Color NCA3D.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kXjfu4otMo7v"
      },
      "source": [
        "# Growing Neural Cellular Automata\n",
        "\n",
        "This notebook contains code to reproduce experiments and figures for the Modeling Morphogenisis in 3d: Neural cellular Automata"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "E8FzkEgQnrqE"
      },
      "outputs": [],
      "source": [
        "#@title Imports and Notebook Utilities\n",
        "import os\n",
        "import io\n",
        "import PIL.Image, PIL.ImageDraw\n",
        "import zipfile\n",
        "import numpy as np\n",
        "import matplotlib.pylab as pl\n",
        "\n",
        "import tensorflow as tf\n",
        "\n",
        "from IPython.display import Image, HTML, clear_output\n",
        "\n",
        "import csv\n",
        "import numpy as np\n",
        "from IPython.display import Image, HTML, clear_output\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def imwrite(f, a, fmt=None):\n",
        "  a = np.asarray(a)\n",
        "  if isinstance(f, str):\n",
        "    fmt = f.rsplit('.', 1)[-1].lower()\n",
        "    if fmt == 'jpg':\n",
        "      fmt = 'jpeg'\n",
        "    f = open(f, 'wb')\n",
        "  np2pil(a).save(f, fmt, quality=95)\n",
        "\n",
        "def np2pil(a):\n",
        "  if a.dtype in [np.float32, np.float64]:\n",
        "    a = np.uint8(np.clip(a, 0, 1)*255)\n",
        "  return PIL.Image.fromarray(a)\n",
        "\n",
        "\n",
        "def imencode(a, fmt='jpeg'):\n",
        "  a = np.asarray(a)\n",
        "  if len(a.shape) == 3 and a.shape[-1] == 4:\n",
        "    fmt = 'png'\n",
        "  f = io.BytesIO()\n",
        "  imwrite(f, a, fmt)\n",
        "  return f.getvalue()\n",
        "\n",
        "\n",
        "def imshow(a, fmt='jpeg'):\n",
        "  display(Image(data=imencode(a, fmt)))\n",
        "\n",
        "\n",
        "\n",
        "def zoom(img, scale=4):\n",
        "  img = np.repeat(img, scale, 0)\n",
        "  img = np.repeat(img, scale, 1)\n",
        "  return img\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hunrodUZ9EDk",
        "outputId": "9c152db0-76e1-4bbd-a160-3f191594911c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Data loading succesfull\n",
            "File name: torus_19x6x19\n",
            "Data shape: (1399, 6)\n"
          ]
        }
      ],
      "source": [
        "#@title Choose and load data from url\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import io\n",
        "import requests\n",
        "\n",
        "MODEL_3D = \"torus_19x6x19.csv\" #@param [\"torus_19x6x19.csv\", \"lizard_44x19x8.csv\", \"bird_15x28x28.csv\", \"spider_29x10x31.csv\", \"knight_18x8x15.csv\"]\n",
        "\n",
        "urls = {\n",
        "    'knight': 'https://raw.githubusercontent.com/Aadityaza/3d-Growing-neural-cellular-automata/main/Entity/knight_18x8x15.csv?token=GHSAT0AAAAAACCHMXX7O5NLF5CQ45A6YLQMZCUWWFA',\n",
        "    'bird': 'https://raw.githubusercontent.com/Aadityaza/3d-Growing-neural-cellular-automata/main/Entity/bird_15x28x28.csv',\n",
        "    'lizard': 'https://raw.githubusercontent.com/Aadityaza/3d-Growing-neural-cellular-automata/main/Entity/lizard_44x19x8.csv?token=GHSAT0AAAAAACCHMXX7CPE6MLJMAJJZLAQWZCUXGJQ',\n",
        "    'torus': 'https://raw.githubusercontent.com/Aadityaza/3d-Growing-neural-cellular-automata/main/Entity/torus_19x6x19.csv?token=GHSAT0AAAAAACCHMXX66DAJD3J6IIOICINOZCUXF4A',\n",
        "    'spider': 'https://raw.githubusercontent.com/Aadityaza/3d-Growing-neural-cellular-automata/main/Entity/spider_29x10x31.csv'\n",
        "}\n",
        "\n",
        "entity=MODEL_3D.split('_')[0]\n",
        "\n",
        "url=urls[entity]\n",
        "\n",
        "df = pd.read_csv(url,encoding= 'unicode_escape')\n",
        "\n",
        "arr = np.array(df)\n",
        "\n",
        "file_name = ((url.split('/')[-1]).split('?')[0]).split('.')[0]\n",
        "dimention = [int(x) for x in file_name.split(\"_\")[-1].split(\"x\")]\n",
        "\n",
        "max_dim=int(np.max(dimention))\n",
        "\n",
        "print('Data loading succesfull')\n",
        "print('File name:', file_name)\n",
        "print('Data shape:', arr.shape)\n",
        "SIZE_X=dimention[0]\n",
        "SIZE_Y=dimention[1]\n",
        "SIZE_Z=dimention[2]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "bgH7-YVcVj7_"
      },
      "outputs": [],
      "source": [
        "#@title load data from local file \n",
        "#@markdown Provide path to your csv file\n",
        "path = \"/content/lizard_24x19x8.csv\" #@param {type:\"string\"}\n",
        "file_name = path.split('/')[-1].split('.')[0]\n",
        "dimention = [int(x) for x in file_name.split(\"_\")[-1].split(\"x\")]\n",
        "max_dim=int(np.max(dimention))\n",
        "\n",
        "with open(path, 'r') as f:\n",
        "    reader = csv.reader(f)\n",
        "    data = list(reader)\n",
        "SIZE_X=dimention[0]\n",
        "SIZE_Y=dimention[1]\n",
        "SIZE_Z=dimention[2]\n",
        "\n",
        "arr = np.array(data).astype(float)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "_-qHKiOKBPSV"
      },
      "outputs": [],
      "source": [
        "#@title Define a function to returns a single layer of voxel \n",
        "def return_zLayer_t(arr,z_l):\n",
        "  z= np.array([arr[arr[:, 2] ==  z_l]])\n",
        "  Zxy=z[0][...,:6].astype(int)\n",
        "\n",
        "  Z=np.zeros((SIZE_X,SIZE_Y))\n",
        "  R=np.zeros((SIZE_X,SIZE_Y))\n",
        "  G=np.zeros((SIZE_X,SIZE_Y))\n",
