{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#|default_exp demo"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Basic setup, once again..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "#|export\n",
    "from fastai.vision.all import *\n",
    "import pathlib\n",
    "import gradio as gr\n",
    "plt = platform.system()\n",
    "if plt == 'Linux': pathlib.WindowsPath = pathlib.PosixPath\n",
    "\n",
    "def is_albani(path):\n",
    "  return parent_label(path) == \"albani\""
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test with own images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "albani = PILImage.create('albani2.jpg')\n",
    "\n",
    "albani.thumbnail((192, 192))\n",
    "albani"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn = load_learner(Path('./resnet18-albani.pkl'))\n",
    "\n",
    "learn.predict(albani)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Gradio demo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running on local URL:  http://127.0.0.1:7873\n",
      "\n",
      "To create a public link, set `share=True` in `launch()`.\n"
     ]
    },
    {
     "data": {
      "text/plain": []
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#|export\n",
    "\n",
    "learn = load_learner(Path('./resnet18-albani.pkl'))\n",
    "categories = ('Dårlig Øl', 'God Øl')\n",
    "\n",
    "def classify_image(img):\n",
    "  pred,idx,probs = learn.predict(img)\n",
    "  return dict(zip(categories, map(float, probs)))\n",
    "\n",
    "title = \"\"\"\n",
    "  # Er du i tvivl om at den øl du sidder med i hånden lige, nu er god?\n",
    "  ### Tvivl ej 68 års invotation inden for machine learning skal nok fortælle dig om øllen er god eller ej\n",
    "\"\"\"\n",
    "image = gr.Image(shape=(192, 192))\n",
    "label = gr.Label()\n",
    "examples = ['albani.jpg', 'albani2.jpg', 'heineken.jpg', 'carlsberg.jpg']\n",
    "\n",
    "iface = gr.Interface(\n",
    "  fn=classify_image, \n",
    "  inputs=image, \n",
    "  outputs=label, \n",
    "  examples=examples, \n",
    "  description=title,\n",
    "  title=\"Er du i tvivl om at den øl du sidder med i hånden lige nu er god? \"\n",
    ")\n",
    "\n",
    "iface.launch(inline=False)\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Build for huggingface"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO: Successfully saved requirements file in ./requirements.txt\n"
     ]
    }
   ],
   "source": [
    "from nbdev.export import nb_export\n",
    "\n",
    "nb_export('model-test.ipynb', '.')\n",
    "! pipreqs . --force\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "fastai",
   "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": "3.10.8"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "b977fc7f635931fb2de4a5e9c6503b6f91cf50b058799eb3f69da04d3aca2546"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}