{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ea42a489",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-27 17:55:39.372674: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'libcudart.so.10.1'; dlerror: libcudart.so.10.1: cannot open shared object file: No such file or directory\n",
      "2022-06-27 17:55:39.372709: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /usr/local/lib/python3.7/dist-packages/tensorflow/python/compat/v2_compat.py:96: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "non-resource variables are not supported in the long term\n",
      "Imported model (for Places365, 128x128 images)\n"
     ]
    }
   ],
   "source": [
    "import tensorflow.compat.v1 as tf\n",
    "tf.disable_v2_behavior()\n",
    "tf.reset_default_graph()\n",
    "import numpy as np\n",
    "from PIL import Image\n",
    "import model\n",
    "import util\n",
    "import os\n",
    "import sys"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "24a28be9",
   "metadata": {},
   "outputs": [],
   "source": [
    "model_PATH='/home/jovyan/work/src/output/models/model2000.ckpt'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "2f04ed3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_demo_image(in_PATH):\n",
    "    img = np.array(Image.open(in_PATH).convert('RGB'))[np.newaxis] / 255.0\n",
    "    img_p = util.preprocess_images_outpainting(img)\n",
    "    return img_p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4cc717e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def inference(model_PATH, img_p):\n",
    "    G_Z = tf.placeholder(tf.float32, shape=[None, IMAGE_SZ, IMAGE_SZ, 4], name='G_Z')\n",
    "    G_sample = model.generator(G_Z)\n",
    "    \n",
    "    saver = tf.train.Saver()\n",
    "    with tf.Session() as sess:\n",
    "        saver.restore(sess, model_PATH)\n",
    "        output, = sess.run([G_sample], feed_dict={G_Z: img_p})\n",
    "        img_norm = (output[0] * 255.0).astype(np.uint8)\n",
    "        img = Image.fromarray(img_norm, 'RGB')\n",
    "        #util.save_image(output[0], out_PATH)\n",
    "    return img"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9ce2fc25",
   "metadata": {},
   "outputs": [],
   "source": [
    "def handle(conf):\n",
    "    \"\"\"\n",
    "    该方法是部署之后，其他人调用你的服务时候的处理方法。\n",
    "    请按规范填写参数结构，这样我们就能替你自动生成配置文件，方便其他人的调用。\n",
    "    范例：\n",
    "    params['key'] = value # value_type: str # description: some description\n",
    "    value_type 可以选择：img, video, audio, str, int, float, [int], [str], [float]\n",
    "    参数请放到params字典中，我们会自动解析该变量。\n",
    "    \"\"\"\n",
    "    base64_str = conf['Photo']\n",
    "    image = load_demo_image(base64_str, image_size, device)\n",
    "    res = inference(model, image)\n",
    "    # add your code\n",
    "    return {'Output': res}\n",
    "    "
   ]
  }
 ],
 "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": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}