{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "585e7c01",
   "metadata": {},
   "outputs": [],
   "source": [
    "#|default_exp app"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7263c290",
   "metadata": {},
   "outputs": [],
   "source": [
    "#export\n",
    "from fastai.vision.all import *\n",
    "import gradio as gr\n",
    "import timm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "58dcbf13",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "PILImage mode=RGB size=224x149"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "im = PILImage.create('basset.jpg')\n",
    "im.thumbnail((224,224))\n",
    "im"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "814eaafb",
   "metadata": {},
   "outputs": [],
   "source": [
    "#export\n",
    "import pathlib\n",
    "\n",
    "temp = pathlib.PosixPath\n",
    "pathlib.PosixPath = pathlib.WindowsPath\n",
    "learn = load_learner('model.pkl')\n",
    "pathlib.PosixPath = temp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "86b8b666",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "('basset_hound',\n",
       " tensor(14),\n",
       " tensor([1.1663e-06, 8.5127e-05, 3.8649e-06, 2.6515e-06, 1.0617e-06, 7.1648e-05,\n",
       "         1.4277e-05, 8.6784e-06, 2.9385e-06, 6.6669e-07, 2.3692e-06, 3.0181e-06,\n",
       "         1.3157e-04, 7.9818e-06, 9.9378e-01, 4.5521e-03, 3.5454e-06, 6.5975e-08,\n",
       "         2.8701e-05, 3.6709e-06, 3.8019e-05, 2.3584e-06, 8.8031e-05, 3.2580e-05,\n",
       "         8.1224e-06, 5.6954e-07, 4.1753e-07, 3.3539e-05, 1.1667e-06, 3.4546e-06,\n",
       "         1.0561e-03, 4.2803e-07, 2.1348e-05, 4.9964e-07, 6.4069e-07, 4.4402e-06,\n",
       "         1.7455e-06]))"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "learn.predict(im)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "15c66394",
   "metadata": {},
   "outputs": [],
   "source": [
    "#export\n",
    "categories = learn.dls.vocab\n",
    "\n",
    "def classify_image(img):\n",
    "    pred,idx,probs = learn.predict(img)\n",
    "    return dict(zip(categories, map(float,probs)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "045b0fe5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "{'Abyssinian': 1.1662716588034527e-06,\n",
       " 'Bengal': 8.512729255016893e-05,\n",
       " 'Birman': 3.864904101646971e-06,\n",
       " 'Bombay': 2.6515112949709874e-06,\n",
       " 'British_Shorthair': 1.0616917052175268e-06,\n",
       " 'Egyptian_Mau': 7.164794078562409e-05,\n",
       " 'Maine_Coon': 1.4276593901740853e-05,\n",
       " 'Persian': 8.67839753482258e-06,\n",
       " 'Ragdoll': 2.938470061053522e-06,\n",
       " 'Russian_Blue': 6.666861054327455e-07,\n",
       " 'Siamese': 2.3692359718552325e-06,\n",
       " 'Sphynx': 3.0181397505657515e-06,\n",
       " 'american_bulldog': 0.00013157303328625858,\n",
       " 'american_pit_bull_terrier': 7.981848284543958e-06,\n",
       " 'basset_hound': 0.9937812685966492,\n",
       " 'beagle': 0.004552144557237625,\n",
       " 'boxer': 3.5454352200758876e-06,\n",
       " 'chihuahua': 6.597473856118086e-08,\n",
       " 'english_cocker_spaniel': 2.870084426831454e-05,\n",
       " 'english_setter': 3.670930709631648e-06,\n",
       " 'german_shorthaired': 3.8018923078197986e-05,\n",
       " 'great_pyrenees': 2.358361371079809e-06,\n",
       " 'havanese': 8.803061064099893e-05,\n",
       " 'japanese_chin': 3.25797482219059e-05,\n",
       " 'keeshond': 8.122364306473173e-06,\n",
       " 'leonberger': 5.695419531548396e-07,\n",
       " 'miniature_pinscher': 4.1752790025384456e-07,\n",
       " 'newfoundland': 3.35386284859851e-05,\n",
