vgg11_sparsity.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "38e925cf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:44:33.042141Z",
     "iopub.status.busy": "2024-06-05T10:44:33.041847Z",
     "iopub.status.idle": "2024-06-05T10:44:46.418979Z",
     "shell.execute_reply": "2024-06-05T10:44:46.417862Z"
    },
    "papermill": {
     "duration": 13.390634,
     "end_time": "2024-06-05T10:44:46.421424",
     "exception": false,
     "start_time": "2024-06-05T10:44:33.030790",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting zfpy\r\n",
      "  Downloading zfpy-1.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (648 bytes)\r\n",
      "Downloading zfpy-1.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (729 kB)\r\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m729.2/729.2 kB\u001b[0m \u001b[31m8.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\r\n",
      "\u001b[?25h\u001b[33mWARNING: Error parsing requirements for aiohttp: [Errno 2] No such file or directory: '/opt/conda/lib/python3.10/site-packages/aiohttp-3.9.1.dist-info/METADATA'\u001b[0m\u001b[33m\r\n",
      "\u001b[0mInstalling collected packages: zfpy\r\n",
      "Successfully installed zfpy-1.0.0\r\n"
     ]
    }
   ],
   "source": [
    "! pip install zfpy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "483a5fe3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:44:46.444771Z",
     "iopub.status.busy": "2024-06-05T10:44:46.443996Z",
     "iopub.status.idle": "2024-06-05T10:44:53.258599Z",
     "shell.execute_reply": "2024-06-05T10:44:53.257467Z"
    },
    "papermill": {
     "duration": 6.828376,
     "end_time": "2024-06-05T10:44:53.260753",
     "exception": false,
     "start_time": "2024-06-05T10:44:46.432377",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn.utils.prune as prune\n",
    "import torchvision.models as models\n",
    "from torchvision.models import VGG11_Weights\n",
    "import numpy as np\n",
    "import zfpy\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import copy\n",
    "from tabulate import tabulate\n",
    "import pandas as pd\n",
    "import pickle\n",
    "import os\n",
    "import torch.nn as nn\n",
    "from torchvision import datasets, transforms\n",
    "from torch.utils.data.sampler import SubsetRandomSampler\n",
    "from tqdm import tqdm\n",
    "import json\n",
    "from io import StringIO\n",
    "\n",
    "\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "print(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c9c9c650",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:44:53.283687Z",
     "iopub.status.busy": "2024-06-05T10:44:53.283204Z",
     "iopub.status.idle": "2024-06-05T10:44:53.287703Z",
     "shell.execute_reply": "2024-06-05T10:44:53.286892Z"
    },
    "papermill": {
     "duration": 0.01782,
     "end_time": "2024-06-05T10:44:53.289647",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.271827",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "TRAIN = False\n",
    "PRUNE = True\n",
    "PRUNE_LOAD = False\n",
    "TRAIN_LOAD = False\n",
    "TEST_JSON = 'test_data.json'\n",
    "PRUNE_DATA_JSON = 'prune_data.json'\n",
    "MODEL_PATH = '/kaggle/input/190-tolerance/org_model.pkl'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2a447f9d",
   "metadata": {
    "papermill": {
     "duration": 0.009997,
     "end_time": "2024-06-05T10:44:53.309819",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.299822",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Helper Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "dba8c935",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:44:53.331933Z",
     "iopub.status.busy": "2024-06-05T10:44:53.331644Z",
     "iopub.status.idle": "2024-06-05T10:44:53.337403Z",
     "shell.execute_reply": "2024-06-05T10:44:53.336640Z"
    },
    "papermill": {
     "duration": 0.019273,
     "end_time": "2024-06-05T10:44:53.339244",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.319971",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def print_statistics(model):\n",
    "    all_statistics = []\n",
    "    for name, param in model.named_parameters():\n",
    "        if 'weight' in name:\n",
    "            weights = param.data.flatten()\n",
    "            statistics = {\n",
    "                'Layer': name,\n",
    "                'Mean': weights.mean().item(),\n",
    "                'Standard Deviation': weights.std().item(),\n",
    "                'Min': weights.min().item(),\n",
    "                'Max': weights.max().item()\n",
    "            }\n",
    "            all_statistics.append(statistics)\n",
    "    \n",
    "    df = pd.DataFrame(all_statistics)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d5c01a1c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:44:53.361483Z",
     "iopub.status.busy": "2024-06-05T10:44:53.361222Z",
     "iopub.status.idle": "2024-06-05T10:44:53.370424Z",
     "shell.execute_reply": "2024-06-05T10:44:53.369648Z"
    },
    "papermill": {
     "duration": 0.02249,
     "end_time": "2024-06-05T10:44:53.372323",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.349833",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def plot_weight_distribution(model, tolerance=1e-3, base=False, prune=-1, prune_zfp=False, bins=100):\n",
    "    model = model.cpu()\n",
    "    fig, axes = plt.subplots(2, 4, figsize=(16, 8))\n",
    "    \n",
    "    if base:\n",
    "        fig.suptitle(f'Weight Distribution of Original Feature Weights')\n",
    "    elif prune > -1:\n",
    "        if prune_zfp:\n",
    "            fig.suptitle(f'Weight Distribution of Feature Weights (ZFP {100*prune} pruned) (Tolerance {tolerance})')\n",
    "        else:\n",
    "            fig.suptitle(f'Weight Distribution of Feature Weights (Base {100*prune} pruned) (Tolerance {tolerance})')\n",
    "    else:\n",
    "        fig.suptitle(f'Weight Distribution of only Feature Weights (Tolerance {tolerance})')\n",
    "\n",
    "    plt.subplots_adjust(left=0.1, right=0.9, top=0.9, bottom=0.1, wspace=0.4, hspace=0.4)\n",
    "    \n",
    "    with pd.option_context('mode.use_inf_as_na', True):\n",
    "        i = 0\n",
    "        for name, param in model.named_parameters():\n",
    "            if 'weight' in name and 'features' in name:\n",