        "  B=np.zeros((SIZE_X,SIZE_Y))\n",
        "\n",
        "  for i in Zxy:\n",
        "    Z[i[0]][i[1]]=1\n",
        "    R[i[0]][i[1]]=i[3]\n",
        "    G[i[0]][i[1]]=i[4]\n",
        "    B[i[0]][i[1]]=i[5]\n",
        "  return R,G,B,Z"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "nOP7pEs5m43W"
      },
      "outputs": [],
      "source": [
        "#@title Stack up layers \n",
        "r_stack,g_stack,b_stack,z_stack=return_zLayer_t(arr,0)\n",
        "for i in range(1,SIZE_Z):\n",
        "  r,g,b,a=return_zLayer_t(arr,i)\n",
        "  r_stack = np.dstack((r_stack,r))\n",
        "  g_stack = np.dstack((g_stack,g))\n",
        "  b_stack = np.dstack((b_stack,b))\n",
        "  z_stack = np.dstack((z_stack,a))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 803
        },
        "id": "DEJjVCoVCGCB",
        "outputId": "f8059586-f962-4c8b-f336-ee0d7c24744f"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAmklEQVR4nO3YzQmEMBRF4fjTpT24twLBpT1YjEWJbmVmYDiQZyKcb5dNCJeXi7FJlTnW9byvu3FsSp3ll7b0Ad7GwCADg8L74bOTokV3nhMGGRhkYFD2+/50Z/2Tu9OcMMjAIAOD+twbLvuee8uqOGGQgUEGBoW/JedhePS7bNo235I1MTDIwKCq/pen9N150Z1EOWGQgUEGBl2crRZKm+JIbAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 0\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAlUlEQVR4nO3YywmAMBBF0fjp0h7cW4Hg0h4sxqJEt+7kMkwgcs/eRB6Th7Erya59v7P3eBvmuctcv89c/I8MDDIwKHzea3dUVLTjnDDIwCADg/B5bq2zvtBOc8IgA4MMDBrpA9t5ZrxHM5wwyMAgA4PCd8l1mpr6LluOw7tkTQYGGRiU+v+7lPodF+2oL04YZGCQgUEPAcsWSiwzh2sAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 1\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAlUlEQVR4nO3YvQmAMBRF4fizpTvYO4FgmR0cxqFEe5vHISpEz1drCJeXi7FJhfacj9I13tSNY1PyfnvXRv7CwCADg/B5rq2zIrTTnDDIwCADg8Lz+7XOikSd5oRBBgYZGNRHDyzb9sY+quGEQQYGGRiE75LzMHzqu2xaV++STzIwyMCgov/bKdXXabSzrpwwyMAgA4NOygsWSryJBnkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 2\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAlUlEQVR4nO3YvQmAMBRF4fizpTvYO4Fg6Q4O41CifSweBxQ1nq8UA+HycjFWCdrmeadr3qzp+4q8X9+1kVIZGGRgUHh+S+usSNRpThhkYJCBQafz+rfOiuSd5oRBBgYZGNTmD6Z1fWIfn+GEQQYGGRgU3iXHrvvVd9mwLN4lr2RgkIFB6H92SuV1WtRZOScMMjDIwKADZtoWSgM+zwQAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 3\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAiElEQVR4nO3YvQmAMBQA4efPltnB3gkEy+zgMA4l2isYDkRR7ysTi3DEFK+KgiXntfTNlzRdV53t13cd5CsMBhkMOvyvf3uzSvZvmjcMMhhkMMhgkMEgg0EGg9r9wjjPT5zjNbxhkMEgg0EGgwwGGQwyGHQ6v46IGFL61XysnyZn+lcyGGQwaAPWMQ5MdTkFuAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 4\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAkklEQVR4nO3YQQ5AMBRFUcUuuwdzK5AY2oPFWJQwFQbNjabSuGcs/Lx8Lyo0CfuyHKlriG4YQs77lZ6vzfmwPzAwyMCgx/uauxNqd+80NwwyMMjAoGBnMW4YZGCQgUH9vG1fz1AVNwwyMMjAoMdZcorR77KLcV09S75hYJCBQcn/67k77d4Jb5Wezw2DDAwyMOgEl6YdaiJNW1kAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 5\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAnklEQVR4nO3XsQ3DIBRF0ZB4S3ZwnwkipfQOHoahLLsnBbrKBwrfU1Kgp6fPl50elWPbzvrsH691TZH3tfTO/4y8/A4sDLIwKEW/+ZbonTY6vxMGWRhkYdDwHVajO212XicMsjDIwqDlW8rsDMjsvE4YZGGQhUHpk/PQ75r3vof+S47O74RBFgZZGPSzT6J3QvTOaumd3wmDLAyyMOgCyX4o4oY6Tv4AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 6\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAn0lEQVR4nO3XsQ2DQBBEUZ9xl/RATgVIDumBYigKQX4E6Gvv1oH/C5GQdkfLyC6vyrGuZ/2MGKapRN5vrfU+79g4/8fAIAODSvQbf5Ldab338cIgA4MMDOreYbXWnZY9vxcGGRhkYNDnu++/niEke34vDDIwyMCgsoxj198x87al/pfsvY8XBhkYZGDQrV+iHZDdWU9a7+OFQQYGGRh0ASy+KOI8alpmAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 7\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAjUlEQVR4nO3YsQmAMBQAUaNu6Q72TiBYZgeHcSjRPhZyJFHBe6Ug5B/R4ocmscd4pM+IbhxDzvullZ6nzTvO/xgMMhgUcr/xO0//02rP4w2DDAYZDDIYZDDIYJDBoH7ZtrfPUFTtebxhkMEgg0EGgwwGGQwyGHTZVc3DkLVPmtb1Uzv90vN4wyCDQQaDTjTxG4ikR+WkAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 8\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAi0lEQVR4nO3YvQmAMBBAYaNu6Q72TiBYukOGyVCifbpHzh/wfaWgeI/TIqmrHPt+1teIYZ5Ty/3RWuep9ZEP+wODQQaDUvQ3Xnv6n3b3PG4YZDDIYJDBIINBBoMMBo1bKW+/Q6i753HDIINBBoMMBhkMMhhkMCit0xR6frTk/Kkz/ej53DDIYJDBoAvmMxjkllDczAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 9\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAjUlEQVR4nO3YsQmAMBQAUaNu6Q72TiBYZgeHcSjRPhZyJFHBe6Ug5B/R4ocmscd4pM+IbhxDzvullZ6nzTvO/xgMMhgUcr/xO0//02rP4w2DDAYZDDIYZDDIYJDBoH7ZtrfPUFTtebxhkMEgg0EGgwwGGQwyGHTZVc3DkLVPmtb1Uzv90vN4wyCDQQaDTjTxG4ikR+WkAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 10\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAn0lEQVR4nO3XsQ2DQBBEUZ9xl/RATgVIDumBYigKQX4E6Gvv1oH/C5GQdkfLyC6vyrGuZ/2MGKapRN5vrfU+79g4/8fAIAODSvQbf5Ldab338cIgA4MMDOreYbXWnZY9vxcGGRhkYNDnu++/niEke34vDDIwyMCgsoxj198x87al/pfsvY8XBhkYZGDQrV+iHZDdWU9a7+OFQQYGGRh0ASy+KOI8alpmAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 11\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAnklEQVR4nO3XsQ3DIBRF0ZB4S3ZwnwkipfQOHoahLLsnBbrKBwrfU1Kgp6fPl50elWPbzvrsH691TZH3tfTO/4y8/A4sDLIwKEW/+ZbonTY6vxMGWRhkYdDwHVajO212XicMsjDIwqDlW8rsDMjsvE4YZGGQhUHpk/PQ75r3vof+S47O74RBFgZZGPSzT6J3QvTOaumd3wmDLAyyMOgCyX4o4oY6Tv4AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 12\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAkklEQVR4nO3YQQ5AMBRFUcUuuwdzK5AY2oPFWJQwFQbNjabSuGcs/Lx8Lyo0CfuyHKlriG4YQs77lZ6vzfmwPzAwyMCgx/uauxNqd+80NwwyMMjAoGBnMW4YZGCQgUH9vG1fz1AVNwwyMMjAoMdZcorR77KLcV09S75hYJCBQcn/67k77d4Jb5Wezw2DDAwyMOgEl6YdaiJNW1kAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 13\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAiElEQVR4nO3YvQmAMBQA4efPltnB3gkEy+zgMA4l2isYDkRR7ysTi3DEFK+KgiXntfTNlzRdV53t13cd5CsMBhkMOvyvf3uzSvZvmjcMMhhkMMhgkMEgg0EGg9r9wjjPT5zjNbxhkMEgg0EGgwwGGQwyGHQ6v46IGFL61XysnyZn+lcyGGQwaAPWMQ5MdTkFuAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 