app.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0fe6c180-71e8-4d58-885f-2154f51d4c41",
   "metadata": {},
   "outputs": [],
   "source": [
    "#|default_exp app"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "72ace8fd-a43e-41ee-a8a3-798a2c3a3695",
   "metadata": {},
   "outputs": [],
   "source": [
    "#|export\n",
    "from fastai.vision.all import *\n",
    "import gradio as gr\n",
    "import timm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "dbaf9b3c-e0ad-4ea1-8dd4-50fec8d51677",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/jpeg": "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",
      "image/png": "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",
      "text/plain": [
       "PILImage mode=RGB size=224x149"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "im = PILImage.create('basset.jpg')\n",
    "im.thumbnail((224,224))\n",
    "im"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "44aaef01-0529-4f17-afe9-d666691a9bf7",
   "metadata": {},
   "outputs": [],
   "source": [
    "#|export\n",
    "learn = load_learner('model.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "ab19a465-146b-417d-875b-02da5ba0aea5",
   "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([1.1834e-07, 3.6242e-07, 4.8684e-08, 2.7621e-07, 1.1813e-08, 1.5572e-06,\n",
       "         3.6965e-07, 1.6971e-06, 4.7838e-08, 1.0521e-07, 1.8542e-07, 1.7373e-08,\n",
       "         6.3202e-09, 7.5863e-08, 9.9992e-01, 5.7861e-05, 4.2675e-08, 3.4805e-07,\n",
       "         4.3144e-06, 5.8502e-08, 1.6396e-06, 1.0636e-07, 9.4755e-08, 6.4813e-08,\n",
       "         6.5368e-07, 9.2687e-08, 1.3983e-07, 1.2649e-07, 1.9477e-07, 1.1417e-06,\n",
       "         5.6963e-08, 3.7510e-08, 4.2811e-06, 2.9909e-08, 8.1630e-09, 2.0253e-08,\n",
       "         7.5729e-07]))"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "learn.predict(im)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "33ea7bbd-2f36-4e2a-a74e-d26f14806cd5",
   "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": 14,
   "id": "60199c5f-60c6-423d-972a-c2770398bc56",
   "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': 1.1833922997084301e-07,\n",
       " 'Bengal': 3.624234352628264e-07,\n",
       " 'Birman': 4.868387648571115e-08,\n",
       " 'Bombay': 2.7620779974313336e-07,\n",
       " 'British_Shorthair': 1.1813022560147601e-08,\n",
       " 'Egyptian_Mau': 1.5571941958114621e-06,\n",
       " 'Maine_Coon': 3.696516159834573e-07,\n",
       " 'Persian': 1.697121660981793e-06,\n",
       " 'Ragdoll': 4.783786522466471e-08,\n",
       " 'Russian_Blue': 1.0520943050096321e-07,\n",
       " 'Siamese': 1.8542475288541027e-07,\n",
       " 'Sphynx': 1.737347865571337e-08,\n",
       " 'american_bulldog': 6.320234557932736e-09,\n",
       " 'american_pit_bull_terrier': 7.586324812791645e-08,\n",
       " 'basset_hound': 0.9999231100082397,\n",
       " 'beagle': 5.7861499954015017e-05,\n",
       " 'boxer': 4.267499065235825e-08,\n",
       " 'chihuahua': 3.4804895676643355e-07,\n",
       " 'english_cocker_spaniel': 4.314403213356854e-06,\n",
       " 'english_setter': 5.8502276800709296e-08,\n",
       " 'german_shorthaired': 1.6395551938330755e-06,\n",
       " 'great_pyrenees': 1.0636112790507468e-07,\n",
       " 'havanese': 9.475484574750226e-08,\n",
       " 'japanese_chin': 6.481283776338387e-08,\n",
       " 'keeshond': 6.536837418025243e-07,\n",
       " 'leonberger': 9.268669742823477e-08,\n",
       " 'miniature_pinscher': 1.3982798918732442e-07,\n",
       " 'newfoundland': 1.264948963353163e-07,\n",
       " 'pomeranian': 1.947668977209105e-07,\n",
       " 'pug': 1.1417224641263601e-06,\n",
       " 'saint_bernard': 5.6963376238172714e-08,\n",
       " 'samoyed': 3.751043209376803e-08,\n",
       " 'scottish_terrier': 4.281065230316017e-06,\n",
       " 'shiba_inu': 2.990866221352917e-08,\n",
       " 'staffordshire_bull_terrier': 8.163000053684755e-09,\n",