    "                weights = param.data.flatten()\n",
    "\n",
    "                # Plot histogram for this layer\n",
    "                ax = axes[i//4, i%4]\n",
    "                sns.histplot(weights, bins=bins, kde=True, ax=ax)\n",
    "                ax.set_title(f'{name}')\n",
    "                ax.set_xlabel('Weight Value')\n",
    "                ax.set_ylabel('Frequency')\n",
    "                ax.set_yscale('log')\n",
    "\n",
    "                i += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "901645fd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:44:53.395236Z",
     "iopub.status.busy": "2024-06-05T10:44:53.394988Z",
     "iopub.status.idle": "2024-06-05T10:44:53.403008Z",
     "shell.execute_reply": "2024-06-05T10:44:53.402141Z"
    },
    "papermill": {
     "duration": 0.020792,
     "end_time": "2024-06-05T10:44:53.404871",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.384079",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# classifier.0.weight 102760448\n",
    "# classifier.3.weight 16777216\n",
    "# classifier.6.weight 4096000\n",
    "\n",
    "def plot_classifier_distribution(model, tolerance=1e-3, base=False, prune=-1, prune_zfp=False, bins=100, classifier_num=6):\n",
    "    model = model.cpu()\n",
    "    plt.figure(figsize=(8, 6))\n",
    "    \n",
    "    layer_name = ''\n",
    "    with pd.option_context('mode.use_inf_as_na', True):\n",
    "        for name, param in model.named_parameters():\n",
    "            if 'classifier' in name and str(classifier_num) in name:\n",
    "                weights = param.data.flatten()\n",
    "\n",
    "                # Plot histogram for this layer\n",
    "                sns.histplot(weights, bins=bins, kde=True)\n",
    "                layer_name = name\n",
    "                plt.xlabel('Weight Value')\n",
    "                plt.ylabel('Frequency')\n",
    "                plt.yscale('log')\n",
    "\n",
    "                break\n",
    "        \n",
    "    if base:\n",
    "        plt.title(f'Weight Distribution of {layer_name}')\n",
    "    elif prune > -1:\n",
    "        if prune_zfp:\n",
    "            plt.title(f'Weight Distribution of {layer_name} (ZFP {100*prune} pruned) (Tolerance {tolerance})')\n",
    "        else:\n",
    "            plt.title(f'Weight Distribution of {layer_name} (Base {100*prune} pruned) (Tolerance {tolerance})')\n",
    "    else:\n",
    "        plt.title(f'Weight Distribution of {layer_name} (Tolerance {tolerance})')\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c992dd97",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:44:53.426429Z",
     "iopub.status.busy": "2024-06-05T10:44:53.426158Z",
     "iopub.status.idle": "2024-06-05T10:44:53.431712Z",
     "shell.execute_reply": "2024-06-05T10:44:53.430794Z"
    },
    "papermill": {
     "duration": 0.018483,
     "end_time": "2024-06-05T10:44:53.433509",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.415026",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def compare_model_stats(model, copied_model):\n",
    "    res = pd.DataFrame()\n",
    "    res['Layer'] = model['Layer']\n",
    "    res['MSE Mean'] = (model['Mean'] - copied_model['Mean'])**2\n",
    "    res['MSE SD'] = (model['Standard Deviation'] - copied_model['Standard Deviation'])**2\n",
    "    res['MSE Min'] = (model['Min'] - copied_model['Min'])**2\n",
    "    res['MSE Max'] = (model['Max'] - copied_model['Max'])**2\n",
    "    \n",
    "    return res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "6aec17a5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:44:53.455104Z",
     "iopub.status.busy": "2024-06-05T10:44:53.454830Z",
     "iopub.status.idle": "2024-06-05T10:44:53.459862Z",
     "shell.execute_reply": "2024-06-05T10:44:53.459066Z"
    },
    "papermill": {
     "duration": 0.018128,
     "end_time": "2024-06-05T10:44:53.461803",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.443675",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def save_pkl(file: str, state_dict: dict) -> float:\n",
    "    with open(file, 'wb') as f:\n",
    "        pickle.dump(state_dict, f)\n",
    "    return os.path.getsize(file)\n",
    "        \n",
    "def load_pkl(file: str) -> dict:\n",
    "    with open(file, 'rb') as f:\n",
    "        return pickle.load(f)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0e44c8e2",
   "metadata": {
    "papermill": {
     "duration": 0.010231,
     "end_time": "2024-06-05T10:44:53.482329",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.472098",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Compress and Decompress functions using ZFP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "907ca2ca",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:44:53.504370Z",
     "iopub.status.busy": "2024-06-05T10:44:53.504087Z",
     "iopub.status.idle": "2024-06-05T10:44:53.509197Z",
     "shell.execute_reply": "2024-06-05T10:44:53.508350Z"
    },
    "papermill": {
     "duration": 0.018181,
     "end_time": "2024-06-05T10:44:53.511024",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.492843",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Function to compress tensor using ZFP\n",
    "def compress_tensor(tensor, tolerance=1e-3):\n",
    "    array = tensor.cpu().numpy()  # Convert tensor to numpy array\n",
    "    compressed = zfpy.compress_numpy(array, tolerance=tolerance)\n",
    "    return compressed\n",
    "\n",
    "# Function to decompress tensor using ZFP\n",
    "def decompress_tensor(compressed, original_shape):\n",
    "    decompressed = zfpy.decompress_numpy(compressed)\n",
    "    tensor = torch.tensor(decompressed).reshape(original_shape)\n",
    "    return tensor"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e15ab7dc",
   "metadata": {
    "papermill": {
     "duration": 0.009815,
     "end_time": "2024-06-05T10:44:53.530931",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.521116",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Train VGG11"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14cc4405",
   "metadata": {
    "papermill": {
     "duration": 0.009869,