14\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAlUlEQVR4nO3YvQmAMBRF4fizpTvYO4Fg6Q4O41CifSweBxQ1nq8UA+HycjFWCdrmeadr3qzp+4q8X9+1kVIZGGRgUHh+S+usSNRpThhkYJCBQafz+rfOiuSd5oRBBgYZGNTmD6Z1fWIfn+GEQQYGGRgU3iXHrvvVd9mwLN4lr2RgkIFB6H92SuV1WtRZOScMMjDIwKADZtoWSgM+zwQAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 15\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAlUlEQVR4nO3YvQmAMBRF4fizpTvYO4FgmR0cxqFEe5vHISpEz1drCJeXi7FJhfacj9I13tSNY1PyfnvXRv7CwCADg/B5rq2zIrTTnDDIwCADg8Lz+7XOikSd5oRBBgYZGNRHDyzb9sY+quGEQQYGGRiE75LzMHzqu2xaV++STzIwyMCgov/bKdXXabSzrpwwyMAgA4NOygsWSryJBnkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 16\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAlUlEQVR4nO3YywmAMBBF0fjp0h7cW4Hg0h4sxqJEt+7kMkwgcs/eRB6Th7Erya59v7P3eBvmuctcv89c/I8MDDIwKHzea3dUVLTjnDDIwCADg/B5bq2zvtBOc8IgA4MMDBrpA9t5ZrxHM5wwyMAgA4PCd8l1mpr6LluOw7tkTQYGGRiU+v+7lPodF+2oL04YZGCQgUEPAcsWSiwzh2sAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 17\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAEwAAAAYCAYAAABQiBvKAAAAmklEQVR4nO3YzQmEMBRF4fjTpT24twLBpT1YjEWJbmVmYDiQZyKcb5dNCJeXi7FJlTnW9byvu3FsSp3ll7b0Ad7GwCADg8L74bOTokV3nhMGGRhkYFD2+/50Z/2Tu9OcMMjAIAOD+twbLvuee8uqOGGQgUEGBoW/JedhePS7bNo235I1MTDIwKCq/pen9N150Z1EOWGQgUEGBl2crRZKm+JIbAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "layer: 18\n"
          ]
        }
      ],
      "source": [
        "#@title Display all layer\n",
        "for i in range(SIZE_X):\n",
        "  img= np.stack((r_stack[i]/255,g_stack[i]/255,b_stack[i]/255,z_stack[i]),axis=2)\n",
        "  imshow(zoom(img))\n",
        "  print('layer: '+str(i))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "p1EdfOHWSjSs"
      },
      "outputs": [],
      "source": [
        "#@title  Parameters\n",
        "#@markdown Choose for options below\n",
        "TRAINING_TYPE = \"structure with color\" #@param [\"structure only\", \"structure with color\"]\n",
        "flag = True if TRAINING_TYPE == \"structure with color\" else False\n",
        "\n",
        "FACE_COLOR='green'#@param['sienna','silver','dimgray','k','green']\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 425
        },
        "id": "9hY5EZLlcoNi",
        "outputId": "eafb192d-5f6c-4140-ea3e-959eb75c0a67"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(19, 6, 19, 4)\n"
          ]
        }
      ],
      "source": [
        "#@title Define target  { vertical-output: true}\n",
        "\n",
        "target=np.stack((r_stack/255,g_stack/255,b_stack/255,z_stack),axis=-1)\n",
        "\n",
        "z_resized = np.pad(z_stack, (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "r_resized = np.pad(r_stack, (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "g_resized = np.pad(g_stack, (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "b_resized = np.pad(b_stack, (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "target_resized=np.stack((r_resized/255,g_resized/255,b_resized/255,z_resized),axis=-1)\n",
        "\n",
        "fig = plt.figure()\n",
        "ax=fig.add_subplot(projection='3d')\n",
        "if flag:\n",
        "  ax.voxels(target_resized[...,3], facecolors=target_resized[..., :3] )\n",
        "else:\n",
        "  ax.voxels(target_resized[...,3], facecolors=FACE_COLOR)\n",
        "\n",
        "# Show the plot\n",
        "plt.show()\n",
        "print(target.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tM2t2wkLI3C7",
        "outputId": "9054b02e-473c-4a8c-e74d-7668e70af60d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Shape of seed:\n",
            "(1, 19, 6, 19, 16)\n"
          ]
        }
      ],
      "source": [
        "#@title Defining seed \n",
        "#@markdown Choose no of channel\n",
        "NUMBER_OF_CHANNELS = \"16\" #@param [16,20,24]\n",
        "Channel=int(NUMBER_OF_CHANNELS)\n",
        "seed = np.zeros([1,SIZE_X,SIZE_Y,SIZE_Z,Channel],np.float32)\n",
        "seed[:,SIZE_X//2,SIZE_Y//2,SIZE_Z//2,:4]=1\n",
        "print(\"Shape of seed:\")\n",
        "print(seed.shape)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 408
        },
        "id": "Q7LO9YEJiDYi",
        "outputId": "c38d972f-51fe-4afd-85e2-4267492acea4"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "#@title Display seed { vertical-output: true}\n",
        "temp = seed[0][...,:4]\n",
        "z_resized = np.pad(temp[...,0], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "\n",
        "\n",
        "fig = plt.figure()\n",
        "ax=fig.add_subplot(projection='3d')\n",
        "if flag:\n",
        "  r_resized = np.pad(temp[...,1], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "  g_resized = np.pad(temp[...,2], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "  b_resized = np.pad(temp[...,3], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "\n",
        "  obj=np.stack((r_resized/255,g_resized/255,b_resized/255,z_resized),axis=-1)\n",
        "  ax.voxels(obj[...,3], facecolors=obj[..., :3] )\n",
        "else:\n",
        "  ax.voxels(z_resized, facecolors=FACE_COLOR )\n",
        "\n",
        "# Show the plot\n",
        "plt.show()\n",
        "plt.close()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7qM_zpV9F8E3"
      },
      "source": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "WJe4gjbq5spG"
      },
      "outputs": [],
      "source": [
        "#@title Building Model  Conv3D multi channel\n",
        "\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from keras.layers import Conv3D\n",
