{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"id": "c7070975-920c-426e-8f2d-19f9e70fec81",
"metadata": {},
"outputs": [],
"source": [
"#|default_exp app"
]
},
{
"cell_type": "markdown",
"id": "fc3cd6f5-7bb2-4ef3-bb8f-5c05043949d9",
"metadata": {},
"source": [
"# Bear Classifier App\n",
"\n",
"This notebook creates uses an exported model `export.pkl` for a bear classifier, to create a python script which can run the model on HuggingFace. "
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "4defcbc0-6c1d-413e-a858-5aaa7c73d994",
"metadata": {},
"outputs": [],
"source": [
"#|export\n",
"from fastai.vision.all import *\n",
"import gradio as gr"
]
},
{
"cell_type": "markdown",
"id": "a9ff5538-2fa7-428e-8464-9a5e26debc09",
"metadata": {},
"source": [
"Let's take a look at an example picture:"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "209f8ab6-3fa8-458a-bcfe-ed60e35fd480",
"metadata": {},
"outputs": [
{
"ename": "FileNotFoundError",
"evalue": "[Errno 2] No such file or directory: 'blackbear.jpg'",
"output_type": "error",
"traceback": [
"\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[1;31mFileNotFoundError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[1;32mIn[3], line 3\u001b[0m\n\u001b[0;32m 1\u001b[0m \u001b[38;5;66;03m#im = PILImage.create('teddybear.jpg')\u001b[39;00m\n\u001b[0;32m 2\u001b[0m \u001b[38;5;66;03m#im = PILImage.create('grizzly.jpg')\u001b[39;00m\n\u001b[1;32m----> 3\u001b[0m im \u001b[38;5;241m=\u001b[39m PILImage\u001b[38;5;241m.\u001b[39mcreate(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mblackbear.jpg\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m 5\u001b[0m im\u001b[38;5;241m.\u001b[39mthumbnail((\u001b[38;5;241m192\u001b[39m, \u001b[38;5;241m192\u001b[39m))\n\u001b[0;32m 6\u001b[0m im\n",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\fastai\\vision\\core.py:125\u001b[0m, in \u001b[0;36mPILBase.create\u001b[1;34m(cls, fn, **kwargs)\u001b[0m\n\u001b[0;32m 123\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(fn,\u001b[38;5;28mbytes\u001b[39m): fn \u001b[38;5;241m=\u001b[39m io\u001b[38;5;241m.\u001b[39mBytesIO(fn)\n\u001b[0;32m 124\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(fn,Image\u001b[38;5;241m.\u001b[39mImage): \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mcls\u001b[39m(fn)\n\u001b[1;32m--> 125\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mcls\u001b[39m(load_image(fn, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mmerge(\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m_open_args, kwargs)))\n",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\fastai\\vision\\core.py:98\u001b[0m, in \u001b[0;36mload_image\u001b[1;34m(fn, mode)\u001b[0m\n\u001b[0;32m 96\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mload_image\u001b[39m(fn, mode\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m):\n\u001b[0;32m 97\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mOpen and load a `PIL.Image` and convert to `mode`\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m---> 98\u001b[0m im \u001b[38;5;241m=\u001b[39m Image\u001b[38;5;241m.\u001b[39mopen(fn)\n\u001b[0;32m 99\u001b[0m im\u001b[38;5;241m.\u001b[39mload()\n\u001b[0;32m 100\u001b[0m im \u001b[38;5;241m=\u001b[39m im\u001b[38;5;241m.\u001b[39m_new(im\u001b[38;5;241m.\u001b[39mim)\n",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\PIL\\Image.py:3227\u001b[0m, in \u001b[0;36mopen\u001b[1;34m(fp, mode, formats)\u001b[0m\n\u001b[0;32m 3224\u001b[0m filename \u001b[38;5;241m=\u001b[39m fp\n\u001b[0;32m 3226\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m filename:\n\u001b[1;32m-> 3227\u001b[0m fp \u001b[38;5;241m=\u001b[39m builtins\u001b[38;5;241m.\u001b[39mopen(filename, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mrb\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m 3228\u001b[0m exclusive_fp \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n\u001b[0;32m 3230\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n",
"\u001b[1;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: 'blackbear.jpg'"
]
}
],
"source": [
"#im = PILImage.create('teddybear.jpg')\n",
"#im = PILImage.create('grizzly.jpg')\n",
"im = PILImage.create('blackbear.jpg')\n",
"\n",
"im.thumbnail((192, 192))\n",
"im"
]
},
{
"cell_type": "markdown",
"id": "8373d8af-73cb-4d72-8a9d-c19228a03355",
"metadata": {},
"source": [
"Let's import the model and create the learner:"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "6e9106b7-9c6c-4b1a-8d15-777b0e44ad60",