     "end_time": "2024-06-05T10:44:53.550909",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.541040",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "#### Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "04c2fc66",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:44:53.572339Z",
     "iopub.status.busy": "2024-06-05T10:44:53.572016Z",
     "iopub.status.idle": "2024-06-05T10:45:12.068028Z",
     "shell.execute_reply": "2024-06-05T10:45:12.067261Z"
    },
    "papermill": {
     "duration": 18.509346,
     "end_time": "2024-06-05T10:45:12.070298",
     "exception": false,
     "start_time": "2024-06-05T10:44:53.560952",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_23/1903179062.py:8: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', False):\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading https://www.cs.toronto.edu/~kriz/cifar-100-python.tar.gz to ./data/cifar-100-python.tar.gz\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 169001437/169001437 [00:15<00:00, 11134843.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting ./data/cifar-100-python.tar.gz to ./data\n"
     ]
    }
   ],
   "source": [
    "def data_loader(data_dir,\n",
    "                batch_size,\n",
    "                random_seed=42,\n",
    "                valid_size=0.1,\n",
    "                shuffle=True,\n",
    "                test=False):\n",
    "      \n",
    "    with pd.option_context('mode.use_inf_as_na', False):\n",
    "        normalize = transforms.Normalize(\n",
    "            mean=[0.4914, 0.4822, 0.4465],\n",
    "            std=[0.2023, 0.1994, 0.2010],\n",
    "        )\n",
    "\n",
    "        # define transforms\n",
    "        transform = transforms.Compose([\n",
    "                transforms.Resize((227,227)),\n",
    "                transforms.ToTensor(),\n",
    "                normalize,\n",
    "        ])\n",
    "    \n",
    "        if test:\n",
    "            dataset = datasets.CIFAR100(\n",
    "              root=data_dir, train=False,\n",
    "              download=True, transform=transform,\n",
    "            )\n",
    "\n",
    "            data_loader = torch.utils.data.DataLoader(\n",
    "                dataset, batch_size=batch_size, shuffle=shuffle\n",
    "            )\n",
    "\n",
    "            return data_loader\n",
    "\n",
    "        # load the dataset\n",
    "        train_dataset = datasets.CIFAR100(\n",
    "            root=data_dir, train=True,\n",
    "            download=True, transform=transform,\n",
    "        )\n",
    "\n",
    "        valid_dataset = datasets.CIFAR10(\n",
    "            root=data_dir, train=True,\n",
    "            download=True, transform=transform,\n",
    "        )\n",
    "\n",
    "        num_train = len(train_dataset)\n",
    "        indices = list(range(num_train))\n",
    "        split = int(np.floor(valid_size * num_train))\n",
    "\n",
    "        if shuffle:\n",
    "            np.random.seed(random_seed)\n",
    "            np.random.shuffle(indices)\n",
    "\n",
    "        train_idx, valid_idx = indices[split:], indices[:split]\n",
    "        train_sampler = SubsetRandomSampler(train_idx)\n",
    "        valid_sampler = SubsetRandomSampler(valid_idx)\n",
    "\n",
    "        train_loader = torch.utils.data.DataLoader(\n",
    "            train_dataset, batch_size=batch_size, sampler=train_sampler)\n",
    "\n",
    "        valid_loader = torch.utils.data.DataLoader(\n",
    "            valid_dataset, batch_size=batch_size, sampler=valid_sampler)\n",
    "\n",
    "        return (train_loader, valid_loader)\n",
    "\n",
    "\n",
    "# CIFAR100 dataset\n",
    "if TRAIN:\n",
    "    train_loader, valid_loader = data_loader(data_dir='./data', batch_size=64)\n",
    "    \n",
    "test_loader = data_loader(data_dir='./data',\n",
    "                          batch_size=64,\n",
    "                          test=True)\n",
    "\n",
    "def test_accuracy(model):\n",
    "    with torch.no_grad():\n",
    "        correct = 0\n",
    "        total = 0\n",
    "        i = 0\n",
    "        \n",
    "#         return 100*3/5\n",
    "        for images, labels in test_loader:\n",
    "            images = images.to(device)\n",
    "            labels = labels.to(device)\n",
    "            outputs = model(images)\n",
    "            _, predicted = torch.max(outputs.data, 1)\n",
    "            \n",
    "            total += labels.size(0)\n",
    "            correct += (predicted == labels).sum().item()\n",
    "            del images, labels, outputs\n",
    "            \n",
    "            i += 1\n",
    "#             if i == 1:\n",
    "#                 break\n",
    "            \n",
    "        return 100 * correct / total"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2e67f23",
   "metadata": {
    "papermill": {
     "duration": 0.018232,
     "end_time": "2024-06-05T10:45:12.107419",
     "exception": false,
     "start_time": "2024-06-05T10:45:12.089187",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "#### Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1ca70b59",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:45:12.145073Z",
     "iopub.status.busy": "2024-06-05T10:45:12.144788Z",
     "iopub.status.idle": "2024-06-05T10:45:19.116750Z",
     "shell.execute_reply": "2024-06-05T10:45:19.115902Z"
    },
    "papermill": {
     "duration": 6.993371,
     "end_time": "2024-06-05T10:45:19.119074",
     "exception": false,
     "start_time": "2024-06-05T10:45:12.125703",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "if TRAIN:\n",
    "    model = models.vgg11(weights=VGG11_Weights.IMAGENET1K_V1).to(device)  \n",
    "    num_epochs = 5\n",
    "    learning_rate = 0.005\n",
    "\n",
    "    # Loss and optimizer\n",
    "    criterion = nn.CrossEntropyLoss()\n",
    "    optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate, weight_decay = 0.005, momentum = 0.9)\n",
    "    total_step = len(train_loader)\n",
    "\n",
    "    for epoch in range(num_epochs):\n",
    "        model.train()\n",
    "        for i, (images, labels) in enumerate(train_loader):  \n",
    "            # Move tensors to the configured device\n",
    "            images = images.to(device)\n",
    "            labels = labels.to(device)\n",
    "            \n",
    "            # Forward pass\n",