        "from keras.models import Sequential, load_model\n",
        "\n",
        "def get_living_mask(x):\n",
        "  alpha = x[..., 3:4]\n",
        "  return  tf.cast(tf.nn.max_pool3d(alpha,3,1,'SAME') > 0.1,tf.float32)\n",
        "\n",
        "class CA(tf.Module):\n",
        "  def __init__(self):\n",
        "    self.model=tf.keras.Sequential([\n",
        "        Conv3D(filters=Channel*3, kernel_size=3, padding='same',input_shape=(SIZE_X,SIZE_Y,SIZE_Z,Channel), activation=tf.nn.relu),\n",
        "        Conv3D(filters=Channel, kernel_size=3, padding='same',kernel_initializer=tf.zeros),\n",
        "    ])\n",
        "  \n",
        "  @tf.function\n",
        "  def __call__(self,x):\n",
        "    #alive_mask= get_living_mask(x)\n",
        "    update_mask = tf.floor(tf.random.uniform(x.shape) + 0.5)\n",
        "    x= x+self.model(x)*update_mask\n",
        "    #x *= alive_mask\n",
        "    return x\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "N6xWnDdnwnEa"
      },
      "outputs": [],
      "source": [
        "#@title Running randomly initialized CA for some ittirations{vertical-output:true}\n",
        "ca=CA()\n",
        "x=seed\n",
        "\n",
        "ittiration=100\n",
        "for i in range(ittiration):\n",
        "  x=ca(x)\n",
        "  "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_q-Tk85aaRmJ",
        "outputId": "595ee9a9-43ff-4c0f-f94b-1ceeed8e11ee"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv3d (Conv3D)             (None, 19, 6, 19, 48)     20784     \n",
            "                                                                 \n",
            " conv3d_1 (Conv3D)           (None, 19, 6, 19, 16)     20752     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 41,536\n",
            "Trainable params: 41,536\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "#@title Model summary S\n",
        "ca.model.summary()\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "51d23it1jGTc"
      },
      "outputs": [],
      "source": [
        "#@title Test / Visualize loss minimization (dont run)\n",
        "\n",
        "trainer=tf.optimizers.Adam(1e-3)\n",
        "\n",
        "ittiration =20\n",
        "\n",
        "for k in range(ittiration):\n",
        "  with tf.GradientTape() as g:\n",
        "    x=seed    \n",
        "    print(\"Target \")   \n",
        "    imshow(zoom(target[7][...,3]))\n",
        "    print(\"NCA in ittiration:\"+str(k))     \n",
        "    \n",
        "    for i in range(100): \n",
        "      x=ca(x)\n",
        "    target_loss = tf.reduce_mean(tf.square(x[...,:4] - tf.cast(target, tf.float32)))\n",
        "    diff_loss=0\n",
        "    for i in tf.range(50):\n",
        "      prev_x=x\n",
        "      x=ca(x)\n",
        "      diff_loss+=tf.reduce_mean(tf.square(x-prev_x))\n",
        "    loss=target_loss+diff_loss \n",
        "    if i %20==0:\n",
        "      imshow(zoom(x[0][...,3][7]))\n",
        "    clear_output(True)\n",
        "      \n",
        "  #params=ca.trainable_variables\n",
        "  #grads=g.gradient(loss, params) \n",
        "  #grads=[g/(tf.norm(g)+1e-8) for g in grads]\n",
        "  #trainer.apply_gradients(zip(grads,params))\n",
        "  #print(loss.numpy())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "A3Z-sxBHwqcX"
      },
      "outputs": [],
      "source": [
        "#@title Defining loss & train step\n",
        "BATCH_SIZE = 1 #@param {type:\"slider\", min:1, max:16, step:1}\n",
        "#defining optimizer\n",
        "lr=tf.keras.optimizers.schedules.PiecewiseConstantDecay([1000],[1e-3,3e-4])\n",
        "trainer=tf.optimizers.Adam(lr)\n",
        "loss_log=[]\n",
        "mse_loss = tf.keras.losses.MeanSquaredError()\n",
        "@tf.function\n",
        "def training_step():\n",
        "  with tf.GradientTape() as g:\n",
        "    x=tf.repeat(seed,BATCH_SIZE,0)\n",
        "    N=\"100\"#@param[50,100,200]\n",
        "    ittiration =int(N)\n",
        "    for i in range(ittiration):\n",
        "      x=ca(x)\n",
        "      \n",
        "      if flag:\n",
        "        loss = mse_loss(x[..., :4], tf.cast(target, tf.float32))\n",
        "      else:\n",
        "        loss = mse_loss(x[..., :4][...,3], tf.cast(target[...,3], tf.float32))  \n",
        "\n",
        "      \n",
        "  params=ca.trainable_variables\n",
        "  grads=g.gradient(loss, params) \n",
        "  grads=[g/(tf.norm(g)+1e-8) for g in grads]\n",
        "  trainer.apply_gradients(zip(grads,params))\n",
        "  return loss,x"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 58
        },
        "id": "zTBOczllbcq8",
        "outputId": "1f414274-2b58-40c0-fe25-3f36e451325c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Target :\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOQAAAAYCAYAAADqB1HvAAAA8ElEQVR4nO3ZQW7CMBAFUEK5JXdg3xNUYpk75DAcqirb1osUxDj+Qe/t2BiPra/JyNMhzPc8//z+/XG5TKP28oh2v89Kq693Pc5r3bFyMeA1AglBBBKCdP8er/7G/k/vGaN3PVvPSFvfT7V3Oy8dEoIIJAQRSAhyql5w9Eyyt3fMVu/9j74f1umQEEQgIYhAQpDyGfJ6u1UvGWXv9e19/6P1Pj8dEoIIJAQRSAhSPkN+Lsufd7Ov83nTd6/2//em9/5H3w/rdEgIIpAQRCAhSNy81c406TPhqzNYWn3V9VTPqO92Xi0dEoIIJAQRSAhyB8jeVugbackMAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "#@title Refrance layer from target to compare training progerss {vertical-output:true}\n",
        "print(\"Target :\")\n",
        "\n",
        "if flag:\n",
        "  imshow(zoom(np.hstack((target[0, ..., :4],target[7, ..., :4],target[9, ..., :4]))))\n",
        "  \n",
        "else:\n",