       " 'pomeranian': 1.1666520549624693e-06,\n",
       " 'pug': 3.4546492315712385e-06,\n",
       " 'saint_bernard': 0.0010560948867350817,\n",
       " 'samoyed': 4.280305461179523e-07,\n",
       " 'scottish_terrier': 2.134833630407229e-05,\n",
       " 'shiba_inu': 4.996368261345197e-07,\n",
       " 'staffordshire_bull_terrier': 6.406941110981279e-07,\n",
       " 'wheaten_terrier': 4.440183147380594e-06,\n",
       " 'yorkshire_terrier': 1.7454595990784583e-06}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classify_image(im)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "caf5eded",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Leul\\AppData\\Local\\Temp\\ipykernel_11884\\1310456111.py:2: GradioDeprecationWarning: Usage of gradio.inputs is deprecated, and will not be supported in the future, please import your component from gradio.components\n",
      "  image = gr.inputs.Image(shape=(192, 192))\n",
      "C:\\Users\\Leul\\AppData\\Local\\Temp\\ipykernel_11884\\1310456111.py:2: GradioDeprecationWarning: `optional` parameter is deprecated, and it has no effect\n",
      "  image = gr.inputs.Image(shape=(192, 192))\n",
      "C:\\Users\\Leul\\AppData\\Local\\Temp\\ipykernel_11884\\1310456111.py:3: GradioDeprecationWarning: Usage of gradio.outputs is deprecated, and will not be supported in the future, please import your components from gradio.components\n",
      "  label = gr.outputs.Label()\n",
      "C:\\Users\\Leul\\AppData\\Local\\Temp\\ipykernel_11884\\1310456111.py:3: GradioUnusedKwargWarning: You have unused kwarg parameters in Label, please remove them: {'type': 'auto'}\n",
      "  label = gr.outputs.Label()\n"
     ]
    }
   ],
   "source": [
    "#export\n",
    "image = gr.inputs.Image(shape=(192, 192))\n",
    "label = gr.outputs.Label()\n",
    "examples = ['basset.jpg']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f7b93fee",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running on local URL:  http://127.0.0.1:7860\n",
      "\n",
      "To create a public link, set `share=True` in `launch()`.\n"
     ]
    },
    {
     "data": {
      "text/plain": []
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#export\n",
    "intf = gr.Interface(fn=classify_image, inputs=image, outputs=label, examples=examples)\n",
    "intf.launch(inline=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "62047859",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sequential(\n",
       "  (0): TimmBody(\n",
       "    (model): ConvNeXt(\n",
       "      (stem): Sequential(\n",
       "        (0): Conv2d(3, 96, kernel_size=(4, 4), stride=(4, 4))\n",
       "        (1): LayerNorm2d((96,), eps=1e-06, elementwise_affine=True)\n",
       "      )\n",
       "      (stages): Sequential(\n",
       "        (0): ConvNeXtStage(\n",
       "          (downsample): Identity()\n",
       "          (blocks): Sequential(\n",
       "            (0): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(96, 96, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=96)\n",
       "              (norm): LayerNorm((96,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=96, out_features=384, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=384, out_features=96, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (1): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(96, 96, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=96)\n",
       "              (norm): LayerNorm((96,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=96, out_features=384, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=384, out_features=96, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (2): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(96, 96, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=96)\n",