       " 'wheaten_terrier': 2.0252580057444902e-08,\n",
       " 'yorkshire_terrier': 7.572912750219984e-07}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classify_image(im)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "b7848874-0411-42c7-8a7d-bcd934d3ba75",
   "metadata": {},
   "outputs": [],
   "source": [
    "#|export\n",
    "image = gr.Image(height=192, width=192) # lecture material deprecated\n",
    "label = gr.Label()\n",
    "examples = ['basset.jpg']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "65b81109-29c5-4f16-b74a-ad981bf72ea3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "* Running on local URL:  http://127.0.0.1:7863\n",
      "\n",
      "To create a public link, set `share=True` in `launch()`.\n"
     ]
    },
    {
     "data": {
      "text/plain": []
     },
     "execution_count": 30,
     "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": 32,
   "id": "1ba99304-4045-49ea-b67e-bfe9aced4ab3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Export Successful\n"
     ]
    }
   ],
   "source": [
    "import nbdev\n",
    "nbdev.export.nb_export('app.ipynb', './') # To save in same dir\n",
    "print('Export Successful')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "62b2b6c7-592d-44b4-9008-0c7536093989",
   "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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=96, out_features=384, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=96, out_features=384, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=96, out_features=384, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=192, out_features=768, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=192, out_features=768, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=192, out_features=768, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=384, out_features=1536, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=768, out_features=3072, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=768, out_features=3072, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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): GlobalResponseNormMlp(\n",
       "                (fc1): Linear(in_features=768, out_features=3072, bias=True)\n",
       "                (act): GELU()\n",
       "                (drop1): Dropout(p=0.0, inplace=False)\n",
       "                (grn): GlobalResponseNorm()\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": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m = learn.model\n",
    "m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "c9969a80-ec04-4f54-9deb-e87d23ca9398",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Parameter containing:\n",
       " tensor([ 4.9261e+00, -8.8871e-03,  1.1226e+00,  3.4869e-01,  1.9858e-01,\n",
       "          4.7761e-01, -8.8127e-03,  1.9133e+00,  3.0253e+00,  1.5562e+00,\n",
       "          5.8044e-01,  2.7812e-03,  3.3807e+00,  1.2346e+00, -5.6217e-03,\n",
       "          2.1261e+00,  1.7260e+00,  7.8030e-01,  2.3202e+00,  3.1628e+00,\n",
       "          1.6729e+00,  1.3232e+00,  3.6975e-01,  2.2164e+00,  1.8384e-01,\n",
       "          1.5510e-01,  1.8206e+00, -7.3682e-03,  2.3788e+00,  3.6732e+00,\n",
       "          4.0146e-01,  1.2551e-02,  3.6649e-01,  1.1996e+00,  7.2464e-01,\n",
       "          2.8378e-01,  2.4163e+00,  3.1019e-01,  7.0898e-01,  6.3262e-01,\n",
       "          7.8462e-01,  7.8624e-04,  2.8187e-01,  5.4559e-01,  8.0706e-01,\n",
       "          3.3379e-01,  7.9569e-01,  6.0122e-01,  1.9981e-01,  3.4817e-01,\n",
       "          2.2388e+00,  1.0888e-02,  1.1574e+00, -4.9267e-03,  3.2374e+00,\n",
       "          5.8425e-01,  2.2246e-01,  4.1916e+00,  2.9523e-01,  8.4290e-01,\n",
       "          2.4097e-04,  6.1834e-03,  6.6644e-01,  2.4106e-01,  1.2585e+00,\n",
       "          2.7601e-01,  6.4914e-01,  2.5466e-01,  4.6883e+00,  6.5842e-01,\n",