        "  imshow(zoom(np.hstack((target[0, ..., 3],target[7, ..., 3],target[9, ..., 3])))) \n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 506
        },
        "id": "A0G2uPycxJKc",
        "outputId": "f123b47b-8b2f-410e-e161-b831c19d3325"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Batch : 0 refrence layer 0,7,9\n"
          ]
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjMAAAGdCAYAAADnrPLBAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACI+ElEQVR4nOz9eZCd933fe76f/TnPWXvfsLFBcIcIiZJgSZQlWRzLUiLJcuzxveVyFHnsmluXqUoN/0hZf8SZ1KTK/0xcqprCXN2bXJdvMjU3yq1MlMRSGDm0LJoiDW6GBIIEsTTWXk+f/Zxn/z3P/HG6WwAJkAAJEujG9+ViUd1odD8Hgut89Pt9Fy3P8xwhhBBCiG1Kv90PIIQQQgjxfkiYEUIIIcS2JmFGCCGEENuahBkhhBBCbGsSZoQQQgixrUmYEUIIIcS2JmFGCCGEENuahBkhhBBCbGvm7X6AWy3LMpaWliiXy2iadrsfRwghhBA3IM9zer0es7Oz6PrNnbXsuDCztLTE7t27b/djCCGEEOI9uHTpErt27bqp37Pjwky5XAaGfxiVSuU2P40QQgghbkS322X37t1b7+M3Y8eFmc2rpUqlImFGCCGE2GbeS4nIHVkA/Od//ufcf//9HDhwgH/1r/7V7X4cIYQQQtzB7riTmTRNeeqpp/jxj39MtVrlscce4xvf+AZjY2O3+9GEEEIIcQe6405mXnzxRR5++GHm5uYolUp8+ctf5kc/+tHtfiwhhBBC3KFueZh59tln+epXv8rs7CyapvH973//bV9z5MgR9u3bh+u6HD58mBdffHHr15aWlpibm9v6eG5ujsXFxVv9mEIIIYTYIW55mBkMBjz66KMcOXLkmr/+ve99j6eeeop/+k//Ka+++iqPPvooX/rSl1hbW3tPPy+KIrrd7lX/CCGEEOLuccvDzJe//GX++T//53zjG9+45q//yZ/8CX/wB3/At771LR566CG++93v4nkef/qnfwrA7OzsVScxi4uLzM7OXvfn/fEf/zHVanXrH5kxI4QQQtxdPtSamTiOeeWVV3jiiSd+8QC6zhNPPMELL7wAwCc/+Ulee+01FhcX6ff7/Jf/8l/40pe+dN3v+e1vf5tOp7P1z6VLlz7w1yGEEEKIO8eH2s20vr6OUoqpqamrPj81NcXJkyeHD2Sa/It/8S/4whe+QJZl/ON//I/fsZPJcRwcx/lAn1sIIYQQd647rjUb4Gtf+xpf+9rXbvdjCCGEEGIb+FCvmcbHxzEMg9XV1as+v7q6yvT09If5KEIIIYTYIT7UMGPbNo899hjPPPPM1ueyLOOZZ57hU5/61If5KO9ZmCiag5gwUbf7UYQQQgjBB3DN1O/3OXPmzNbH586d49ixY4yOjrJnzx6eeuopvvnNb/Lxj3+cT37yk3znO99hMBjwrW9961Y/yi232A44utCgH6aUXJPD82PM1Qq3+7GEEEKIu9otDzMvv/wyX/jCF7Y+fuqppwD45je/yZ/92Z/x27/929Trdf7oj/6IlZUVDh06xNNPP/22ouCbdeTIEY4cOYJSH8yJSZgoji406PgJE2WHei/i6EKDrxycwbWMD+RnCiGEEOLdaXme57f7IW6lbrdLtVql0+nc0q3ZzUHMn/9sidGijWsZW9dNX310lpGifct+jhBCCHE3ej/v33fcbqY7lWcblFyTei8iTBT1XkTJNSnYciojhBBC3E4SZm6Qaxkcnh+j6lk0BzFVz+Lw/JhcMQkhhBC32R05Z+ZONVcr8JWDMwSxomAbEmSEEEKIO4CEmZvkWhJihBBCiDvJjrlmOnLkCA899BCf+MQnbvejCCGEEOJDJN1MQgghhLjtpJtJCCGEEHctCTNCCCGE2NYkzAghhBBiW5MwI4QQQohtTcKMEEIIIbY1CTNCCCGE2NZ2TJiROTNCCCHE3UnmzAghhBDitpM5M0IIIYS4a0mYEUIIIcS2JmFGCCGEENuahBkhhBBCbGsSZoQQQgixrUmYEUIIIcS2JmFGCCGEENvajgkzMjRPCCGEuDvJ0DwhhBBC3HYyNE8IIYQQdy0JM0IIIYTY1iTMCCGEEGJbkzAjhBBCiG1NwowQQgghtjUJM0IIIYTY1iTMCCGEEGJbkzAjhBBCiG1NwowQQgghtrUdE2Y+jHUGYaJoDmLCRH1gP0MIIYQQN0fWGdygxXbA0YUG/TCl5Jocnh9jrla4Zd9fCCGEuJvJOoMPWJgoji406PgJo0Wbjp9wdKEhJzRCCCHEHUDCzA3wY0U/TJkoO7iWwUTZoR+mBLGEGSGEEOJ2kzBzAzzboOSa1HsRYaKo9yJKrknBNm73owkhhBB3PQkzN8C1DA7Pj1H1LJqDmKpncXh+DNeSMCOEEELcbubtfoDtYq5W4CsHZwhiRcE2JMgIIYQQdwgJMzfBtSTECCGEEHcauWYSQgghxLYmYUYIIYQQ25qEGSGEEEJsaxJmhBBCCLGtSZgRQgghxLa2Y8LMh7FoUgghhBB3Hlk0KYQQQojbThZNCiGEEOKuJWFGCCGEENuahBkhhBBCbGsSZoQQQgixrcluphsUJoqWH0MOI0VbdjQJIYQQdwgJMzdgsR3w9PFlTix1AXh4rsqvPTLNXK1wm59MCCGEEHLN9C7CRPHc6TrHFzvoGmgaHL/c5rnTdcJE3e7HE0IIIe56EmbehR8rmv0Yy9AYLTqMlxwsQ6c1iAliCTNCCCHE7SZh5l14tsFoySZROc1BxHo/IlEZI0Wbgi11M0IIIcTtJmHmXbiWweMHJjg4VyXLIc/h4K4ajx+YkCJgIYQQ4g4gBcA3YK5W4Hd+aS9tPyYHRjzpZhJCCCHuFBJmbpBrGUxXpXtJCCGEuNPINZMQQgghtjUJM0IIIYTY1nZMmDly5AgPPfQQn/jEJ273owghhBDiQ6TleZ7f7oe4lbrdLtVqlU6nQ6VSud2PI4QQQogb8H7ev3fMyYwQQggh7k4SZoQQQgixrUmYEUIIIcS2JmFGCCGEENuahBkhhBBCbGsSZoQQQgixrUmYuUlhomgOYsJE3e5HEUIIIQSym+mmLLYDji406IcpJdfk8PwYczXZ1ySEEELcTnIyc4PCRHF0oUHHTxgt2nT8hKMLDTmhEUIIIW4zCTM3yI8V/TBlouzgWgYTZYd+mBLEEmaEEEKI20nCzA3ybIOSa1LvRYSJot6LKLkmBdu43Y8mhBBC3NUkzNwg1zI4PD9G1bNoDmKqnsXh+TFcS8KMEEIIcTtJAfBNmKsV+MrBGYJYUbCNDzzIhInCjxXeh/CzhBBCiO1KwsxNcq0PJ1hI55QQQghxY+Sa6Q4knVNCCCHEjZMwcweSzikhhBDixkmYuQNJ55QQQghx4yTM3IGkc0oIIYS4cVIAfIf6sDunhBBCiO1Kwswd7MPqnBJCCCG2M7lmEkIIIcS2JmFGCCGEENvajgkzR44c4aGHHuITn/jE7X6U9yVMFM1BLDNlhBBCiBuk5Xme3+6HuJW63S7VapVOp0OlUrndj3NTZOqvEEKIu9X7ef/eMScz251M/RVCCCHeGwkzdwiZ+iuEEEK8NxJm7hAy9VcIIYR4byTM3CFk6q8QQgjx3sjQvPcgTBR+rPBu8WRemforhBBC3DwJMzfpg+44kqm/QgghxM2Ra6ab8GF3HMnMGSGEEOLdycnMTbhWx1FzEBPE6oZOU668ntr8fte7qpKZM0IIIcSNkTBzE67sOJooO9R7EVXPuqGOoyvDSaIy0DQsXbtmULnyBGjz5xxdaPCVgzNyBSWEEEK8hVwz3YT32nF0ZTgpuybHFzscv9ym5JrXvKqSmTNCCCHEjZOTmZv0XjqONsNJzbPoRymGpqHrGrqmXfOq6v2cAAkhhBB3Gwkz79HNLLTybINEZfzkVB1dg/MNn/GSTZbn1HvxNYPK/dNlXlvsyMwZIYQQ4l1ImLlJm7UvrUGMZeo8fmCc+fHSu/9GTQPA0HXGSzYF26Tlx4x49lVB5craGsfS+fg9o8yPFyXICCGEENchYeYmbNa+XGz4NAcR672Yk8tdfv+z88xPXD/Q+LHC0jV++b4JdE0jy3Nag5hfuX+SmVphK6hcq/D35HKX+fHih/UShRBCiG1HCoBvgh8rWoOY5iAiVjnTNZeWn/D8mfV3nAWzWQPT8RNMXaPjJ4wU7auCzOb3l8JfIYQQ4uZImLkJnm1gmTrrvRjPNvBjxUTZJlLZOwaOG+2CkmWTQgghxM2Ta6ab4FoGjx8Y5+Ryl+VOyETZZsRzGPHsdw0cN9oFJYW/QgghxM2RMHOT5sdL/P5n53n+zDqRyt5WwPtO3mnvkhT+CiGEEO+NhJn3YH6ixFjRpjGIGSvaVD37fX0/KfwVQggh3jsJM+/Brd6b9NbC35pnsdIJafsx01XZxySEEEK8EykAvkkfxObsKwt/l9oBPzlV5/Ran5+cqrPYDm7h0wshhBA7j4SZm/RBtE9vdjt5jsGxS20ADu2p4UfqfQclIYQQYqeTa6ab9EHtTZqrFfjlAxO0BjGTFZeKaxEm6m17m4QQQghxNTmZuUk3OjNmM4jczKnKSNFmsuLS8ROZMyOEEELcIDmZeQ82Z8a0/Zg8H4aQK73XAuHNoHR0oSFzZoQQQogbJGHmPWoM4msGls0C4UYvolywaPQiji40+MrBmRsKJTc6XE8IIYQQQxJm3oNrzYXZDCx+rLjc9FnvR6QNMHUIEnVTdS9XDtcLE4UfKzwJNkIIIcQ1SZh5DzY7mqqeRZLlVD1rq6PJ0GClG9IYxMzVCiy2A9IsR9du/ufc6nk2QgghxE4kBcDvgWcbJFnOs6fqPH9mnWdP1UmynIJtoHKYrhaYqjiEScZUxWG6WiDLr//9rlUs/EHMsxFCCCF2IjmZea/yzXSSX/WxoUHZNUmUxWTFJYgV4yXnuh1J1zt9aQ1i1rohkxV3a56NtGkLIYQQbydh5j3wY4Vl6HzuvgmCWNENE1KVsVDv88LZBi+db27MoXH5xD2j1+1I2jx9We9HVFyL9f6w9ubQ7hovnW9yarXPyZUeH9s7glL5LZlnI4QQQuw0Embeg83BeccXO7y50qU1SCi5Jo1BzFovouSYuJZBnGY4psZY8dqLKP1YcbkVsN4PSRWYBvTDFD9KSVTOod01Xr3Y4tjFNofnrx+KhBBCiLvZHVkz841vfIORkRF+8zd/83Y/yjW5lsFDsxVeX+7SGCSMFC1SlfHc6TVUljFadJiquBQd8x1XHRgarHQCVrsRrqWz2o243PIZxMN1CbO1Ap+7b4L7pkp87r4JKf4VQgghruGODDP/6B/9I/71v/7Xt/sx3pFl6NQKFvdPlciBVpCw2A45udzjUmvYmp2ojJJrEibqmoW7KofpistUeaNYuOywq1agaA/XJYSJou0nTFZcat61T3eEEEKIu90dGWY+//nPUy6Xb/djvKPxok21YPHmap/FVoAfKYqOia5rrHQD0ixjz1iRKM35i9dX+eHxZRbbwVWdS55tsGvUY/eIx0d2Vdk94rFvosSn7x1/13UJQgghhBi66TDz7LPP8tWvfpXZ2Vk0TeP73//+277myJEj7Nu3D9d1OXz4MC+++OKteNY7StWz+cS+MQZRuhFO0o1aGZ2Hpsv8/mfmmau56LDVWv308WX+47FF/vxnS/zw+DKNQczh+THGyg5xmjFWdjg8P8b8RImvHJzhq4/O8pWDM3K9JIQQQryDmy4AHgwGPProo/ze7/0ev/Ebv/G2X//e977HU089xXe/+10OHz7Md77zHb70pS/x5ptvMjk5CcChQ4dI0/Rtv/dHP/oRs7Oz7+FlfPjCRKHyjAemyryx0iVKMlp+QjdMcC0DP0mJkoyJsoNrGdQ8i78+vc7+iRJ7x7yrpgZfa33BlVOAhRBCCHF9Nx1mvvzlL/PlL3/5ur/+J3/yJ/zBH/wB3/rWtwD47ne/yw9+8AP+9E//lD/8wz8E4NixY+/taa8hiiKiKNr6uNvt3rLv/U78WBElGR/bO0J9ELLYDklSxUzFwzZ0Xl/qYps65xs+szWX5XYAwEzt7XNjNhdV+huFwtfawC0rDYQQQohru6Wt2XEc88orr/Dtb39763O6rvPEE0/wwgsv3MofteWP//iP+Wf/7J99IN/7nWy2Z0eJYv94iVY/wXBMNE0jy3MutnzGPJuFdZ+Fep/7p8s8PFel4yfYhk69F23NjXmntQUL9T4/PbNOrDJGPFtWGgghhBBvcUsLgNfX11FKMTU1ddXnp6amWFlZueHv88QTT/Bbv/Vb/PCHP2TXrl3vGIS+/e1v0+l0tv65dOnSe37+m+Faxla9y1jJRtNAAyqehYbGG8s9dE3nswfG2T9RpOgYfGx3Dc8xrirsBa67tmBhvc+/+usFnj/b4EJjwIWGLysNhBBCiLe4I4fm/bf/9t9u+Gsdx8FxnA/waa5vrlbg0O4a9W7I7jGPKMmoFmwcU2O1G1PxTMquRdlNefVCm36YMlK0+fg9o8yPF3GtYbDphyk1zyJRGTXPohemtP2Y506v0x4kzNRc/FjR8iMqBVNWGgghhBBXuKUnM+Pj4xiGwerq6lWfX11dZXp6+lb+qDtCmCiOXWqjaxr3jBWpuhZ5njMIFUGccnShyYXmgJfONwkSxahn40eKk8u/qOvxbINEZfzkVJ3nz67zk1N1EpWR55CkGeNle6tept6LcQxdVhoIIYQQV7ilYca2bR577DGeeeaZrc9lWcYzzzzDpz71qVv5o+4IfqzohykztQIPz1aZrDhcbPpEacbH9lYB+OnpdRbbAboGb671MAzt7VOBNW3zP2x97NkGI0Wb0aKDbWistENGPItP3zsupzJCCCHEFW76mqnf73PmzJmtj8+dO8exY8cYHR1lz549PPXUU3zzm9/k4x//OJ/85Cf5zne+w2Aw2Opu2kk2i4CHSyUdXNNApTmdIOa1RcVY0UFlORMlh4Jl0PYTGv0Wn9o/tnW64scKS9f45fsm0DeKh/thSpazVVNTdk0OTJX5zIFx5sdLt/MlCyGEEHecmw4zL7/8Ml/4whe2Pn7qqacA+OY3v8mf/dmf8du//dvU63X+6I/+iJWVFQ4dOsTTTz/9tqLgW+3IkSMcOXIEpT684tjNIuCjCw1WuyGdMEE3NFSWo7KcM2s9PMfk0/vHWOvFdMOYXNM4OFfdOl3ZDEQdP2Gi7FDvxVtdTiNF+5ozaIQQQgjxC1qe5/ntfohbqdvtUq1W6XQ6VCqVD+VnholiqR3wg+NLnFzu4ceKOFUE8XDFwcf3jTJTdbnY8JmuufzmY7uvCibv1JothBBC3A3ez/v3HdnNtN24lsFsrcBk2eXs2oBqQacXgB8pXNvg9GqPN5a7eLbJSMmhMYivCitztYKcwAghhBDv0R25aHI7ci2Dxw9McHCuSp5DO0iYqLp8fO8IicrIyPmle0bR4ZqzYlxreK10M0HmyqWVQgghxN1KTmZuoblagd/5pb18Zr3PX7yxhmcZLNQHrHRCdF2jHSbMVgtbawxuNrhcudJg82qqNYixTJ3HpThYCCHEXUrCzC3mWgb7xktMlju8cLZBEKd0w4QoyfjxyTU+tneE/ROl686KudYeprfW1BzaU+PYxTYXGz7NQcR6L+bkcpff/+w88xMSaIQQQtxdJMx8AFzL4JG5Ks+fadDyE2qejWXqtIOE1XbI3/vYrmueylyrEHisaG+tOxh2O0U8d3qdIE5pDiJilTNdc1nuhDx/Zp3ZWkFqboQQQtxVdkyYuR2t2e9k/0SJR3dX6UcpcyMuQZyhaTn7J0p49nCNgXfFrBlD422h5ehCg0/Nj71t3UFzEEMO672Y6Y1VBxNlm0hlsupACCHEXWfHhJknn3ySJ598cqu163ZzLYPPPzDJmbU+9V7MRNlmxHMwdZ2fnKoTJRl+kpKoDM82sQydtp9s7WyaKDs0BzGaxta6A8vQSFTOwbkqn7t/kgtNn+VOuPW9RzxbVh0IIYS46+yYMHMnmh8v8fufnef5M+tEKqPkmPQjhR8psjznRydWSVXGI3NVRjyblh/jWjqz1QL1XjQcnmcZW+sOVAZRqkiznH3jxau+94hnc3h+TE5lhBBC3HUkzHzA5idKzNYKBLGiE8T8558v45pwYrFHunEqE8QZppFQ8yyyHFZ7ISOezaE9NdYHMTA8jTmz1idKNV5f6rJQ7/PQbHXre8t8GiGEEHcrCTMfAtcyaAxi/utrK/zoxArrvYicHE3TmCxBuWBwsREwXrSpFixsw2C8ZPPT0+v4ccpCfcBSJySMUxKVo+saz56uMz9RwrUkxAghhLi7SZj5EISJ4rnTdV5b6qLBcJlklKJpGrahs7A2IM1zpmouczWPk8tdfnB8mUrBZLrikgOXmwPKrsVYyca1TE6t9Gn7MdNVWXsghBDi7iZh5kPgx4pmP0bTcoqWyYhnY+galqFjGBqmrvHwTJUHpysYusZaL6QTDIuB4zRHZRlFx2Sm6rJr1KPtJ+Q57KilWkIIIcR7JOsMPgSebTBasslzjSAdDtFzLYOKazFaGAabRi/i9FqfTpDQ8VNGPYtYZXi2QdtPmaq4mLpOa5Dgxyn3T5cZ8WxZaSCEEOKut2NOZu60OTNX2tzb1A9TwkTR9hNMQ8c0dSxTZ6rqMlK0qfdCPFtnrGwzaxVQKuPc2oBAKR6cKVPvRnSCmJJrYho6J5a6XGgMZNu2EEKIu5qW5/mOuq14PyvEP2jDIBNzoTngL1+v8/xCnYJlsmesyP6JIjrwqw9PkwPHLrWpd0NOrfaZHnG5d7zEX55cQ2U5n79vgkY/4nR9wL0TJfaMeVut3F85OCMFwUIIIbad9/P+vWNOZrYD1zKYrhaoeTanV/vsbhVJs4zz631OrXR5cKZCzi/auZfaAaa5xlTZJVEZrqUDGoNYcabe58RSF02DsbKzNWRPJgALIYS420jNzG3gx4osg8f2jhAlOanKcC2Dmmdx7FKbMBkGktlagRHPpt6LAEhUTpgojl1q8cZKj0RlnFzpcvRcg4vNAWiga7f5xQkhhBAfMgkzt4FnG5RckzBRjJdsdo94PDhT4f7pCv0wJYiHdT+uZXB4foyqZ9ELUw7OVZmfKHJ2rY+GxgMzZQqWwRuLXU6t9Wn7Cc+cXGNhvS9FwUIIIe4acs10G2yGlOdO1zlbH2AYOvsnS3T8ZLjC4Ir9SnO1Al85OLM15ffp44u0gpgsA5VljJUcEitj90iB+fEiZ9b6vHy+yf7J0taKAykKFkIIsZPJycxtMlcr8PVDc/z+Z+/hU/vHIIeqZ11zv5JrGYwUbc43Bvx/X7xMkuakWU5zELFQ7zPq2dwzXsLQNZqDiJafULJNOn7C0YWGnNAIIYTY0eRk5jZyLYOHZqvMT5Tedb9SmCh+fHKNXpCyZ9RjrRcyiBVanpGjcXKly3TF5VLTZ6Rk41oGRceUomAhhBA7noSZO8CN7FfyY0XPj0mzjKVOjMqHaxGmykXyPOcnb9ZJVYYfK0qORZRm7Bstsn+ydNW1lRBCCLHT7JgwcycPzbsVDA3aYcpI0SYD1johrq1TK5o0BhFLbZ9Y5Ri6RpAoOgsRy+2AT94zKqcyQgghdjQZmrdNNAcx/5+/ucB6P6QXKM43e5hoxFnOUiugEyZk2fBrXVvHMXVqBYeP7q7xP/7KvcxPlG7vCxBCCCHegQzNuwt4tsGukQKupVOdsZgdcVls+ryx0iFWGaahkZKTKoiTDA2NSsGkFcb85M01ChtFxHJKI4QQYqeRbqZtYrOde7zkEKUZ+ydK/A9fuJfP3TdJ0TGxDQNDB02DNAdT14jSjM4g5oWFBv/x2CI/PL7MYju43S9FCCGEuKXkZGYbeevMGdcy+O8O72WpE/LmSg8/MhjEiijJMAwNzzLoBAmxyqkUftGq/cUHJlH58LRHTmqEEEJsd1IzswMs1Ps888YqL51vEqsMlWX0AkWaZfTCFNc0+MjuX7SAT5YdshzZtC2EEOKOITUzd7n5iRIFy0BlOSNFm1MrfdZ6IadXuwRJRqwyXjzX5GeX21Rdi4/sqvHwXHXrpEY2bQshhNjOpGZmhxgp2kxWXPxIsXfMo+3HrPVidC0nzSBIMvpBSmuQcHyxw88utTEM7apdUEIIIcR2JGFmh7hyKWWiMmqexVjJZqpSwLN0XFPDsjSKrkGcZjT7Ma9eaOFYugzVE0IIsa3JNdMOslkgvNQOSLMMyzC43PQxDY1+oBgvO0xXXMIkwzKHOfbgXFWumIQQQmxrOybM7PQJwDfKtQxmawUmyi69IEUDDB1GPZtKwULXdT59oEaawnjZkWF6Qgghtj3pZtqhXrnQ4t+/conmIKZoG3zm3gmW2j6n630MTefhuSpfeGCCWsGWFm0hhBC3nXQziauEieJCY8C+8SKP7qrRGES8fKHJvvEin79vkqV2QKIUL51rEiWZtGgLIYTY1qQAeAfyY0U/TJmtFhgrOYwWHTpBQtW1KLsWs7UCbyx1udj0Kbm/GKYXJnf3FZ0QQojtSU5mdiDPNii5JvVeRM2zWO0ElByTTphQdExOLvdYbAcYhkaYKPaOFbdatOW6SQghxHYjJzM70Gabdpbn/PXpdRbWfSoFiyyHxZbPSjekVrTQ0Wj7ibRoCyGE2NbkZGaHGivalFyTiZJDL0pYbAYstwNmawWiVFGyTaI0I89zDF3fatEOE4UfKykKFkIIsW1ImNmhNutmgiQly2G8YnPsUptU5TiWzuvLPZJE4dom94x5REnGQr3PsUtt+mEqRcFCCCG2Dblm2qE828AyddZ7MZ5t0AuGCycdWyfLIUkVcZajAxfbAf/rTxf4n39ylgsNn9GiLUXBQgghtg0JMzuUaxk8fmCcWtFiuRPiWDqebRLEGVmeU3ZN9o56jBZtTE0jTjKWOyH1Xoiua0yUHdnbJIQQYluQMLODzY+X+P