       "              (norm): LayerNorm((96,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=96, out_features=384, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=384, out_features=96, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "          )\n",
       "        )\n",
       "        (1): ConvNeXtStage(\n",
       "          (downsample): Sequential(\n",
       "            (0): LayerNorm2d((96,), eps=1e-06, elementwise_affine=True)\n",
       "            (1): Conv2d(96, 192, kernel_size=(2, 2), stride=(2, 2))\n",
       "          )\n",
       "          (blocks): Sequential(\n",
       "            (0): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192)\n",
       "              (norm): LayerNorm((192,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=192, out_features=768, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=768, out_features=192, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (1): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192)\n",
       "              (norm): LayerNorm((192,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=192, out_features=768, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=768, out_features=192, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (2): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192)\n",
       "              (norm): LayerNorm((192,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=192, out_features=768, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=768, out_features=192, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "          )\n",
       "        )\n",
       "        (2): ConvNeXtStage(\n",
       "          (downsample): Sequential(\n",
       "            (0): LayerNorm2d((192,), eps=1e-06, elementwise_affine=True)\n",
       "            (1): Conv2d(192, 384, kernel_size=(2, 2), stride=(2, 2))\n",
       "          )\n",
       "          (blocks): Sequential(\n",
       "            (0): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(384, 384, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=384)\n",
       "              (norm): LayerNorm((384,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=1536, out_features=384, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (1): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(384, 384, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=384)\n",
       "              (norm): LayerNorm((384,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=1536, out_features=384, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (2): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(384, 384, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=384)\n",
       "              (norm): LayerNorm((384,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=1536, out_features=384, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (3): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(384, 384, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=384)\n",
       "              (norm): LayerNorm((384,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=1536, out_features=384, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (4): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(384, 384, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=384)\n",
       "              (norm): LayerNorm((384,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=1536, out_features=384, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (5): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(384, 384, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=384)\n",
