{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "import sys, os\n",
    "import gradio as gr\n",
    "import numpy as np\n",
    "sys.path.append('stylegan3')\n",
    "import utils\n",
    "\n",
    "def to_uint8(im, ndim=2):\n",
    "    im -= np.min(im)\n",
    "    im /= np.max(im)\n",
    "    im *= 255.\n",
    "    im = np.asarray(im, dtype=np.uint8)\n",
    "    if ndim == 3:    \n",
    "        if im.ndim == 2:\n",
    "            im = np.expand_dims(im, axis=-1)\n",
    "        elif im.ndim == 3:\n",
    "            if im.shape[0] == 1: \n",
    "                np.transpose(im, (1,2,0))\n",
    "        im = np.tile(im, (1,1,3)) #make fake RGB\n",
    "        return im\n",
    "    elif ndim ==2:\n",
    "        if im.ndim == 2:\n",
    "            return im\n",
    "        if im.ndim == 3:\n",
    "            if im.shape[0] == 1:  #[1, H, W]\n",
    "                return im[0,...]\n",
    "            elif im.shape[2] == 1: #[H, W, 1]\n",
    "                return im[...,0]\n",
    "            else:\n",
    "                raise AssertionError(f\"Unexpected image passed to to_uint8 with shape: {np.shape(im)}.\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running on local URL:  http://127.0.0.1:7868\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-11 17:33:10.390 INFO    paramiko.transport: Connected (version 2.0, client OpenSSH_7.6p1)\n",
      "2022-10-11 17:33:11.271 INFO    paramiko.transport: Authentication (publickey) successful!\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running on public URL: https://23528.gradio.app\n",
      "\n",
      "This share link expires in 72 hours. For free permanent hosting, check out Spaces: https://huggingface.co/spaces\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div><iframe src=\"https://23528.gradio.app\" width=\"900\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tip: The inputs and outputs flagged by the users are stored in the flagging directory, specified by the flagging_dir= kwarg. You can view this data through the interface by setting the examples= kwarg to the flagging directory; for example gr.Interface(..., examples='flagged')\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(<gradio.routes.App at 0x7fcded9da880>,\n",
       " 'http://127.0.0.1:7868/',\n",
       " 'https://23528.gradio.app')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "\n",
    "in_gpu = False\n",
    "num_images = 1\n",
    "G = utils.load_default_gen(in_gpu=in_gpu)\n",
    "sampler = utils.SampleFromGAN(G=G, z_shp=[num_images, G.z_dim], in_gpu=in_gpu)\n",
    "\n",
    "def sample_GAN():\n",
    "    im = sampler()\n",
    "    im = im.numpy()\n",
    "    im = np.transpose(im, (1,2,0))\n",
    "    im = np.squeeze(im) #if single channel (yes), drop it.\n",
    "    # print(f\"sample_linearBP: im shape: {im.shape}; min: {np.min(im)}, max: {np.max(im)}.\")\n",
    "    im = to_uint8(im, ndim=2)\n",
    "    # print(f'1. uint image shape: {im.shape}')\n",
    "    return im\n",
    "\n",
    "\n",
    "title=\"Generate fake linear array images\"\n",
    "description=\"Generate fake linear array images.\"\n",
    "\n",
    "with gr.Blocks() as demo:\n",
    "    gr.Markdown(description)\n",
    "    with gr.Row():\n",
    "        with gr.Column(scale=2):\n",
    "            button_gen = gr.Button(\"Generate fake linear image\")\n",
    "        with gr.Column(scale=2):\n",
    "            output_im = gr.Image(type=\"numpy\", shape=(256, 256), image_mode=\"L\", label=\"fake image\", interactive=False) #grayscale image\n",
    "    button_gen.click(sample_GAN, inputs=None, outputs=output_im)\n",
    "    \n",
    "demo.launch(share=True, show_tips=True, enable_queue=True)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.0 ('torch')",
   "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": "3.9.0"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "d9a708e2b965f26293af085a2040b513f9ca1674b4051c9a67720a0b5ff04f14"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}