3PzvPwTJkgUbiWjm3qxGmGaRgUbJNelJJmOY1BRCuIOXapzZvLHZbbASXXlKJgIYQQdzypmdnhZmsFJioumqYzU3Np9CL6cUqqci40ByQqozGI0TVIs5wkzXj6xCqHdtf42L5RKQIWQghxx5OTmR3OjxVRkrF3zKNgGag85/z6gIKl85G5Kg9Ml4f7mzQADccysE2dsmtyZq0nNTNCCCHueDvmZEYWTV7b5gC9M2t96r2Q1xY7GIbOR3ZXWevEnF7rYegamq5R0CFKM7pBwko3pDlIODhX5aHZ6u1+GUIIIcR17ZiTmSeffJLXX3+dl1566XY/yh3FtQwO7amx1gtZ6UY4lsFU2eFi06cdxGRApWATp4p6L6TtJ6RZTpwOC4WPL3bkdEYIIcQdbceczIjrqxVs9k+WeGimwum1PoMopTVIGESKeydKREnGIEpwzWx4xeSYhEnG4fkKUZLJmgMhhBB3tB1zMiOuz7MNRjybROXsnyiRZjlZBoYO/ThF18EydGZHCsxUC5iGhmtohPFwwF6YKDmdEUIIcceSMHMX2NzVVPUsAB6YKbN/sghAs58QpzmJyvCTDEOHRj/m5Fqf1W5Ay0/4i9dX+eHxZRbbwe18GUIIIcQ1SZi5S8zVCnzl4Ay/+vAUI56NbehUCxZ7RwpUChb3TpQgz4lVzlzNxTI0Xr3Q5tWLTZJcyURgIYQQdywJM3cR1zKwTYMoyZitFai4Noahk+egaRqebXBgqkyOhqZB04+43PD58et1+nFMaxDLRGAhhBB3HCkAvststmq3/YS9Yx6vXozQNY1De6qcXjNZ7oSsdQM6QYrKc3Kg5Se0/JhH5qq0/JiRon27X4YQQgixRcLMXWazfuboQoN+mPKp/WMc3FVjfrzI+fUB/++/PMO5Ohi6RtEy6IYJtqFjmRqOpfPS+SZjRRuVD4ORdDkJIYS43STM3IU262eCWFG4IpBMVlwOzJToxymL7YAgTnGMYau2axr0Q8WPT66x1gupujYl1+Tw/BhztcJtfkVCCCHuZlIzc5dyLYORon3VyYpnG0yWXSoFi71jHiOeRcHSyTWNimeR5TlrvYhTK31KrilFwUIIIe4IcjIjtriWweMHJuiHKSeWutijBgemK7QGMSXbIs8zRjwTy9DQNY2JskNzoyj4etdNYaLwYyVXUkIIIT4wEmbEVeZqBX7nl/ZycrnL315qE6eK45c7LHcCwmS48sBzLHaPepRsk7GyQ8G+dkhZbAc8d7pOazAsGn78wIRcSQkhhLjl5JpJXNPC+gByGPFsusGwhmYQpcQqZ6UT8u9fXeRnix32jhWveeISJoqnX1vhhbMNztYHvHC2wdOvrciVlBBCiFtOwox4Gz9W9MMUQ9f42aUO9X6AHyYMohSV5Zg6OIZGohRn1nrXDCitQcyJxQ6GrjFatDF0jROLHVp+vPU1YaJoDmIJOEIIId4XuWYSb+PZBo6l88LZBs1+xFonpB0qtCCl4BjYho6ua+QZnF3rcX69z77x0tUnNNov/pXnG//+xadZbAdb7eHSFSWEEOL9kJMZ8TauZfDIXJVU5VxuB1imQdk1yHOI4gzT0Gn7MX97qc3Tr63yL350mv/t+fOcWOpctZTyvunhUsuWH5FmOQ/PVqh5NmGiOLrQoOMnjBZt6YoSQgjxvuyYk5kjR45w5MgRlJI3xFth/0SJB2ZKnFrrsX+yxFo3wrNT4kQRJ4ogUdiGRk7O2XqX5U7AsUst7psqU7BNPMvA0nUenKmQ5/lWAbBrGTQHMf0wZaLs4FrGDXVFCSGEENej5Xme3+6HuJW63S7VapVOp0OlUrndj7OtnVzp8n//Tydo+wll16DRj4kSRaJyolSRZqDU8N9F26BcsChYBkXX5IkHprCt4TLLw/OjOIaBZxv4iSJKFEfPNfEjRc2zWGoHjJcdvn5oTsKMEELcpd7P+/eOOZkRt94D0xX+0RcP8O9fvkwnTNA1jZlKgYVGj9OrfVKVQwYp0IsUaKCynCBV/OTUGgemyni2SZJmNAcxZ+t9ojTDs012jXoAHF/sAODaJo1BLHUzQgghbpqEGfGOPrV/nIdmKpyrDzh6vsFk2SU6oTi/7qOUIr7ia20DelFCxbHQNI3FVkikFBXHZLUXcmZtgGVq7Kl5nKv3cUydPWMee0a9rbqZrxyckdMZIYQQN0UKgMW7qno2D8xWmCi7tP2EQ7trVF0Ly9DxTJ2CCZapEaRgaBpVz8LQNcZKFp6lU3JN6v2YQZTS9hOawbAduxMkzFRcKq7FRNmhH6YEsdQ8CSGEuDkSZsQN2dy2XfUsCrbJFx+aYteYR9WzKBdsdo8UKFk642WH6UqBTx8Y48BEielqgcWOz9m13kbhb8JKJ2C1F1ItWHTCZGOycETJNa87TVgIIYS4HrlmEjfsym3bugYHfr7EqdUeg1Bxue3jGDol16Idxpy43OMT94zy1Ufn+D9evoima1S94ebtlp/gWiaf2j+Kpuk0BzFVz+Lw/NjWFZPsdBJCCHGjJMyIm+JavwgXn39gioJtcma1x8J6nzSHNM+puhZ5nuOYGvvGPO6fLqOjca4xYKyUEyUZu2oF0gx+7eFJshwKV4QWGagnhBDiZsg1k3jP5moFvvjAJNWCRaVg4zkmXT9hsR3QGEQstQI0DSbKLmMlB0PXIIeJssMn5keIkowsh5GiDUBzENPxY44uNGj0ImxTp9GLZKCeEEKIdyQnM+J9UTmYhs50xeFSyydVihwNP1I0BjGFjVqbRGWsdAMSlfPYPTX0XN+qkdncrt3sxziWzlI7JE4VaQNMHYJEyUA9IYQQ1yVhRrwvnm0wUrSpFmxKjklrkGEZOvOTRUY9myBRzNUKfP3QHAfnqhxf7AyXWJoah/bUAHj6+DLHFztYhsYgUix3AsZLDjM1l0vNgCjN0LV3fg4hhBB3Lwkz4n1x33LyMl1x2TNeYKUdsdqL+MmpOp+4Z5SCZTDi2Ty6q8rL51tEKuPYxTad8YQTS11MXWPEcwgTnzDJMA2NU6t9siynY+ostgOqnn27X64QQog7kIQZ8b5defLy6sUWL51rEqUZ900XudTwef7M+sbabA002DdW5MBkiXov4m8vtlB5hqbpaBoYuo5r6aRZxkTJwdQhR+P4Yof5iZJcNQkhhHgbKQAWt4RrGTw0W+XhmSr9MKUxiHjuVIOfnl3nzZUeWQYqzzhX71Pvhag8xzQ0On7MPWNFVJbTHAznCR+cq5JlOYYGoyWHj+0dFgvLQD0hhBDXIicz4pYJE8WJ5Q6tIEGpjBRYagd4jslY0cYwNNa6ERcaPt0g4VLTxzJ1PnHPGA/MVCDP6YQJl5oBa72Yei9mz1gJpfKNYX1yKiOEEOLt5GRG3DJ+rOgFKTNVd9imHSYkKsNPFJdbwcb2bZM4VZxv+timzkTJYbkdUHIMvvDgJM1+TKIyHpypkAHPn61jm9pVA/WEEEKIK8nJjLhlPNtgtGSjaxq9MCVOM1zHRAeWOyFF1+DQnhG6QcJaN2S87ACw0glZagW0J2P6UcpcrYBnm7iWznI75ON7R2VonhBCiOuSMCNuGdcyePzABEstn2MX2xi6xp7RImXHRNfg//L4PewdLfKff77E5VbApabPciekNYhZ7oTYpo6maZxf99k16rLajZgoO0xX3dv90oQQQtzBJMyIW2quVuAbH9vF6bUBkDNTK9D2E/IcKgWLy+2AfpSSqow3V7okKmei7DBVdnh9uYtSOa0gpnExYt9Ekb/32G4cy6A5iGVPkxBCiGuSMCNuuelqgcf2jXL8cptukKKynD1jRZ47vc6rF9pkec5Hdlfx45R+lPLR3SMYhsbxyx3KrsXXPjJLox8zXXMp2jr/x8uXiFVGyTF5ZK7KfmnRFkIIcQUJM+KWcy2DX3tkmpJj0BrElFyTKM3p+AmGDlqusdoNGS+5dII+a72INMvoRym7RguMFh1qns3rSx1evtAiSjIKtk4/VDx/psGju6t8+t5xJsuunNYIIYRAy/M8v90PcSt1u12q1SqdTodKpXK7H+euFm7sVAoTxV+8vkrZNXltsctyx2e9H2MZOs1BTJ6DyhVRmjNbLTA74tLxU9Y35tHcN11hqRXgxymWqeOaBirP+eQ9o+wbK8pWbSGE2AHez/u3tGaLD4xrDfc2jRRtSq5J20+YrbmsdCPCRDHiWWgaZORUXIvJioNGzqmV4WC9yYqLaxlcWO+TqoyWnzDM3sMBexfXB6x0AtmqLYQQd7kdc8105MgRjhw5glLypnan2dzfdHShwWonZPdogfsmy1xs+tiGTsHSSFRO1TGoeRa1NMc2dcruMOwstwMGkSLLc2ZqLhfWAyDnYivAMQ1SlctWbSGEuIvtmJOZJ598ktdff52XXnrpdj+KuIa5WoGvHJzhGx+b49P7xwFI1HChpKFB2TWo92JMTUflOVkOo0WbpXZIy08wDagWLNqDlChRKJVTdAyafsxKN5St2kIIcRfbMWFG3Plcy2C6WuDxAxMbA/M0Ros2u0aL2LqBa+k0BwnDiyQ4sdQB4MBUiQMTZSquScHSma65TFeHhcIzG/8521GVX0IIIW7GjrlmEtvHlVu2jy926Icp3SDm1FqfOM2ZKDmkKuP8WohuQCdIuBgEhEnGiGfwsT2jTFcKjJVsGoOIkmvJyYwQQtzF5GRG3BabW7a/fmiOv/ORGaqejZZr7B3zSFTOpaZPpkOew7n6gEYvwtAhUTnr/YiqZ7LaDbnQ8Fluhzx9YoXFdnC7X5YQQojbQMKMuK1cy8A2DfIcxss2fqywDI12mDI/VmSmVoCNUxfL0PAci0stn6JjkOXgJ4r1fsgLZxs8fXxZupqEEOIuJNdM4rbz7GELdy9MaQ4iVjsR1YLF3vEis1WXy62A5iCmVrAZxIogTvnPP1um6cdMVwqMVByag4gTS13afsx0VWbOCCHE3UTCjLjtNlu3AcquyYGpMvdOlVhuh/TDlM/sH+dvzjVIs4w8z7lnY51BlGTUexGzNZfN+l+pAxZCiLuPhBlxR9hs3Q5iRWFjRcHmBGFdg30/91ioD2j0IwxDp+QamLrGufUB670ITdM4uKvGiGff7pcihBDiQyZhRtwxXOvqPUtXfvz5B6YwjTpHFxJUlrN/okyzEONaBrtGPSbLDo8fmJDBeUIIcReSMCO2hc127vunyvztxTaJytgz5vG1j84yUrDRNVA5dPwYlSMLKIUQ4i4iYUZsG41BzJurPfphQtE1ObSnxvx4icX2cD/T5abPSjdkulpg10hBFlAKIcRdQsKM2BbCRPH08WWOL3bQNQgSRRCl/J1HZ/npmXWWOyHL7ZBumGAaGq6lc3ShwVcOzsgJjRBC7HASZsS20PJjTix1iVOFH2esdUPeWOpytj7gzdUekcoIk4zpsoNjGewb1WgNYoJ4OHfGj5VcPQkhxA4lYUZsDzmoLGetF2HoGmGqiBLFG6tdFpshcZqi6zrNfsRqLyKMFZMVh5MrXS42fZr9mNGSzeMHJuTqSQghdhgJM2JbGCna3DtV4tx6H8vQURlMVt2NtmwwDR2VZ8QJtP0Y29AY8Wz+7YsXyQDX1HlzNacfpvzOL+2