    "            outputs = model(images)\n",
    "            loss = criterion(outputs, labels)\n",
    "            \n",
    "            # Backward and optimize\n",
    "            optimizer.zero_grad()\n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "            \n",
    "            if i % 100 == 0:\n",
    "                print ('Epoch [{}/{}], Step [{}/{}], Loss: {:.4f}' \n",
    "                            .format(epoch+1, num_epochs, i+1, total_step, loss.item()))\n",
    "        \n",
    "        # Validation\n",
    "        model.eval()\n",
    "        with torch.no_grad():\n",
    "            correct = 0\n",
    "            total = 0\n",
    "            for images, labels in valid_loader:\n",
    "                images = images.to(device)\n",
    "                labels = labels.to(device)\n",
    "                outputs = model(images)\n",
    "                _, predicted = torch.max(outputs.data, 1)\n",
    "                total += labels.size(0)\n",
    "                correct += (predicted == labels).sum().item()\n",
    "                del images, labels, outputs\n",
    "        \n",
    "            print('Accuracy of the network on the {} validation images: {} % (Epoch {})'.format(5000, 100 * correct / total, epoch+1))\n",
    "else:\n",
    "#     model = models.vgg11(weights=VGG11_Weights.IMAGENET1K_V1).to(device)  \n",
    "    model = models.vgg11()\n",
    "    model.load_state_dict(load_pkl(MODEL_PATH))\n",
    "    model.eval()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "423049cc",
   "metadata": {
    "papermill": {
     "duration": 0.01852,
     "end_time": "2024-06-05T10:45:19.156644",
     "exception": false,
     "start_time": "2024-06-05T10:45:19.138124",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Distribution of Original Weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "dec0b714",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:45:19.194432Z",
     "iopub.status.busy": "2024-06-05T10:45:19.194088Z",
     "iopub.status.idle": "2024-06-05T10:45:19.198044Z",
     "shell.execute_reply": "2024-06-05T10:45:19.197248Z"
    },
    "papermill": {
     "duration": 0.024978,
     "end_time": "2024-06-05T10:45:19.199854",
     "exception": false,
     "start_time": "2024-06-05T10:45:19.174876",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# classifier.0.weight 102760448\n",
    "# classifier.3.weight 16777216\n",
    "# classifier.6.weight 4096000\n",
    "# plot_weight_distribution(model, base=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "43175e38",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:45:19.237440Z",
     "iopub.status.busy": "2024-06-05T10:45:19.237171Z",
     "iopub.status.idle": "2024-06-05T10:45:19.493630Z",
     "shell.execute_reply": "2024-06-05T10:45:19.492716Z"
    },
    "papermill": {
     "duration": 0.277814,
     "end_time": "2024-06-05T10:45:19.495790",
     "exception": false,
     "start_time": "2024-06-05T10:45:19.217976",
     "status": "completed"
    },
    "tags": []
   },
   "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>Layer</th>\n",
       "      <th>Mean</th>\n",
       "      <th>Standard Deviation</th>\n",
       "      <th>Min</th>\n",
       "      <th>Max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>features.0.weight</td>\n",
       "      <td>-6.118355e-04</td>\n",
       "      <td>0.182518</td>\n",
       "      <td>-0.576514</td>\n",
       "      <td>0.733067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>features.3.weight</td>\n",
       "      <td>-7.280580e-04</td>\n",
       "      <td>0.031691</td>\n",
       "      <td>-0.366674</td>\n",
       "      <td>0.426103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>features.6.weight</td>\n",
       "      <td>-5.120513e-04</td>\n",
       "      <td>0.017840</td>\n",
       "      <td>-0.239951</td>\n",
       "      <td>0.290745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>features.8.weight</td>\n",
       "      <td>-9.444316e-04</td>\n",
       "      <td>0.014110</td>\n",
       "      <td>-0.204781</td>\n",
       "      <td>0.409000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>features.11.weight</td>\n",
       "      <td>-5.817962e-04</td>\n",
       "      <td>0.011325</td>\n",
       "      <td>-0.171319</td>\n",
       "      <td>0.202021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>features.13.weight</td>\n",
       "      <td>-1.003245e-03</td>\n",
       "      <td>0.008708</td>\n",
       "      <td>-0.144776</td>\n",
       "      <td>0.156960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>features.16.weight</td>\n",
       "      <td>-5.544542e-04</td>\n",
       "      <td>0.009031</td>\n",
       "      <td>-0.161565</td>\n",
       "      <td>0.128053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>features.18.weight</td>\n",
       "      <td>-1.142678e-03</td>\n",
       "      <td>0.008698</td>\n",
       "      <td>-0.103508</td>\n",
       "      <td>0.119273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>classifier.0.weight</td>\n",
       "      <td>-1.109735e-04</td>\n",
       "      <td>0.002568</td>\n",
       "      <td>-0.024549</td>\n",
       "      <td>0.042916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>classifier.3.weight</td>\n",
       "      <td>-2.723727e-04</td>\n",
       "      <td>0.004957</td>\n",
       "      <td>-0.030718</td>\n",
       "      <td>0.038001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>classifier.6.weight</td>\n",
       "      <td>4.801769e-07</td>\n",
       "      <td>0.008999</td>\n",
       "      <td>-0.059792</td>\n",
       "      <td>0.142456</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Layer          Mean  Standard Deviation       Min       Max\n",
       "0     features.0.weight -6.118355e-04            0.182518 -0.576514  0.733067\n",
       "1     features.3.weight -7.280580e-04            0.031691 -0.366674  0.426103\n",
       "2     features.6.weight -5.120513e-04            0.017840 -0.239951  0.290745\n",
       "3     features.8.weight -9.444316e-04            0.014110 -0.204781  0.409000\n",
       "4    features.11.weight -5.817962e-04            0.011325 -0.171319  0.202021\n",