       "              (norm): LayerNorm((384,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=1536, out_features=384, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (6): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(384, 384, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=384)\n",
       "              (norm): LayerNorm((384,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=1536, out_features=384, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (7): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(384, 384, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=384)\n",
       "              (norm): LayerNorm((384,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=1536, out_features=384, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (8): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(384, 384, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=384)\n",
       "              (norm): LayerNorm((384,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=1536, out_features=384, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "          )\n",
       "        )\n",
       "        (3): ConvNeXtStage(\n",
       "          (downsample): Sequential(\n",
       "            (0): LayerNorm2d((384,), eps=1e-06, elementwise_affine=True)\n",
       "            (1): Conv2d(384, 768, kernel_size=(2, 2), stride=(2, 2))\n",
       "          )\n",
       "          (blocks): Sequential(\n",
       "            (0): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(768, 768, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=768)\n",
       "              (norm): LayerNorm((768,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=768, out_features=3072, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=3072, out_features=768, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (1): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(768, 768, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=768)\n",
       "              (norm): LayerNorm((768,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=768, out_features=3072, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=3072, out_features=768, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "            (2): ConvNeXtBlock(\n",
       "              (conv_dw): Conv2d(768, 768, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=768)\n",
       "              (norm): LayerNorm((768,), eps=1e-06, elementwise_affine=True)\n",
       "              (mlp): Mlp(\n",
       "                (fc1): Linear(in_features=768, out_features=3072, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (norm): Identity()\n",
       "                (fc2): Linear(in_features=3072, out_features=768, bias=True)\n",
       "                (drop2): Dropout(p=0.0, inplace=False)\n",
       "              )\n",
       "              (shortcut): Identity()\n",
       "              (drop_path): Identity()\n",
       "            )\n",
       "          )\n",
       "        )\n",
       "      )\n",
       "      (norm_pre): Identity()\n",
       "      (head): NormMlpClassifierHead(\n",
       "        (global_pool): SelectAdaptivePool2d (pool_type=avg, flatten=Identity())\n",
       "        (norm): LayerNorm2d((768,), eps=1e-06, elementwise_affine=True)\n",
       "        (flatten): Flatten(start_dim=1, end_dim=-1)\n",
       "        (pre_logits): Identity()\n",
       "        (drop): Dropout(p=0.0, inplace=False)\n",
       "        (fc): Identity()\n",
       "      )\n",
