pytorch-image-models/blseresnext-test.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from fastai.vision import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import timm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#setup paths\n",
    "path = Path(untar_data(URLs.CIFAR_100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#hyperparameters\n",
    "learning_rate = 0.0005133\n",
    "dropout = 0.4068\n",
    "bs = 32\n",
    "epochs=10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "new() received an invalid combination of arguments - got (bool, int), but expected one of:\n * (torch.device device)\n * (torch.Storage storage)\n * (Tensor other)\n * (tuple of ints size, torch.device device)\n      didn't match because some of the arguments have invalid types: (\u001b[31;1mbool\u001b[0m, \u001b[31;1mint\u001b[0m)\n * (object data, torch.device device)\n      didn't match because some of the arguments have invalid types: (\u001b[31;1mbool\u001b[0m, \u001b[31;1mint\u001b[0m)\n",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-5-7a35ae156aa7>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[1;31m#timm.models.blresnet_model(50,2,4)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[1;31m#timm.models.blresnext_model(50,4,32,2,4,pretrained = True)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 5\u001b[1;33m \u001b[0mtimm\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmodels\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mblseresnext_model\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m50\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m4\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m32\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m4\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;32mG:\\Notebooks\\my-timm\\timm\\models\\blseresnext.py\u001b[0m in \u001b[0;36mblseresnext_model\u001b[1;34m(depth, basewidth, cardinality, alpha, beta, use_se, num_classes, pretrained)\u001b[0m\n\u001b[0;32m    299\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    300\u001b[0m     model = bLSEResNeXt(Bottleneck, layers, basewidth, cardinality,\n\u001b[1;32m--> 301\u001b[1;33m                         alpha, beta, num_classes,use_se)\n\u001b[0m\u001b[0;32m    302\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mpretrained\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    303\u001b[0m         url = model_urls['blseresnext-{}-{}x{}d-a{}-b{}'.format(depth, cardinality,\n",
      "\u001b[1;32mG:\\Notebooks\\my-timm\\timm\\models\\blseresnext.py\u001b[0m in \u001b[0;36m__init__\u001b[1;34m(self, block, layers, basewidth, cardinality, alpha, beta, use_se, num_classes)\u001b[0m\n\u001b[0;32m    199\u001b[0m                                        num_channels[3] * block.expansion, layers[3], basewidth, cardinality, stride=2)\n\u001b[0;32m    200\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mgappool\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mnn\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mAdaptiveAvgPool2d\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 201\u001b[1;33m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfc\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mnn\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mLinear\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mnum_channels\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m3\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m*\u001b[0m \u001b[0mblock\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mexpansion\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mnum_classes\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    202\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    203\u001b[0m         \u001b[1;32mfor\u001b[0m \u001b[0mm\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmodules\u001b[0m\u001b[1;33m(\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[1;32m~\\Anaconda3\\envs\\fastai_v1\\lib\\site-packages\\torch\\nn\\modules\\linear.py\u001b[0m in \u001b[0;36m__init__\u001b[1;34m(self, in_features, out_features, bias)\u001b[0m\n\u001b[0;32m     70\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0min_features\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0min_features\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     71\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mout_features\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mout_features\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 72\u001b[1;33m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mweight\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mParameter\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mTensor\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mout_features\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0min_features\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[0;32m     73\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mbias\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     74\u001b[0m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbias\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mParameter\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mTensor\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mout_features\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[1;31mTypeError\u001b[0m: new() received an invalid combination of arguments - got (bool, int), but expected one of:\n * (torch.device device)\n * (torch.Storage storage)\n * (Tensor other)\n * (tuple of ints size, torch.device device)\n      didn't match because some of the arguments have invalid types: (\u001b[31;1mbool\u001b[0m, \u001b[31;1mint\u001b[0m)\n * (object data, torch.device device)\n      didn't match because some of the arguments have invalid types: (\u001b[31;1mbool\u001b[0m, \u001b[31;1mint\u001b[0m)\n"
     ]
    }
   ],
   "source": [
    "#models\n",
    "#model = models.resnet50\n",
    "#timm.models.blresnet_model(50,2,4)\n",
    "#timm.models.blresnext_model(50,4,32,2,4,pretrained = True)\n",
    "timm.models.blseresnext_model(50,4,32,2,4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tfms = get_transforms(do_flip=True, flip_vert=True, max_rotate=25, max_lighting=0.1, max_zoom=1.05, max_warp=0.9)\n",
    "\n",
    "data = (ImageList.from_folder(path)\n",
    ".split_by_rand_pct()\n",
    ".label_from_folder()\n",
    ".transform(get_transforms())\n",
    ".databunch()\n",
    ".normalize(imagenet_stats))\n",
    "\n",
    "learn = cnn_learner(data,\n",
    "model,\n",
    "pretrained=False,\n",
    "ps=dropout,\n",
    "metrics=accuracy).to_fp16()\n",
    "\n",
    "learn.fit_one_cycle(epochs, max_lr=learning_rate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn.save(‘inity1’)\n",
    "learn.export(‘init1y.pkl’)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:fastai_v1] *",
   "language": "python",
   "name": "conda-env-fastai_v1-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}