       "5    features.13.weight -1.003245e-03            0.008708 -0.144776  0.156960\n",
       "6    features.16.weight -5.544542e-04            0.009031 -0.161565  0.128053\n",
       "7    features.18.weight -1.142678e-03            0.008698 -0.103508  0.119273\n",
       "8   classifier.0.weight -1.109735e-04            0.002568 -0.024549  0.042916\n",
       "9   classifier.3.weight -2.723727e-04            0.004957 -0.030718  0.038001\n",
       "10  classifier.6.weight  4.801769e-07            0.008999 -0.059792  0.142456"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "orig_stats = print_statistics(model)\n",
    "orig_stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7996a187",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:45:19.536022Z",
     "iopub.status.busy": "2024-06-05T10:45:19.535708Z",
     "iopub.status.idle": "2024-06-05T10:45:49.525755Z",
     "shell.execute_reply": "2024-06-05T10:45:49.524507Z"
    },
    "papermill": {
     "duration": 30.012562,
     "end_time": "2024-06-05T10:45:49.527926",
     "exception": false,
     "start_time": "2024-06-05T10:45:19.515364",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model Size: 506.84027194976807 MB\n",
      "Test Accuracy: 59.51\n"
     ]
    }
   ],
   "source": [
    "model = model.to(device)\n",
    "\n",
    "weights = {}\n",
    "for name, param in model.named_parameters():\n",
    "    if param.requires_grad:\n",
    "        weights[name] = param.data            \n",
    "org_size = save_pkl('org_model_param_size.pkl', weights)\n",
    "\n",
    "if TRAIN:\n",
    "    save_pkl('org_model.pkl', model.state_dict())\n",
    "\n",
    "org_data = {\n",
    "    'tolerance': 0,\n",
    "    'test_accuracy': test_accuracy(model),\n",
    "    'compression_ratio': 1,\n",
    "    'model_size': org_size / 1024**2,  # Size in MB\n",
    "    'stats': orig_stats\n",
    "}\n",
    "\n",
    "print('Model Size:', org_data['model_size'], 'MB')\n",
    "print(f'Test Accuracy: {org_data[\"test_accuracy\"]}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "414d357f",
   "metadata": {
    "papermill": {
     "duration": 0.018484,
     "end_time": "2024-06-05T10:45:49.565969",
     "exception": false,
     "start_time": "2024-06-05T10:45:49.547485",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Pruning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "68961535",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:45:49.604223Z",
     "iopub.status.busy": "2024-06-05T10:45:49.603938Z",
     "iopub.status.idle": "2024-06-05T10:45:49.607860Z",
     "shell.execute_reply": "2024-06-05T10:45:49.607018Z"
    },
    "papermill": {
     "duration": 0.025184,
     "end_time": "2024-06-05T10:45:49.609665",
     "exception": false,
     "start_time": "2024-06-05T10:45:49.584481",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# parameters_to_prune = (\n",
    "#     (model.features[0], 'weight'),\n",
    "#     (model.features[3], 'weight'),\n",
    "#     (model.features[6], 'weight'),\n",
    "#     (model.features[8], 'weight'),\n",
    "#     (model.features[11], 'weight'),\n",
    "#     (model.features[13], 'weight'),\n",
    "#     (model.features[16], 'weight'),\n",
    "#     (model.features[18], 'weight')\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "0692874a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T10:45:49.648141Z",
     "iopub.status.busy": "2024-06-05T10:45:49.647897Z",
     "iopub.status.idle": "2024-06-05T11:07:25.554830Z",
     "shell.execute_reply": "2024-06-05T11:07:25.553941Z"
    },
    "papermill": {
     "duration": 1295.943179,
     "end_time": "2024-06-05T11:07:25.571504",
     "exception": false,
     "start_time": "2024-06-05T10:45:49.628325",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing: 100%|██████████| 20/20 [21:35<00:00, 64.79s/item]\n"
     ]
    }
   ],
   "source": [
    "def calculate_sparsity(model):\n",
    "    total_params = 0\n",
    "    zero_params = 0\n",
    "\n",
    "    for param in model.parameters():\n",
    "        total_params += param.numel()\n",
    "        zero_params += torch.sum(param == 0).item()\n",
    "\n",
    "    sparsity = zero_params / total_params\n",
    "    return sparsity\n",
    "\n",
    "def global_prune(c_model, amount):\n",
    "    parameters_to_prune = []\n",
    "    \n",
    "    for name, module in c_model.named_modules():\n",
    "        if isinstance(module, (torch.nn.Conv2d, torch.nn.Linear)):\n",
    "            parameters_to_prune.append((module, 'weight'))\n",
    "            if module.bias is not None:\n",
    "                parameters_to_prune.append((module, 'bias'))\n",
    "    \n",
    "    prune.global_unstructured(\n",
    "        parameters_to_prune,\n",
    "        pruning_method=prune.L1Unstructured,\n",
    "        amount=amount,\n",
    "    )\n",
    "\n",
    "    for layer, param in parameters_to_prune:\n",
    "        prune.remove(layer, param)\n",
    "        \n",
    "    \n",
    "prune_values = [i/20 for i in range(0, 20)]\n",
    "TOLERANCE = 0.0002  \n",
    "\n",
    "if PRUNE and (TRAIN or not PRUNE_LOAD):\n",
    "    prune_results = []\n",
    "    for i, amount in tqdm(enumerate(prune_values), total=len(prune_values), desc=\"Processing\", unit=\"item\"):\n",
    "        copied_model = copy.deepcopy(model)\n",
    "        global_prune(copied_model, amount)\n",
    "        before_sparisty = calculate_sparsity(copied_model)\n",
    "        old_stats = print_statistics(copied_model)\n",
    "        old_accuracy = test_accuracy(copied_model.to(device))\n",
    "        \n",
    "#         if i >= 17:\n",
    "#             plot_classifier_distribution(copied_model, tolerance=TOLERANCE, prune=amount, prune_zfp=False, bins=100)\n",
    "\n",
    "        # Compress weights\n",
    "        compressed_weights = {}\n",
    "        for name, param in copied_model.named_parameters():\n",
    "            if param.requires_grad:\n",
    "                compressed_weights[name] = compress_tensor(param.data, tolerance=TOLERANCE)\n",
    "\n",
    "        pkl_path = f'zfp_model_prune_{amount}.pkl'\n",
    "        zfp_size = save_pkl(pkl_path, compressed_weights)\n",
    "\n",
    "        # Decompress weights\n",
    "        for name, param in copied_model.named_parameters():\n",