       "    )\n",
       "  )\n",
       "  (1): Sequential(\n",
       "    (0): AdaptiveConcatPool2d(\n",
       "      (ap): AdaptiveAvgPool2d(output_size=1)\n",
       "      (mp): AdaptiveMaxPool2d(output_size=1)\n",
       "    )\n",
       "    (1): fastai.layers.Flatten(full=False)\n",
       "    (2): BatchNorm1d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (3): Dropout(p=0.25, inplace=False)\n",
       "    (4): Linear(in_features=1536, out_features=512, bias=False)\n",
       "    (5): ReLU(inplace=True)\n",
       "    (6): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (7): Dropout(p=0.5, inplace=False)\n",
       "    (8): Linear(in_features=512, out_features=37, bias=False)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m = learn.model\n",
    "m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "624c9c73",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Parameter containing:\n",
       " tensor([ 1.2555e+00,  1.9176e+00,  1.2180e+00,  1.0397e+00, -6.9788e-04,\n",
       "          7.6582e-01,  8.8732e-01,  1.6315e+00,  7.0553e-01,  3.2894e+00,\n",
       "          7.8567e-01, -1.2476e-03,  1.0006e+00, -1.7705e-03,  3.2974e+00,\n",
       "         -8.8328e-04,  1.9839e+00,  1.0212e+00,  4.4530e+00,  2.5682e-01,\n",
       "          2.7254e+00,  9.2568e-01,  1.2383e+00,  4.4638e-03,  1.7874e+00,\n",
       "          5.4293e-01,  4.6262e+00,  1.2076e-02, -1.1175e-03,  3.4506e+00,\n",
       "          1.3523e+00,  4.1264e+00,  2.6865e+00,  4.1198e+00,  3.4007e+00,\n",
       "          8.4976e-01,  7.3672e-01,  3.9789e+00,  1.2856e+00,  6.3987e-01,\n",
       "          2.6902e+00,  1.1184e+00,  1.1697e+00,  5.5227e-01,  2.3343e+00,\n",
       "          2.8715e-03,  9.6929e-01,  2.6237e-03,  1.1977e+00,  1.7900e+00,\n",
       "          4.0211e-01,  4.5062e-01,  9.7191e-01,  3.9893e+00,  6.5823e-01,\n",
       "          6.8813e-01,  9.8591e-01,  2.7062e+00,  1.2165e+00,  7.6212e-01,\n",
       "          3.3005e+00,  1.6206e+00,  9.5593e-01,  2.1220e+00,  6.2895e-01,\n",
       "          4.0337e+00,  8.9186e-01, -2.1240e-03,  4.0865e+00,  1.0657e+00,\n",
       "          1.3974e+00,  1.6683e+00,  5.7347e-04,  7.6544e-01,  8.8428e-01,\n",
       "          6.4261e-01,  1.3443e+00,  7.1664e-01,  5.4760e-01,  2.0902e+00,\n",
       "          1.1953e+00,  3.0861e-01,  2.9652e-01,  1.4724e+00,  4.0840e+00,\n",
       "         -6.7567e-04,  1.1469e+00,  3.8848e+00,  3.6010e+00,  4.8350e-01,\n",
       "          2.1793e-01,  1.1833e-03,  6.4875e-01,  3.0072e+00,  3.0463e+00,\n",
       "          4.8006e-03], requires_grad=True),\n",
       " Parameter containing:\n",
       " tensor([-9.6541e-02, -4.1685e-02,  4.1656e+00, -1.1127e-02,  3.7692e-03,\n",
       "         -2.5879e-02, -3.1345e-02, -8.0998e-02, -1.4048e-01, -6.3114e-02,\n",
       "          3.2136e-01, -3.3704e-01, -5.8832e-02, -4.7559e-03, -4.6798e-02,\n",
       "         -2.7067e-02, -4.0918e-02, -3.9427e-02,  8.6198e-03, -2.4087e-02,\n",
       "          9.2235e-03, -1.6377e-01, -4.0150e+00,  5.2956e-01, -5.3388e-01,\n",
       "          2.8046e+00,  3.7292e-02, -9.4003e-03, -1.6751e-03, -1.1742e-01,\n",
       "         -1.3906e-01,  1.8982e-02, -9.4769e-02, -1.3003e-01, -1.9367e-01,\n",
       "         -6.8046e-02, -3.7449e-02, -1.2956e-01,  1.5339e-01,  2.0702e-04,\n",
       "         -6.4451e-02,  5.7224e-02, -9.2490e-02, -1.1461e+00, -5.2695e-02,\n",
       "         -3.2731e-03, -1.8366e-01,  2.4825e-02,  4.1198e-02, -5.8960e-02,\n",
       "         -4.1485e-02, -5.6968e-02, -4.4273e-02, -1.3014e-02, -1.1162e-01,\n",
       "          7.1587e-03, -3.7926e-02, -1.5821e-01, -9.8600e-02, -1.8663e-01,\n",
