{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pytorch Image Models (timm)\n",
    "\n",
    "> `timm` is a deep-learning library created by [Ross Wightman](https://twitter.com/wightmanr) and is a collection of SOTA computer vision models, layers, utilities, optimizers, schedulers, data-loaders, augmentations and also training/validating scripts with ability to reproduce ImageNet training results. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Install"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```\n",
    "pip install timm\n",
    "```\n",
    "\n",
    "Or for an editable install, \n",
    "\n",
    "```\n",
    "git clone https://github.com/rwightman/pytorch-image-models\n",
    "cd pytorch-image-models && pip install -e .\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## How to use"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import timm \n",
    "import torch\n",
    "\n",
    "model = timm.create_model('resnet34')\n",
    "x     = torch.randn(1, 3, 224, 224)\n",
    "model(x).shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is that simple to create a model using `timm`. The `create_model` function is a factory method that can be used to create over 300 models that are part of the `timm` library."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To create a pretrained model, simply pass in `pretrained=True`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pth\" to /Users/amanarora/.cache/torch/hub/checkpoints/resnet34-43635321.pth\n"
     ]
    }
   ],
   "source": [
    "pretrained_resnet_34 = timm.create_model('resnet34', pretrained=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To create a model with a custom number of classes, simply pass in `num_classes=<number_of_classes>`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 10])"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import timm \n",
    "import torch\n",
    "\n",
    "model = timm.create_model('resnet34', num_classes=10)\n",
    "x     = torch.randn(1, 3, 224, 224)\n",
    "model(x).shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### List Models with Pretrained Weights\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`timm.list_models()` returns a complete list of available models in `timm`. To have a look at a complete list of pretrained models, pass in `pretrained=True` in `list_models`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(271,\n",
       " ['adv_inception_v3',\n",
       "  'cspdarknet53',\n",
       "  'cspresnet50',\n",
       "  'cspresnext50',\n",
       "  'densenet121'])"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "avail_pretrained_models = timm.list_models(pretrained=True)\n",
    "len(avail_pretrained_models), avail_pretrained_models[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are a total of **271** models with pretrained weights currently available in `timm`!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Search for model architectures by Wildcard"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is also possible to search for model architectures using Wildcard as below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['densenet121',\n",
       " 'densenet121d',\n",
       " 'densenet161',\n",
       " 'densenet169',\n",
       " 'densenet201',\n",
       " 'densenet264',\n",
       " 'densenet264d_iabn',\n",
       " 'densenetblur121d',\n",
       " 'tv_densenet121']"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_densenet_models = timm.list_models('*densenet*')\n",
    "all_densenet_models"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}