    "            if param.requires_grad:\n",
    "                param.data = decompress_tensor(compressed_weights[name], param.data.shape)\n",
    "\n",
    "#         if i >= 17:\n",
    "#             plot_classifier_distribution(copied_model, tolerance=TOLERANCE, prune=amount, prune_zfp=True, bins=100)\n",
    "        \n",
    "        prune_results.append({\n",
    "            'before_sparsity': before_sparisty,\n",
    "            'after_sparsity': calculate_sparsity(copied_model),\n",
    "            'compression_ratio': org_size / zfp_size,\n",
    "            'model_size': zfp_size / 1024**2,  # Size in MB\n",
    "            'old_stats': old_stats,\n",
    "            'new_stats': print_statistics(copied_model),\n",
    "            'before_accuracy': old_accuracy,\n",
    "            'after_accuracy': test_accuracy(copied_model.to(device))\n",
    "#             .to(device)\n",
    "        })\n",
    "    \n",
    "    # save prune_results\n",
    "    prune_results_copy = copy.deepcopy(prune_results)\n",
    "    for r in prune_results_copy:\n",
    "        r['old_stats'] = r['old_stats'].to_json(orient='records')\n",
    "        r['new_stats'] = r['new_stats'].to_json(orient='records')\n",
    "    \n",
    "    with open(PRUNE_DATA_JSON, 'w') as json_file:\n",
    "        json.dump(prune_results_copy, json_file, indent=4) \n",
    "        \n",
    "elif PRUNE:\n",
    "    # load prune_results\n",
    "    with open(PRUNE_DATA_JSON, 'r') as json_file:\n",
    "        prune_results = json.load(json_file)\n",
    "        \n",
    "        for r in prune_results:\n",
    "            r['old_stats'] = pd.read_json(StringIO(r['old_stats']), orient='records')\n",
    "            r['new_stats'] = pd.read_json(StringIO(r['new_stats']), orient='records')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f4a8729b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T11:07:25.613260Z",
     "iopub.status.busy": "2024-06-05T11:07:25.612969Z",
     "iopub.status.idle": "2024-06-05T11:07:25.616602Z",
     "shell.execute_reply": "2024-06-05T11:07:25.615824Z"
    },
    "papermill": {
     "duration": 0.026537,
     "end_time": "2024-06-05T11:07:25.618456",
     "exception": false,
     "start_time": "2024-06-05T11:07:25.591919",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# plot_weight_distribution(copied_model, tolerance=TOLERANCE, prune=0.95, prune_zfp=False, bins=7)\n",
    "# print(org_size)\n",
    "# prune_results[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "2de93a0b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T11:07:25.659681Z",
     "iopub.status.busy": "2024-06-05T11:07:25.659381Z",
     "iopub.status.idle": "2024-06-05T11:07:26.291802Z",
     "shell.execute_reply": "2024-06-05T11:07:26.290900Z"
    },
    "papermill": {
     "duration": 0.655439,
     "end_time": "2024-06-05T11:07:26.293985",
     "exception": false,
     "start_time": "2024-06-05T11:07:25.638546",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if PRUNE:\n",
    "    compression_ratios = [result[\"compression_ratio\"] for result in prune_results]\n",
    "    model_sizes = [result[\"model_size\"] for result in prune_results]\n",
    "    \n",
    "\n",
    "    data = [\n",
    "        (compression_ratios, org_data['compression_ratio'], 'green', 'Compression Ratio'),\n",
    "        (model_sizes, org_data['model_size'], 'red', 'Model Size (MB)'),\n",
    "    ]\n",
    "\n",
    "    for values, baseline, color, label in data:\n",
    "        plt.figure(figsize=(10, 6))\n",
    "\n",
    "        plt.plot(prune_values, values, color=color, marker='o', label=label)\n",
    "        plt.axhline(y=baseline, color='purple', linestyle='--', label=f'Baseline {label}')\n",
    "\n",
    "        plt.title(f'{label} vs Prune Amount with Baseline with Tolerance {TOLERANCE}')\n",
    "#         plt.xscale('log', base=2)\n",
    "        plt.xlabel('Prune Amount')\n",
    "        plt.ylabel(label)\n",
    "        plt.legend()\n",
    "        plt.grid(True)\n",
    "\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8542c9c8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T11:07:26.338897Z",
     "iopub.status.busy": "2024-06-05T11:07:26.338496Z",
     "iopub.status.idle": "2024-06-05T11:07:26.658308Z",
     "shell.execute_reply": "2024-06-05T11:07:26.657407Z"
    },
    "papermill": {
     "duration": 0.344332,
     "end_time": "2024-06-05T11:07:26.660249",
     "exception": false,
     "start_time": "2024-06-05T11:07:26.315917",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if PRUNE:\n",
    "    before_accuracies = [result[\"before_accuracy\"] for result in prune_results]\n",
    "    after_accuracies = [result[\"after_accuracy\"] for result in prune_results]\n",
    "    baseline = org_data['test_accuracy']\n",
    "\n",
    "    data = [\n",
    "        (before_accuracies, 'green', 'Pruned Accuracy'),\n",
    "        (after_accuracies,'red', 'ZFP Pruned Accuracy'),\n",
    "    ]\n",
    "    \n",
    "    plt.figure(figsize=(10, 6))\n",
    "    plt.axhline(y=baseline, color='purple', linestyle='--', label=f'Baseline Accuracy')\n",
    "\n",
    "    for values, color, label in data:\n",
    "        plt.plot(prune_values, values, color=color, marker='o', label=label)\n",
    "\n",
    "    plt.title(f'ZFP vs Prune Accuracy with Baseline and Tolerance {TOLERANCE}')\n",
    "    plt.xlabel('Prune Amount')\n",
    "    plt.ylabel(label)\n",
    "    plt.legend()\n",
    "    plt.grid(True)\n",
    "\n",
    "    plt.show()\n",
    "    \n",
    "#     print(baseline, before_accuracies, after_accuracies)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "70c2b820",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T11:07:26.707037Z",
     "iopub.status.busy": "2024-06-05T11:07:26.706737Z",
     "iopub.status.idle": "2024-06-05T11:07:27.083678Z",
     "shell.execute_reply": "2024-06-05T11:07:27.082854Z"
    },
    "papermill": {
     "duration": 0.402482,
     "end_time": "2024-06-05T11:07:27.086056",
     "exception": false,
     "start_time": "2024-06-05T11:07:26.683574",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if PRUNE:\n",
    "    avg_mean = []\n",
    "    avg_sd = []\n",
    "    avg_min = []\n",
    "    avg_max = []\n",
    "#     sparsity_dif = []\n",