       "          1.4914e-03,  4.7689e+00,  3.3615e+00,  2.8126e-01,  5.7262e-01,\n",
       "          5.2182e-01,  2.9029e+00,  4.1523e-01,  6.5620e-01,  3.0521e+00,\n",
       "          9.0350e-03,  1.2095e-01,  1.3768e+00,  3.0706e-01,  3.1004e+00,\n",
       "          4.7427e-01,  1.2646e+00,  3.0499e-01,  3.0932e+00,  3.6443e-01,\n",
       "          2.8410e+00,  2.4744e-01,  2.1994e-01,  2.9492e+00,  4.2457e-01,\n",
       "          1.6487e-01], requires_grad=True),\n",
       " Parameter containing:\n",
       " tensor([ 2.1153e-02,  7.1709e-01,  3.1521e-01, -9.8799e-01,  3.5578e-03,\n",
       "         -2.4537e+00, -8.3321e-01,  2.0586e+00,  5.8227e-02,  1.3452e-01,\n",
       "          9.8413e-03, -2.0268e-03,  2.7724e-02, -1.6198e+00,  2.3026e-01,\n",
       "          4.9697e-02,  7.0849e-02, -4.2198e-02, -4.3060e-03,  3.1771e-02,\n",
       "         -6.2846e-03, -1.3942e-03,  5.4756e-03, -1.1883e-01,  4.6279e-02,\n",
       "          3.5501e-02, -9.3745e-02,  1.2097e+00,  2.5377e-01,  4.7725e-02,\n",
       "          2.1996e-03,  3.2848e-01,  1.1331e-02, -5.7414e-02,  1.7438e-02,\n",
       "          2.7921e-02,  5.0494e-01,  1.0246e-02, -2.0180e-02, -2.7485e-03,\n",
       "          2.8702e-02,  2.9092e-01,  2.0980e-02, -6.7508e-01, -8.7712e-02,\n",
       "          5.2626e-02,  3.3390e-02,  2.0174e+00,  2.7407e-02, -1.1986e-02,\n",
       "          7.9388e-02,  4.9428e-02,  7.8484e-02,  2.8448e-03,  3.8502e-02,\n",
       "          1.0036e-01,  7.8661e-03, -4.8637e-03,  2.2506e-02, -7.0692e-03,\n",
       "          3.3882e-01, -5.3049e-01,  2.8798e-02, -1.4062e-02,  2.2939e+00,\n",
       "         -4.0604e-02,  1.0977e-01, -5.6906e-03,  7.3702e-03,  1.1824e-02,\n",
       "         -3.2739e-04,  3.8008e-02,  3.6318e-02, -6.7542e-04,  9.2092e-02,\n",
       "          1.1813e+00, -1.3646e-01, -1.5180e+00,  1.4595e-01,  6.0408e-02,\n",
       "         -5.0736e-02,  2.3435e-02, -7.4179e-02,  3.6609e-04,  5.8384e-02,\n",
       "         -2.3146e-01,  2.8343e-01, -1.9405e-02, -6.3685e-02, -2.4096e-02,\n",
       "         -4.4026e-02, -1.9353e-02,  1.3453e-02, -3.6983e-02, -3.6457e-02,\n",
       "          3.6066e-03], requires_grad=True)]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "l = m.get_submodule('0.model.stem.1')\n",
    "list(l.parameters())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "bfc377fb-fb0a-46b6-95d0-499b5f17abbc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Parameter containing:\n",
       " tensor([[-0.0549, -0.0274, -0.0116,  ...,  0.0161,  0.0373, -0.0029],\n",
       "         [ 0.0018, -0.0521,  0.0143,  ..., -0.0252, -0.0422,  0.0049],\n",
       "         [-0.0103,  0.0004, -0.0078,  ..., -0.0394, -0.0060, -0.0004],\n",
       "         ...,\n",
       "         [-0.0040, -0.0124, -0.0328,  ..., -0.0215, -0.0049,  0.0024],\n",
       "         [ 0.0305,  0.0005,  0.0191,  ...,  0.0493, -0.0117, -0.0040],\n",
       "         [-0.0076,  0.0013, -0.0087,  ...,  0.0120,  0.0137,  0.0040]],\n",
       "        requires_grad=True),\n",
       " Parameter containing:\n",
       " tensor([-0.8710, -0.7155, -1.0197, -1.0517, -0.5571, -0.8992, -0.7452, -1.1259,\n",
       "         -0.9349, -1.1854, -0.8042, -0.7971,  0.1005, -1.5095, -0.9240, -0.6481,\n",
       "          0.6850, -1.1375, -0.0261, -1.0961, -0.8624, -0.7149, -0.6821,  0.2149,\n",
       "         -1.6132, -0.6693, -0.4569, -0.0345, -0.6725, -0.4703, -0.8043,  0.0036,\n",
       "         -0.4164,  1.1497, -0.9721, -0.2357, -0.8716, -1.0910, -0.5575,  0.1451,\n",
       "         -1.3999,  0.1282, -0.4641, -0.5791,  0.0431, -1.1841,  0.1650, -1.1270,\n",
       "         -1.4342, -1.0155, -1.0422, -1.0554, -0.9646, -0.8282, -0.1248, -0.9160,\n",
       "         -0.1742, -1.1329, -0.5556, -0.9600, -0.5444, -0.1855, -0.9322, -0.9431,\n",
       "         -0.6303, -0.1016, -0.4250, -1.1639, -1.0388, -0.8601,  0.2182, -0.5836,\n",
       "         -0.5519, -0.9180, -1.5489, -1.1364, -0.5259,  0.1579, -1.8885, -0.3252,\n",