       "         -1.1151e-01, -1.8172e-01, -3.3822e-02, -2.6476e-02,  1.4198e+00,\n",
       "         -3.2238e-02, -4.2690e-02, -2.6961e-01, -4.7163e-02, -2.5859e-04,\n",
       "          2.6605e-01,  1.8801e-01,  6.9712e-01, -3.0929e-01,  8.1798e-02,\n",
       "         -1.0849e+00,  1.4461e-02, -4.4108e-02, -8.0469e-02, -6.7796e-02,\n",
       "         -1.3042e-01, -1.8991e-02, -1.7602e-02, -3.8776e-02, -7.1110e-02,\n",
       "          1.3405e-02, -5.8258e-02, -3.5131e-02, -7.9756e-02, -7.2881e-02,\n",
       "         -1.0433e-02, -3.7696e-01, -1.2890e-02, -2.6144e-02, -6.3093e-02,\n",
       "          3.8806e-05], requires_grad=True)]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "l = m.get_submodule('0.model.stem.1')\n",
    "list(l.parameters())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "2f43fd12",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Parameter containing:\n",
       " tensor([[ 2.2842e-02, -1.5819e-03,  4.0467e-02,  ...,  1.8076e-03,\n",
       "          -4.5074e-02,  7.9816e-03],\n",
       "         [-1.4371e-01,  1.6982e-02,  2.6013e-02,  ...,  1.2654e-02,\n",
       "          -1.0438e-01,  5.6389e-02],\n",
       "         [-6.5495e-02, -3.2666e-02,  5.7539e-03,  ..., -4.1672e-02,\n",
       "           6.5967e-02, -4.0074e-02],\n",
       "         ...,\n",
       "         [-8.7754e-03,  6.9858e-02,  1.2250e-04,  ...,  4.1555e-03,\n",
       "           4.1464e-02, -1.9192e-02],\n",
       "         [ 1.9366e-03,  3.2158e-02,  2.9872e-02,  ..., -2.9866e-02,\n",
       "          -3.0545e-02,  5.5596e-02],\n",
       "         [ 1.2102e-01, -3.5394e-02, -4.3373e-03,  ..., -6.3677e-03,\n",
       "           2.3553e-02, -1.1467e-02]], requires_grad=True),\n",
       " Parameter containing:\n",
       " tensor([-0.4048, -0.7420, -0.4234, -0.1651, -0.3027, -0.1899, -0.5534, -0.6271,\n",
       "         -0.3009, -0.4252, -0.5995, -0.4106, -0.2172, -1.7936, -0.3170, -0.1164,\n",
       "         -0.4483, -0.2846, -0.4342, -0.4946, -0.4065, -1.1400, -0.6754, -1.7237,\n",
       "         -0.2954, -0.2655, -0.2187, -0.3913, -0.4150, -0.4771,  0.2365, -0.7542,\n",
       "         -0.5851, -0.1819, -1.5272, -0.3625, -2.4689, -2.3461, -0.6109, -0.4113,\n",
       "         -0.6962, -0.5763, -0.5877, -0.0319, -2.0354, -0.2858, -0.3954, -0.8403,\n",
       "         -2.2400, -1.0875, -0.2295, -0.9002, -0.7585, -0.8833, -0.3753, -0.4548,\n",
       "         -0.3835, -0.4048, -2.0231, -1.0264, -0.4107, -1.1565, -0.2225, -0.4250,\n",
       "         -0.2496, -0.4224, -0.0975, -1.4017, -0.6888, -0.4371, -0.2933, -0.4641,\n",
       "         -0.4958, -1.2534, -1.0721, -1.2968, -0.6276, -1.4161, -2.3083, -2.4539,\n",
       "         -0.4260, -0.9988, -0.4636, -0.3146, -0.2418, -0.8743, -0.2827, -1.4206,\n",
       "         -0.3258, -0.3202, -0.0603, -0.1895, -0.2496, -0.6130, -0.2975, -2.1465,\n",
       "         -0.4129, -0.3676, -1.9815, -0.3813, -0.3786, -0.2292, -0.3699, -0.3256,\n",
       "         -0.5586, -2.4192, -0.4590, -1.7748, -0.3996, -0.4093, -0.3516, -0.5332,\n",
       "         -1.6534, -1.8191,  0.6264, -0.4059,  0.5871, -2.2074, -0.2440, -2.4539,\n",
       "         -0.2284, -0.6866,  0.6988,  0.6477, -0.6445, -0.3454, -0.3277, -0.5702,\n",
       "         -0.5175, -0.2773, -0.4089, -0.3018, -0.4872, -0.4952, -0.4072, -0.4355,\n",
       "         -0.5102, -0.4130, -2.0919, -0.2825, -0.5830, -1.5834,  0.6139, -0.8506,\n",
       "         -0.4669, -2.1359, -0.3417, -0.3765, -0.3345, -0.3961, -0.3884, -0.5669,\n",
       "         -0.2223, -1.3058, -0.4601, -0.3927, -0.4667, -0.4215, -0.4755, -0.2866,\n",