    "\n",
    "    for model_data in prune_results:\n",
    "        comp = compare_model_stats(model_data['old_stats'], model_data['new_stats'])\n",
    "\n",
    "        avg_mean.append(comp['MSE Mean'].sum())\n",
    "        avg_sd.append(comp['MSE SD'].sum())  \n",
    "        avg_min.append(comp['MSE Min'].sum())\n",
    "        avg_max.append(comp['MSE Max'].sum())\n",
    "#         sparsity_dif.append( (model_data['after_sparsity'] - model_data['before_sparsity'])**2 )\n",
    "    \n",
    "    data = [\n",
    "        (avg_mean, 'blue', 'Mean'),\n",
    "        (avg_sd, 'green', 'Standard Deviation'),\n",
    "        (avg_min, 'red', 'Min'),\n",
    "        (avg_max, 'purple', 'Max'),\n",
    "#         (sparsity_dif, 'orange', 'Global Sparsity')\n",
    "    ]\n",
    "\n",
    "    plt.figure(figsize=(10, 6))\n",
    "    plt.title(f'Prune Amount vs Model Weight MSE for Tolerance {TOLERANCE}')\n",
    "    plt.xlabel('Prune Amount')\n",
    "    plt.ylabel('Value')\n",
    "\n",
    "    for values, color, label in data:\n",
    "        plt.plot(prune_values, values, color=color, marker='o', label=label)\n",
    "\n",
    "    plt.legend()\n",
    "    plt.grid(True)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "50c55da0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T11:07:27.136255Z",
     "iopub.status.busy": "2024-06-05T11:07:27.135947Z",
     "iopub.status.idle": "2024-06-05T11:07:27.428954Z",
     "shell.execute_reply": "2024-06-05T11:07:27.428120Z"
    },
    "papermill": {
     "duration": 0.32107,
     "end_time": "2024-06-05T11:07:27.431214",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.110144",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if PRUNE:    \n",
    "    plt.figure(figsize=(10, 6))\n",
    "    plt.title(f'Sparsity before and after ZFP for Tolerance {TOLERANCE}')\n",
    "    plt.xlabel('Prune Amount')\n",
    "    plt.ylabel('Sparsity')\n",
    "    \n",
    "    data = [\n",
    "        ([r['before_sparsity'] for r in prune_results], 'blue', 'Before'),\n",
    "        ([r['after_sparsity'] for r in prune_results], 'red', 'After'),\n",
    "    ]\n",
    "    for values, color, label in data:\n",
    "        plt.plot(prune_values, values, color=color, marker='o', label=label)\n",
    "\n",
    "    plt.legend()\n",
    "    plt.grid(True)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b6887e70",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T11:07:27.481050Z",
     "iopub.status.busy": "2024-06-05T11:07:27.480350Z",
     "iopub.status.idle": "2024-06-05T11:07:27.484207Z",
     "shell.execute_reply": "2024-06-05T11:07:27.483368Z"
    },
    "papermill": {
     "duration": 0.0307,
     "end_time": "2024-06-05T11:07:27.486079",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.455379",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# print(prune_results[-4])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c717c04e",
   "metadata": {
    "papermill": {
     "duration": 0.023756,
     "end_time": "2024-06-05T11:07:27.533805",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.510049",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Perform Test Accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "76d8280c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T11:07:27.582647Z",
     "iopub.status.busy": "2024-06-05T11:07:27.582349Z",
     "iopub.status.idle": "2024-06-05T11:07:27.590879Z",
     "shell.execute_reply": "2024-06-05T11:07:27.590034Z"
    },
    "papermill": {
     "duration": 0.034979,
     "end_time": "2024-06-05T11:07:27.592659",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.557680",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "tolerance_values = [1/2**i for i in range(0, 27)]\n",
    "# tolerance_values = [1/2**i for i in range(0, 2)]\n",
    "\n",
    "if not PRUNE and (TRAIN or not TRAIN_LOAD):\n",
    "    results = []\n",
    "\n",
    "    for i, tolerance in tqdm(enumerate(tolerance_values), total=len(tolerance_values), desc=\"Processing\", unit=\"item\"):\n",
    "        copied_model = copy.deepcopy(model)\n",
    "\n",
    "        # Compress weights\n",
    "        compressed_weights = {}\n",
    "        for name, param in copied_model.named_parameters():\n",
    "            if param.requires_grad:\n",
    "                compressed_weights[name] = compress_tensor(param.data, tolerance=tolerance)\n",
    "\n",
    "        pkl_path = f'zfp_model_{tolerance}.pkl'\n",
    "        zfp_size = save_pkl(pkl_path, compressed_weights)\n",
    "\n",
    "        # Decompress weights\n",
    "        for name, param in copied_model.named_parameters():\n",
    "            if param.requires_grad:\n",
    "                param.data = decompress_tensor(compressed_weights[name], param.data.shape)\n",
    "        \n",
    "        results.append({\n",
    "            'tolerance': tolerance,\n",
    "            'test_accuracy': test_accuracy(copied_model.to(device)),\n",
    "#             'test_accuracy': 0.25,\n",
    "            'compression_ratio': org_size / zfp_size,\n",
    "            'model_size': zfp_size / 1024**2,  # Size in MB\n",
    "            'stats': print_statistics(copied_model)\n",
    "        })\n",
    "\n",
    "#         if i % 9 == 0:\n",
    "#             plot_weight_distribution(copied_model, tolerance=tolerance)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "226bc4c2",
   "metadata": {
    "papermill": {
     "duration": 0.023513,
     "end_time": "2024-06-05T11:07:27.639732",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.616219",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Save or Load Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "e01c8deb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T11:07:27.688090Z",
     "iopub.status.busy": "2024-06-05T11:07:27.687837Z",
     "iopub.status.idle": "2024-06-05T11:07:27.694441Z",
     "shell.execute_reply": "2024-06-05T11:07:27.693626Z"
    },
    "papermill": {
     "duration": 0.033026,
     "end_time": "2024-06-05T11:07:27.696365",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.663339",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "if not PRUNE:\n",
    "    if TRAIN or not TRAIN_LOAD:\n",
    "        # save results\n",