       "         -1.0519, -0.3035, -1.0424,  0.0470, -0.5451, -0.7240, -0.0128, -0.2205,\n",
       "         -0.3006, -1.3354, -1.4727, -0.8168,  0.1200, -1.1841, -1.0891, -1.0196,\n",
       "         -0.7406, -0.8362,  0.2723, -0.6673, -0.8509, -0.9288, -0.7535,  0.3409,\n",
       "         -0.8399, -0.8583, -1.1104,  0.0253, -0.8192, -0.6778, -0.8823, -0.8687,\n",
       "         -0.6498, -1.0601, -0.7346, -1.5595, -1.2032, -0.8686, -1.1627, -0.2544,\n",
       "         -0.9174, -0.6721, -1.1432, -0.1054, -1.3272, -0.2871, -0.8455, -0.9920,\n",
       "         -1.0520, -0.0887, -0.7570, -1.0305, -1.6802, -1.0804, -1.0056, -1.2723,\n",
       "         -0.9601, -0.7422, -2.0823, -1.6651,  1.2235, -1.2329,  1.6335, -0.2864,\n",
       "         -0.1724,  0.8120, -1.0294, -1.9505, -0.6095,  0.1257,  0.0601, -1.1687,\n",
       "         -1.1579, -1.0781, -0.7345, -1.3378, -0.6893, -0.0307, -0.7308, -1.6322,\n",
       "         -0.7487, -0.7780, -0.4587, -0.7457, -0.6080, -1.0781, -0.5075,  0.0575,\n",
       "         -0.9883,  0.0602, -1.1019, -0.0195, -0.5551, -1.0399,  0.1955, -1.9137,\n",
       "         -1.0614, -0.1942, -0.8215, -0.5152, -0.8812, -1.1114, -2.1140,  0.0240,\n",
       "         -0.9122, -0.3440, -0.6731, -0.7381, -1.1557, -0.7441, -0.9894, -0.7292,\n",
       "         -1.2092, -0.9573, -0.5806, -1.3261,  0.4017, -0.9669, -0.9634,  0.0251,\n",
       "          0.5222,  0.1061,  0.0963, -0.6362, -1.4618,  0.4937, -1.1745, -0.7144,\n",
       "          0.1577, -0.7498, -0.1472,  0.0356, -0.0550, -0.6171, -0.2946, -0.2751,\n",
       "         -0.1773, -0.8614, -0.8889, -0.8137, -0.4144, -1.0673, -0.9342, -0.8888,\n",
       "         -1.0049, -0.5802, -1.0733, -0.9503, -0.8324,  0.2892, -0.6218, -1.5728,\n",
       "         -0.9147, -0.8388, -0.2166, -0.8134, -1.2772, -0.8947, -0.8611, -0.4659,\n",
       "         -0.2381, -0.4285, -1.1688, -1.1162, -0.0031, -1.0269, -1.0406,  0.0294,\n",
       "         -0.7894, -0.7894, -0.0619, -1.1784, -0.6535, -1.0078,  0.1786, -0.5676,\n",
       "         -0.6162, -0.0451, -1.1747, -1.0208, -0.0352, -0.0367, -0.7860, -1.0950,\n",
       "         -0.7335, -2.0821, -0.2957, -0.5171, -1.0888,  0.6528, -1.8124, -0.5819,\n",
       "         -0.7847, -0.7482, -0.9316, -1.7908, -0.7554, -1.3785, -0.8370, -0.8694,\n",
       "         -1.0020, -1.1709,  0.2856,  0.3897, -1.3105, -0.1779, -1.4977, -0.6688,\n",
       "         -1.5143,  0.3117, -0.3610, -0.1486, -0.8931, -0.8731, -0.0662, -0.0780,\n",
       "          0.0266, -1.0102, -0.9474, -0.3682, -0.8499,  0.2566, -0.7618, -0.9397,\n",
       "         -1.1221, -0.8798, -0.9013, -0.8465, -1.0566, -0.4912, -0.6898, -0.7957,\n",
       "         -0.8566, -1.3176, -1.7420, -1.2748, -0.4087, -1.3843, -0.8035, -0.5892,\n",
       "         -0.8959, -0.9003,  0.2714, -0.4426, -0.6128, -0.4060, -0.0892, -1.0747,\n",
       "         -1.1280, -0.7393, -0.9338, -0.8511,  0.2821, -0.3915, -0.8365, -0.8208,\n",
       "         -1.1843, -0.2385, -0.4321, -1.4915, -1.2293,  0.0047, -1.3581, -0.6742,\n",
       "         -0.8112, -0.9943, -0.9222, -0.1512, -1.8718, -0.1446, -0.7375, -1.2760,\n",
       "         -1.1872, -0.4652, -1.4105, -1.0674,  0.0847, -0.1263, -0.4053, -0.2346,\n",
       "         -1.1281, -0.6818, -1.4250, -1.2238,  0.1913,  0.1989,  0.4873, -1.0196,\n",
       "         -0.3932, -0.8783, -1.0143, -1.2413, -1.0231, -0.8865,  0.3379, -0.5213,\n",
       "         -0.4395, -1.1520, -0.9902, -0.4052, -1.1923, -1.2056, -0.2370, -0.6130],\n",
       "        requires_grad=True)]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "l = m.get_submodule('0.model.stages.0.blocks.1.mlp.fc1')\n",
    "list(l.parameters())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "10dcaad1-0714-4a25-9a11-61aea1bf6c6f",
   "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": "3.12.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}