"metadata": {},
"outputs": [],
"source": [
"#|export\n",
"learn = load_learner('export.pkl')"
]
},
{
"cell_type": "markdown",
"id": "bad6e35f-536a-4de4-b914-1ca4e3ec3fd5",
"metadata": {},
"source": [
"With the learner we can to the predictions (inference):"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "52a0bb5a-e5a9-4905-a45f-4f613701c207",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"('black', TensorBase(0), TensorBase([9.9979e-01, 1.9130e-04, 1.6887e-05]))"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"learn.predict(im)"
]
},
{
"cell_type": "markdown",
"id": "c9519ddb-f6dd-4e8c-ba67-3ba748593911",
"metadata": {},
"source": [
"The available categories are contained in the vocab:"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "82cd137c-b375-4902-a572-95a5d723fb3f",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['black', 'grizzly', 'teddy']"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"learn.dls.vocab"
]
},
{
"cell_type": "markdown",
"id": "49c774b1-96c0-4a3a-82e9-8453de8b26cd",
"metadata": {},
"source": [
"This is the function to classify the images:"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "1f479df5-838d-4f33-909c-703d39912d9c",
"metadata": {},
"outputs": [],
"source": [
"#|export\n",
"def classify_image(img):\n",
" pred,pred_idx,probs = learn.predict(img)\n",
" return dict(zip(learn.dls.vocab, map(float, probs)))"
]
},
{
"cell_type": "markdown",
"id": "15b2bff4-0a3e-4f6d-b628-341759c6508e",
"metadata": {},
"source": [
"Testing the function:"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "9049900f-0cd1-4a33-9a5c-e5bc08062dbb",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"{'black': 0.9997918009757996,\n",
" 'grizzly': 0.0001913038140628487,\n",
" 'teddy': 1.688681004452519e-05}"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"classify_image(im)"
]
},
{
"cell_type": "markdown",
"id": "16d81693-3786-4dee-8edf-92fad120bc91",
"metadata": {},
"source": [
"## Gradio App\n",
"\n",
"Now it is time to create the gradio app:"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "c95ee1ca-5feb-4011-b99c-3241468d3e3a",
"metadata": {},
"outputs": [],
"source": [
"# commented, because it produced warnings\n",
"\n",
"#image = gr.inputs.Image(shape=(192,192))\n",
"#label = gr.outputs.Label()"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "5ca3c2b5-1691-434f-acf8-f0cb91bf32b7",
"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": [
"(, 'http://127.0.0.1:7860/', None)"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#|export\n",
"image = gr.components.Image(shape=(192,192))\n",
"label = gr.components.Label()\n",
"examples = ['teddybear.jpg', 'grizzly.jpg', 'blackbear.jpg']\n",
"\n",
"intf = gr.Interface(fn=classify_image, inputs=image, outputs=label, examples=examples)\n",
"intf.launch(inline=False)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "63b8cbb0-40f8-495d-995e-9f946e28ae98",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Closing server running on port: 7860\n"
]
}
],
"source": [
"intf.close()"
]
},
{
"cell_type": "markdown",
"id": "585bbd25-7895-40df-b15c-790f1ca058a2",
"metadata": {},
"source": [
"## Export\n",
"\n",
"Finally, we export the code in the cells which are marked with `#|export`:"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "dcc8edc2-d153-4221-92bd-3e32277a65b7",
"metadata": {},
"outputs": [],
"source": [
"# commented, because it does not work\n",
"\n",
"# import notebook2script from nbdev.export\n",
"# notebook2script('app.ipynb')"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "5031b82c-67fa-4cac-b1a4-9c52646795b1",
"metadata": {},
"outputs": [],
"source": [
"from nbdev import nbdev_export"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "85bbbb28-b48d-4531-9179-f98ebb751508",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Export successful\n"
]
}
],
"source": [
"nbdev_export('app.ipynb')\n",
"print('Export successful')"
]
}
],
"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"
},
"toc": {
"base_numbering": 1,
"nav_menu": {},
"number_sections": true,
"sideBar": true,
"skip_h1_title": false,
"title_cell": "Table of Contents",
"title_sidebar": "Contents",
"toc_cell": false,
"toc_position": {},
"toc_section_display": true,
"toc_window_display": false
},
"vscode": {
"interpreter": {
"hash": "d9da906b64701e68312bc07fbc15a3a13814f930718c2c6b0e41a29d035806a3"
}
}
},
"nbformat": 4,
"nbformat_minor": 5
}