    "        results_copy = copy.deepcopy(results)\n",
    "        org_data_copy = copy.deepcopy(org_data)\n",
    "        results_copy.append(org_data_copy)\n",
    "\n",
    "        for r in results_copy:\n",
    "            r['stats'] = r['stats'].to_json(orient='records')\n",
    "\n",
    "        with open(TEST_JSON, 'w') as json_file:\n",
    "            json.dump(results_copy, json_file, indent=4)  \n",
    "    else:\n",
    "        # load results\n",
    "        with open(TEST_JSON, 'r') as json_file:\n",
    "            results = json.load(json_file)\n",
    "\n",
    "            for r in results:\n",
    "                r['stats'] = pd.read_json(StringIO(r['stats']), orient='records')\n",
    "\n",
    "            org_data = results.pop()\n",
    "\n",
    "# print(results[0].keys())\n",
    "# print(type(org_data['stats']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ed98c6a",
   "metadata": {
    "papermill": {
     "duration": 0.023654,
     "end_time": "2024-06-05T11:07:27.744914",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.721260",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Graphing Tolerance vs Accuracy, Compression Ratio, Model Size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "82c0d90f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T11:07:27.793408Z",
     "iopub.status.busy": "2024-06-05T11:07:27.793154Z",
     "iopub.status.idle": "2024-06-05T11:07:27.801286Z",
     "shell.execute_reply": "2024-06-05T11:07:27.800625Z"
    },
    "papermill": {
     "duration": 0.034523,
     "end_time": "2024-06-05T11:07:27.803070",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.768547",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "if not PRUNE:\n",
    "    tolerances = [result[\"tolerance\"] for result in results]\n",
    "    accuracy = [result[\"test_accuracy\"] for result in results]\n",
    "    compression_ratios = [result[\"compression_ratio\"] for result in results]\n",
    "    model_sizes = [result[\"model_size\"] for result in results]\n",
    "\n",
    "    data = [\n",
    "        (accuracy, org_data['test_accuracy'], 'blue', 'Test Accuracy'),\n",
    "        (compression_ratios, org_data['compression_ratio'], 'green', 'Compression Ratio'),\n",
    "        (model_sizes, org_data['model_size'], 'red', 'Model Size (MB)'),\n",
    "    ]\n",
    "\n",
    "    for values, baseline, color, label in data:\n",
    "        plt.figure(figsize=(10, 6))\n",
    "\n",
    "        plt.plot(tolerances, values, color=color, marker='o', label=label)\n",
    "        plt.axhline(y=baseline, color='purple', linestyle='--', label=f'Baseline {label}')\n",
    "\n",
    "        plt.title(f'{label} vs Tolerance (Log Scale) with Baseline')\n",
    "        plt.xscale('log', base=2)\n",
    "        plt.xlabel('Tolerance (Log Scale)')\n",
    "        plt.ylabel(label)\n",
    "        plt.legend()\n",
    "        plt.grid(True)\n",
    "\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d1aa8f3a",
   "metadata": {
    "papermill": {
     "duration": 0.023783,
     "end_time": "2024-06-05T11:07:27.850777",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.826994",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Comparing Mean, Standard Deviation, Min, Max"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "c6bd4caa",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-06-05T11:07:27.900508Z",
     "iopub.status.busy": "2024-06-05T11:07:27.900224Z",
     "iopub.status.idle": "2024-06-05T11:07:27.908662Z",
     "shell.execute_reply": "2024-06-05T11:07:27.907993Z"
    },
    "papermill": {
     "duration": 0.035579,
     "end_time": "2024-06-05T11:07:27.910478",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.874899",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "if not PRUNE:\n",
    "\n",
    "    abs_mean = []\n",
    "    abs_sd = []\n",
    "    abs_min = []\n",
    "    abs_max = []\n",
    "    orig_stats = org_data['stats']\n",
    "\n",
    "    for tolerance, model_data in zip(tolerances, results):\n",
    "        comp = compare_model_stats(orig_stats, model_data['stats'])\n",
    "\n",
    "        abs_mean.append(comp['MSE Mean'].sum())\n",
    "        abs_sd.append(comp['MSE SD'].sum())  \n",
    "        abs_min.append(comp['MSE Min'].sum())\n",
    "        abs_max.append(comp['MSE Max'].sum())\n",
    "    \n",
    "    \n",
    "    # plotting\n",
    "    data = [\n",
    "        (abs_mean, 'blue', 'Mean'),\n",
    "        (abs_sd, 'green', 'Standard Deviation'),\n",
    "        (abs_min, 'red', 'Min'),\n",
    "        (abs_max, 'purple', 'Max')\n",
    "    ]\n",
    "\n",
    "    plt.figure(figsize=(10, 6))\n",
    "    plt.title(f'Tolerance vs Total Layer MSE for Mean, Standard Deviation, Min, and Max')\n",
    "    plt.xscale('log', base=2)\n",
    "    plt.xlabel('Tolerance (Log Scale)')\n",
    "    plt.ylabel('Total MSE')\n",
    "\n",
    "    for values, color, label in data:\n",
    "        plt.plot(tolerances, values, color=color, marker='o', label=label)\n",
    "\n",
    "\n",
    "    plt.legend()\n",
    "    plt.grid(True)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02a38f18",
   "metadata": {
    "papermill": {
     "duration": 0.023941,
     "end_time": "2024-06-05T11:07:27.958321",
     "exception": false,
     "start_time": "2024-06-05T11:07:27.934380",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Resources - Works Cited\n",
    "- https://blog.paperspace.com/vgg-from-scratch-pytorch/\n",
    "- https://pypi.org/project/zfpy/\n",
    "- https://pytorch.org/hub/pytorch_vision_vgg/\n",
    "- https://github.com/alex-grindrod/eblc-model-compression\n",
    "- https://github.com/LLNL/zfp"
   ]
  }
 ],
 "metadata": {
  "kaggle": {
   "accelerator": "gpu",
   "dataSources": [
    {
     "sourceId": 181577038,
     "sourceType": "kernelVersion"
    }
   ],
   "dockerImageVersionId": 30716,
   "isGpuEnabled": true,
   "isInternetEnabled": true,
   "language": "python",
   "sourceType": "notebook"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 1379.769214,
   "end_time": "2024-06-05T11:07:30.135927",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2024-06-05T10:44:30.366713",
   "version": "2.5.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}