       "         -1.5804, -0.1787, -0.4366, -0.3172,  1.5730, -0.4045, -0.4839, -0.2577,\n",
       "         -0.5612, -0.4266, -0.2577, -0.3175, -0.4621, -1.9551, -1.9143, -0.3960,\n",
       "          0.3989, -2.3519, -0.9689, -0.2830, -1.9003, -0.4181,  0.0159, -1.1110,\n",
       "         -0.4921, -0.3177, -1.8910, -0.3101, -0.8135, -2.3346, -0.3844, -0.3848,\n",
       "         -0.1974, -0.4446, -1.6233, -2.5484, -0.3177, -1.2715, -1.1478,  0.6150,\n",
       "         -0.3750, -0.3950, -2.0748, -0.4657, -0.3781, -0.4956, -0.3282, -1.9218,\n",
       "         -2.0019, -0.5306, -0.2555, -1.1161, -0.3515, -2.2184, -1.1394,  0.5364,\n",
       "         -0.3218, -2.0387, -0.4656,  0.1847, -0.5831, -0.3128,  0.6181, -0.2124,\n",
       "         -2.3537, -0.9700, -0.9784, -0.3668, -0.4501, -1.9563, -0.2662, -1.1757,\n",
       "         -0.4199, -0.9023, -0.3605, -0.5171, -1.1881, -0.4190, -0.4770, -1.5559,\n",
       "         -0.4012, -0.6517, -0.4818, -0.2424,  0.6908, -0.5082, -0.4302, -0.6066,\n",
       "         -0.3999, -0.3329, -0.3595, -1.6108, -0.2373, -0.2467, -0.4546,  0.1807,\n",
       "         -0.3227, -0.3918, -0.3515, -0.3755, -1.2178, -0.4000, -0.3578, -0.2881,\n",
       "         -1.7484, -0.2363, -0.1599, -0.2640, -0.9768, -1.3065, -0.4148, -0.2663,\n",
       "         -0.3932, -0.4627, -0.2173,  0.2139, -0.5733, -0.2766, -0.3660, -0.5172,\n",
       "         -0.3484, -0.3363, -0.6446,  0.6867, -0.3739, -0.2901, -2.0863, -0.4883,\n",
       "         -0.2597, -1.0496, -1.6618, -0.3397, -0.5109, -0.5658, -0.3027, -0.5049,\n",
       "         -0.2877, -0.2841, -0.1984, -0.6909, -0.2871, -2.1121, -0.8929, -0.2300,\n",
       "         -1.5013, -0.4734, -2.2292, -0.4021, -0.2926, -0.4199,  0.6645, -0.3047,\n",
       "         -0.1687, -0.3748, -0.6432, -2.3346, -0.3101, -1.2731, -0.8194, -1.0594,\n",
       "         -0.0932, -1.6386,  0.3426, -0.8484, -0.4909, -0.5002, -1.0632, -0.3531,\n",
       "         -1.1564, -0.3842, -0.3172, -0.6433, -0.9083, -0.6567, -0.6490,  0.6336,\n",
       "         -0.2663, -1.3202, -1.1623, -1.2032, -2.0576, -0.3001, -1.3596, -0.4614,\n",
       "         -0.5023, -0.4948, -0.3156, -0.3271, -0.2668, -0.4280, -0.3297, -0.3011,\n",
       "         -1.6635,  0.6434, -0.9456,  0.6098, -0.4234,  0.3917, -0.4943, -0.4286,\n",
       "         -0.2588, -0.4950, -2.1991, -0.2600, -0.3934, -0.4566, -0.5817, -0.3488,\n",
       "         -0.7372, -0.3590, -0.4895, -2.0106,  0.4557, -0.8057, -1.7748, -0.3512,\n",
       "         -0.5359, -0.2102, -0.3954, -0.4780, -1.1455, -0.3976, -2.2115, -0.2838],\n",
       "        requires_grad=True)]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "l = m.get_submodule('0.model.stages.0.blocks.1.mlp.fc1')\n",
    "list(l.parameters())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a0236406",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: nbdev in c:\\users\\leul\\anaconda3\\lib\\site-packages (2.3.12)\n",
      "Requirement already satisfied: fastcore>=1.5.27 in c:\\users\\leul\\anaconda3\\lib\\site-packages (from nbdev) (1.5.29)\n",
      "Requirement already satisfied: execnb>=0.1.4 in c:\\users\\leul\\anaconda3\\lib\\site-packages (from nbdev) (0.1.5)\n",
      "Requirement already satisfied: astunparse in c:\\users\\leul\\anaconda3\\lib\\site-packages (from nbdev) (1.6.3)\n",
      "Requirement already satisfied: ghapi>=1.0.3 in c:\\users\\leul\\anaconda3\\lib\\site-packages (from nbdev) (1.0.3)\n",
      "Requirement already satisfied: watchdog in c:\\users\\leul\\anaconda3\\lib\\site-packages (from nbdev) (2.1.6)\n",
