1)\n",
    "        f1 = f1_score(preds, ys, average=\"micro\")\n",
    "        scores.append(f1)\n",
    "\n",
    "    return np.array(scores).mean(axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "938f2a2e-a5b8-4b01-86b4-75c826d501d6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f4a29b6a237148949046ddf61341aa7b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.10808939873417721"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# test the evaluation function\n",
    "evaluate(state, testloader)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "26435602-bd07-4537-bfc4-3d1ae2f3e76c",
   "metadata": {},
   "source": "Since we're using scikit-learn for evaluation anyway, it will also be nice if we can get a detailed classification report."
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "b6633801-58d4-4bee-aa31-053723031ed2",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "\n",
    "\n",
    "def custom_classification_report(params, test_loader):\n",
    "    preds = list()\n",
    "    actual = list()\n",
    "    for batch in tqdm(test_loader):\n",
    "        xs, ys = batch\n",
    "        pred = test_step(state, xs)\n",
    "        pred = jnp.argmax(pred, axis=-1).tolist()\n",
    "        preds.extend(pred)\n",
    "        actual.extend(ys.tolist())\n",
    "\n",
    "    clf = classification_report(preds, actual, target_names=classes)\n",
    "    print(clf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "f604de31-3f14-4782-95d3-8c0cbed0a4be",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4d168af542ab4c10bb9896802b099b2d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       plane       0.00      0.17      0.00         6\n",
      "         car       0.45      0.12      0.19      3799\n",
      "        bird       0.00      0.05      0.00        21\n",
      "         cat       0.02      0.11      0.04       223\n",
      "        deer       0.18      0.08      0.11      2350\n",
      "         dog       0.01      0.16      0.02        57\n",
      "        frog       0.01      0.11      0.01        75\n",
      "       horse       0.17      0.09      0.12      1923\n",
      "        ship       0.25      0.16      0.20      1541\n",
      "       truck       0.00      0.00      0.00         5\n",
      "\n",
      "    accuracy                           0.11     10000\n",
      "   macro avg       0.11      0.10      0.07     10000\n",
      "weighted avg       0.28      0.11      0.15     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "custom_classification_report(state.params, testloader)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13d89990-9a91-4ed0-9881-bc6f3fcb772d",
   "metadata": {},
   "source": "Perfect. Now we can move to the train loop and add the evaluation there as well so that we can check how the f1 score changes during training."
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "173d4d62-e222-40cf-9d3a-2bd8ed2209ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm.auto import trange\n",
    "\n",
    "def train(state, epochs, train_loader, test_loader):\n",
    "    steps = 0\n",
    "    losses = []\n",
    "    f1s = []\n",
    "\n",
    "    # =============\n",
    "    for e in trange(5):\n",
    "        for batch in tqdm(train_loader):\n",
    "            loss, state = train_step(state, batch)\n",
    "            steps += 1\n",
    "\n",
    "            # log every 200 steps\n",
    "            if steps % 200 == 0:\n",
    "                losses.append(loss)\n",
    "                \n",
    "                # run evaluation\n",
    "                print(\"Evaluating ... \")\n",
    "                score = evaluate(state, test_loader)\n",
    "                \n",
    "                f1s.append(score)\n",
    "    \n",
    "                print(f\"Epoch : {e + 1} :: Step : {steps} :: Loss : {loss}  :: F1 : {score}\")\n",
    "\n",
    "    # ============\n",
    "    return state, losses, f1s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "030f6f56-e33c-447f-883e-f3d130abbf91",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "06cf629546594c469fb2d9619bdf45c1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/5 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "baae6846a88846e58fa598313507057f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/391 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating ... \n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "db5426ea7a7c4f959662a04ffec807c3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 1 :: Step : 200 :: Loss : 1.420501708984375  :: F1 : 0.4317642405063291\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3e3bda930e0546f8a32dbe8e5c4bf79a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/391 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating ... \n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f246e9dd736462490c1b50972e5b3ba",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 2 :: Step : 400 :: Loss : 1.4548004865646362  :: F1 : 0.4642009493670886\n",
      "Evaluating ... \n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8e5c5476bbc8481f8436ec915f7bc4f6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 2 :: Step : 600 :: Loss : 1.6473182439804077  :: F1 : 0.48951740506329117\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "01aa43abc7764f7e9d05e81c585c0ca7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/391 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating ... \n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "09be35d855384813b13a0d174a37a076",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 3 :: Step : 800 :: Loss : 1.521551251411438  :: F1 : 0.48358386075949367\n",
      "Evaluating ... \n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dc2fc275fa1f4062be2fc7f7acca7a87",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 3 :: Step : 1000 :: Loss : 1.4645955562591553  :: F1 : 0.49624208860759494\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1bb98356c3c345db8030f967d964b987",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/391 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating ... \n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ccd4653faebd4d21800376d19df02769",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 4 :: Step : 1200 :: Loss : 1.4096702337265015  :: F1 : 0.5100870253164557\n",
      "Evaluating ... \n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1e1813b052ec4977973d49470b926a19",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 4 :: Step : 1400 :: Loss : 1.2829735279083252  :: F1 : 0.5018789556962026\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c1a1f002bd564ee69a6c0bce9c77184b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/391 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating ... \n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2f6024925c1947199e8ce13284cc0faa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 5 :: Step : 1600 :: Loss : 1.3250913619995117  :: F1 : 0.5089992088607594\n",
      "Evaluating ... \n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "817cc5c75d8f49e5be661cc7720ca2f6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 5 :: Step : 1800 :: Loss : 1.2746849060058594  :: F1 : 0.5096914556962026\n"
     ]
    }
   ],
   "source": [
    "state, losses, f1s = train(state, 5, trainloader, testloader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "4f167270-6d8d-4603-b1cf-a1154c1c5de4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f6440434d70>]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 448,
       "width": 855
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "fig = plt.figure(figsize=(10, 5))\n",
    "fig.add_subplot()\n",
    "plt.xlabel(\"Every 200 Steps\")\n",
    "plt.ylabel(\"Loss/Train\")\n",
    "plt.plot(np.arange(len(losses)), losses)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "f2454df0-7e3d-4e4d-a51f-4e3d38ece68a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f64404f2720>]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 448,
       "width": 855
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig2 = plt.figure(figsize=(10, 5))\n",
    "fig2.add_subplot()\n",
    "plt.xlabel(\"Every 200 Steps\")\n",
    "plt.ylabel(\"f1 score\")\n",
    "plt.plot(np.arange(len(f1s)), f1s)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf854206-3e08-4524-940a-6c2384c204af",
   "metadata": {},
   "source": "##### Class wise classification results"
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "4382d168-c41c-489c-bc56-b1414070dbd7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7bc6bd5256334423bc6a1fcb0836d53d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/79 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       plane       0.50      0.59      0.54       844\n",
      "         car       0.61      0.63      0.62       970\n",
      "        bird       0.44      0.38      0.41      1157\n",
      "         cat       0.33      0.38      0.35       862\n",
      "        deer       0.46      0.43      0.44      1076\n",
      "         dog       0.37      0.51      0.43       732\n",
      "        frog       0.61      0.54      0.57      1134\n",
      "       horse       0.59      0.62      0.60       955\n",
      "        ship       0.74      0.57      0.64      1308\n",
      "       truck       0.55      0.57      0.56       962\n",
      "\n",
      "    accuracy                           0.52     10000\n",
      "   macro avg       0.52      0.52      0.52     10000\n",
      "weighted avg       0.53      0.52      0.52     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "custom_classification_report(state.params, testloader)"
   ]
  }
 ],
 "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": "3.12.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}