      "Requirement already satisfied: asttokens in c:\\users\\leul\\anaconda3\\lib\\site-packages (from nbdev) (2.0.5)\n",
      "Requirement already satisfied: PyYAML in c:\\users\\leul\\anaconda3\\lib\\site-packages (from nbdev) (6.0)\n",
      "Requirement already satisfied: ipython in c:\\users\\leul\\anaconda3\\lib\\site-packages (from execnb>=0.1.4->nbdev) (8.15.0)\n",
      "Requirement already satisfied: pip in c:\\users\\leul\\anaconda3\\lib\\site-packages (from fastcore>=1.5.27->nbdev) (23.2.1)\n",
      "Requirement already satisfied: packaging in c:\\users\\leul\\anaconda3\\lib\\site-packages (from fastcore>=1.5.27->nbdev) (23.1)\n",
      "Requirement already satisfied: six in c:\\users\\leul\\anaconda3\\lib\\site-packages (from asttokens->nbdev) (1.16.0)\n",
      "Requirement already satisfied: wheel<1.0,>=0.23.0 in c:\\users\\leul\\anaconda3\\lib\\site-packages (from astunparse->nbdev) (0.38.4)\n",
      "Requirement already satisfied: backcall in c:\\users\\leul\\anaconda3\\lib\\site-packages (from ipython->execnb>=0.1.4->nbdev) (0.2.0)\n",
      "Requirement already satisfied: decorator in c:\\users\\leul\\anaconda3\\lib\\site-packages (from ipython->execnb>=0.1.4->nbdev) (5.1.1)\n",
      "Requirement already satisfied: jedi>=0.16 in c:\\users\\leul\\anaconda3\\lib\\site-packages (from ipython->execnb>=0.1.4->nbdev) (0.18.1)\n",
      "Requirement already satisfied: matplotlib-inline in c:\\users\\leul\\anaconda3\\lib\\site-packages (from ipython->execnb>=0.1.4->nbdev) (0.1.6)\n",
      "Requirement already satisfied: pickleshare in c:\\users\\leul\\anaconda3\\lib\\site-packages (from ipython->execnb>=0.1.4->nbdev) (0.7.5)\n",
      "Requirement already satisfied: prompt-toolkit!=3.0.37,<3.1.0,>=3.0.30 in c:\\users\\leul\\anaconda3\\lib\\site-packages (from ipython->execnb>=0.1.4->nbdev) (3.0.36)\n",
      "Requirement already satisfied: pygments>=2.4.0 in c:\\users\\leul\\anaconda3\\lib\\site-packages (from ipython->execnb>=0.1.4->nbdev) (2.15.1)\n",
      "Requirement already satisfied: stack-data in c:\\users\\leul\\anaconda3\\lib\\site-packages (from ipython->execnb>=0.1.4->nbdev) (0.2.0)\n",
      "Requirement already satisfied: traitlets>=5 in c:\\users\\leul\\anaconda3\\lib\\site-packages (from ipython->execnb>=0.1.4->nbdev) (5.7.1)\n",
      "Requirement already satisfied: colorama in c:\\users\\leul\\anaconda3\\lib\\site-packages (from ipython->execnb>=0.1.4->nbdev) (0.4.6)\n",
      "Requirement already satisfied: parso<0.9.0,>=0.8.0 in c:\\users\\leul\\anaconda3\\lib\\site-packages (from jedi>=0.16->ipython->execnb>=0.1.4->nbdev) (0.8.3)\n",
      "Requirement already satisfied: wcwidth in c:\\users\\leul\\anaconda3\\lib\\site-packages (from prompt-toolkit!=3.0.37,<3.1.0,>=3.0.30->ipython->execnb>=0.1.4->nbdev) (0.2.5)\n",
      "Requirement already satisfied: executing in c:\\users\\leul\\anaconda3\\lib\\site-packages (from stack-data->ipython->execnb>=0.1.4->nbdev) (0.8.3)\n",
      "Requirement already satisfied: pure-eval in c:\\users\\leul\\anaconda3\\lib\\site-packages (from stack-data->ipython->execnb>=0.1.4->nbdev) (0.2.2)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install nbdev"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "1d300cfb",
   "metadata": {},
   "outputs": [],
   "source": [
    "import nbdev\n",
    "nbdev.export.nb_export('app.ipynb', './')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f15512ba",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "211253ef",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "86bc6af4",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3b222084",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}