VExohhNhBZJ2B2BZcy+DvfGSGB2cqlF2LmZqLrQ//+uZ5hq6BYxo4JniOiWsPt3AvrA8ghxHPQQNevdhitSs7nIQQYieRMCO2jfnxEv/Xz+3niQcneXCmjKZrOJYBmkaWa+iA61hUXZMLTZ+fnKxT74Z0g4R+lHKp5XO5GfCXJ+uylFIIIXYQuWYS28r8RImxos1//tkSE0UXXYcXzqxzdn1AwTaYLrs0/Jh6N8RPMvwo5aULTS63fRzT4J6JIkGcSqeTEELsIBJmxLajcshy2DPm4VoGUxWXlxYatMOEMMnQdUizHNvQqI24LLdD1roh902VMXWdlU5IqnKCWEmYEUKIHUCumcS249kGJdek3osIE8V6LwJd45HZGp89MI5r6vSClLJjYRsGUxUHTdMxNJ2xos1SJ+R8Y0Ci1O1+KUIIIW4BCTNi29ncsl31LJqDGMc2mK4WKLkmJ1f6DOKUOFVcag+IEkWWQblgsmfMo9GPCeKUtp/woxOrUjsjhBA7gIQZsS1tbtn+6qOzfP3RWaYqDi+db7JQ72HpBrMjHo5h0I8SakWLX7l/kvmJIpauYRo6E2WH9UHEc6frhImc0AghxHYmNTNi27pyq/Yjc1WePVUnzWCsZPPgbJXFlo9SGY/tG2HEc1jpRpxe64MGYZLRj1LO1X32jHp8dM/I1vcKEyUTg4UQYhuRMCN2hP0TJT61f4znzzRIs4zTq13WuhHjZYeRok2mclqD4cfNfkQ/SugEMeQa/+aF81xs+jx+YAKAowsN+mFKyTVlWaUQQmwDEmbEjuBaBl94YIpBmPKj11cJY8VYyWGy5LBQH7B/okSQKHaPFriwPqAXxbQDRdHWiVJFzbPx45RBlKKjs2fMo96LpIVbCCG2AQkzYseYqxX42qE5kizHcwxOLHa52BhQ78esdiJ0DRabAY6tsdrLANB1jTTLeOVCi59dapFmMF5y0I0JdtUKNDeWVUqYEUKIO5cUAIsdZaRoM1sroNKcPIcgycjynEY/Yr0fc2qtRy9IUVmGY2hkWY6W51xoDFjvx5Rcg/VBzF+fqnOx4VNyTQq2BBkhhLiTSZgRO8pm27ZjG0Sp4uHZCvdMFJmpFVB5hgHEqdoIOoowzWj6CZqmUfNsHMNgtGCy3g/JyTk8P3bVqUyYKJqDWDqghBDiDiLXTGLHmasV+Pqjs+jA+iDi9GqPxXZAy0/Q8pwwzSm7Ft0wJlYZeg6GrrPUDugECZ6lU3EtbFOn7ceMFW0A3ljucuxim4ycEc++oeJg6YwSQogPnoQZsSNVPZsvPDjFj0+u8ten1klUzljJ5lLDx9R1ZqsFPMeg0Y/RGNbO9MMUP1aUHZNDe0f4+eUOxy61OTBZohPGHLvUQamceyZKTJVdEpXx9UNzuJZxzdCy2A6kM0oIIT4EEmbEjjVXK/DFB6ZYagW0g4RukNL1E/qxIs0zLEPHsw0SlZEq0DXQNA1Dg1fOtygXLJI049ULLVSWY1sGZcfktcUOi57Pai/g4FyVqmfz3Ok6zX7MaMnm8QMTjBVtji406PgJE2VHOqOEEOIDJGFG7GgjRZt7p8qs9yOqrsWuEZcLDR9Ng+VOiGFodAYJcZpgmRpZrpGR0x4kBPGwLmYQpWR5zogODV/hGjpBbJCmOa9ebNGPUk4u97AMjTdXc/phylcOztAPUybKDoauUbANWtIZJYQQHwgJM2JH2ywI3rzuuW+6wv/5k3vwLIPlTsiP31zlv762SpxmpHlO1TWJ04wwVURKYesaaDmJyulFKaBhF3SqBZMHZkvUuyFn6wNcy2DEc2gOIk4sdXnioUlKrsmZtT7NQcR6L6ZWtGj5MSMbNThCCCFujTsuzFy6dInf/d3fZW1tDdM0+Sf/5J/wW7/1W7f7scQ2trnHKYgVhStqWqarBebHi4x4DmfrPU4u9+iFipmKSZJnXGr6hElGqnIcUyPLcxxDZ6biMFF1eO1SD02HRGU4loGmQb7xM23T4NCeGi+fb9LyE0ZLFkXb4qXzTWZrBTmdEUKIW+iOCzOmafKd73yHQ4cOsbKywmOPPcZXvvIVisXi7X40sY1ducfpSlXP5lcfnubogsWo57DYCpgbLeDoGv/ltRWWOiEFSyfPoWAbfHRPDU3XafRjojTFRKcXp+Q56JqGynIO7qox4tkwiJmpFZipuaz3Ylp+xNGFiHvGioyVHcaLNlVPTmmEEOL9uuPCzMzMDDMzMwBMT08zPj5Os9mUMCM+MFee3CRK0Y8UqVKs+zGvXGjRHsSgDdu3qwULy9BZbgXDQmKVEMYKHbhnvMiu0QL3T5c5tdLl5Qst3lzpstQOmaq4eLZBvR/yL/7iTUY8m5Jj8sv3TfC5+yYk1AghxPtw00Pznn32Wb761a8yOzuLpml8//vff9vXHDlyhH379uG6LocPH+bFF198Tw/3yiuvoJRi9+7d7+n3C3GjXMvATxTPnWnw5z9b4n959hwL9QFJmuFYwzqatp9w9HyTowtN1noRWZbT9hOCJN8oAu7y0rkm//NfneX/9u9+xo9OrJKT48eKejcky4Zf1wlSVJbzNwsN/p//9U3+H3/+Oq9caN3uPwIhhNi2bvpkZjAY8Oijj/J7v/d7/MZv/Mbbfv173/seTz31FN/97nc5fPgw3/nOd/jSl77Em2++yeTkJACHDh0iTdO3/d4f/ehHzM7OAtBsNvn7f//v8y//5b+82UcU4qaFieLoQoNGL+J8w+fUao88z+kECRrDGpjJsoXKcsJEkQONQUgYZ4BG0dE4uz6g4loUbJ2OH6MDU2UXz9YZxCktP2a1EzJdcVlsBxg6aBqsdkP+/SuXuHeiSNWzCRNFy48hB882UBv/ljobIYS4tpsOM1/+8pf58pe/fN1f/5M/+RP+4A/+gG9961sAfPe73+UHP/gBf/qnf8of/uEfAnDs2LF3/BlRFPHrv/7r/OEf/iGf/vSn3/Vroyja+rjb7d7gKxHiF/xY0Q9TXNtguRNQtA10TYM8p96LmCjbeI6FTk4v0gjilLVIESU5lqERJzlRouiR0+jnxEpR70UE4yl+lOGYOgXbQNc1moOEnJwcDc8ymCw7NAcxzUFMP1Y8fXyZE0td/DjFsQz2T5TYNVKQoXtCCHEdt3Q3UxzHvPLKKzzxxBO/+AG6zhNPPMELL7xwQ98jz3P+wT/4B/zKr/wKv/u7v/uuX//Hf/zHVKvVrX/kSkq8F55tUHJNmv0IlWUEicK1daZrBVzbIE6hHyZ0w5RemFDvx5iaRtE2sAydOE1JVU4/VOi6RoZGmmecawzwHIPPHJhgrOiwf7yEyoY7oaI0wzZ1Xl/u0Q4SemHCc6frHF/skOc56/3hKobz631WOsNpwrITSggh3u6Whpn19XWUUkxNTV31+ampKVZWVm7oe/z0pz/le9/7Ht///vc5dOgQhw4d4vjx49f9+m9/+9t0Op2tfy5duvS+XoO4O23Oo5muFRgtOpiGjmeblF2Lv/uRWX7lgUkMQyNOMwqWgQY4lsmuEY/pioum6YyUbCqeRcHSqTgmUxWXPSMen54fZakdECaK8bLDwV0jfP7AOLtqBRKVM1q0eXCmwrHLHdZ6IZahUXYtyCFKMxbqAy6s+1xu+luD/IQQQvzCHdfN9Pjjj5Nl2Q1/veM4OI7zAT6RuFvM1Qp8/dAcB+eqw8m+YcpIcbieQAdMQ+NSKxiuP8hgvR8TqQzIma65PL5/nItNnyTLsXSNHI1P3zvGwzNV/s3fXCDLoWDrPDRbIU4yZmoFKgWTsaJLluecr/fRdYanNklMYxARJBnTFZemH2MYGrp2u/+UhBDiznNLw8z4+DiGYbC6unrV51dXV5menr6VP0qID4RrGTw0W2V+onTVkL0wUYyWHE6t9JmuuQxGFFmWozamBh/aM0q1aDOb56z1Isg17p0q8el7x5mtFrhvusTPL3VQWc7L55s8MlNhplbAjxSNfsRfn67TDhNGPRuV5RvPYjJWMhkvO1QLJuMllyx/lxcghBB3oVsaZmzb5rHHHuOZZ57h13/91wHIsoxnnnmGf/gP/+Gt/FFCfKDeOmTPtQwePzDOyeUuy52QuRGXfWMF1noR902VqRZsXjrfJFEZ94wVsG0TzzY4drENe8AyDNIsY70fESYKQ9f52lSFeq/LM2+sUe9FlByTuorI8pw9ox5ffGASzzYZK9l0woTxkkPBlo4mIYR4q5sOM/1+nzNnzmx9fO7cOY4dO8bo6Ch79uzhqaee4pvf/CYf//jH+eQnP8l3vvMdBoPBVneTENvV/HiJ3//sPM+fWSdSw63bVc9h10iBvznbYKUTEKuMxVbAeNnhyw9P0/ET/urkGl0/YaRgU7RNNB3OrPX5D397GdPQsA2dyYpDsx8TJIqJkkOU5pjGcEFlP06pFizuny7f7j8CIYS4I910mHn55Zf5whe+sPXxU089BcA3v/lN/uzP/ozf/u3fpl6v80d/9EesrKxw6NAhnn766bcVBd9qR44c4ciRIyglBZLigzM/UWK2ViCIFboGz5xc42JjwLnGgEjl2IbG6iCiEyS8VGoxXrK51PSJ04yVbsg9EyUur/tYho5uQC9UhKlipRPSC1MgJ82G277PNQYcmNDIgSBWPPtmndcWOzx+YEJatIUQ4gpanuc76ha+2+1SrVbpdDpUKpXb/Thih1tsBzz92jL/6dgipq5h6DrL7QCV54yVbFqDhPGyw6f3j/LsqQaNfrRxomNz/1SJMFG8dL5JEKUEaYauQc2zqRZs9k14fO7eCX5yus56P2bfWJEszzk4V+V3fmmvDNETQuwo7+f9+47rZhJiO5mrFfjNj+0iiFJ+vtSl3g2peDZhkqJUjqFrjBVtumHKbM0lVhlFSyfJcpIsp+0P9z5VPBtXZagsR2UZa72AWsHkbxYaLHdCDE2j7JoMopQTS13afsx0dXg6EyYKP1YyJVgIcdeSMCPE+1T1bL7x2G5qxVX+5myDWOXo2nCIUzdMMXSNE4tdBrFirGjzsb0jtAYxFxs+3SBlvOTQCxNsdFIycjTiVHGm3udiazifZq7mYRoaaZaTqIxOmGCbBm0/5qXzTZr9mNGSLVdQQoi7koQZIW6BuVqB33xsNx/bM8KrF1u8eqFNlud8ZKLIT8+sk6iM0aKNZej0woTpioMfKUqOSZLlZCpjuRtRdEx0QHc0EpVhaJDlkGYZ59cHLLYDyo7Jv/zJArO1AhcaA4Iko1oweXM1px+m7+kKSk53hBDbmYQZIW6RK2fUfGzPCM+ernP8cocsH3ZCPTBbZq0T0ejHzFULjJcdmoOIpU5IwTaZLms8OFthod6n3o8YLzkUHZORJMMw4PjlDkGsME2dxiCmEyRcavl4tsnu0QLNfsyrF1s88dAkU5XCDYeTxfZwVUI/TCm5puyAEkJsOxJmhLjFXMtgfqLE8cUO902WqRYsGoOYtW7EdNXlwFSZT+0f5X/68QK9MOW+qRKXmgGzNYNHd1fpRyntIEHXNWKVk+YZa+2ITpjimDpRqmj2M6I0JVXQHAxn1PSjYZ3Of3h1Cdc2IM+3JhhfL5xsbgvv+Ak1z2KpHfDc6TpfPzQnJzRCiG1jx4QZac0WdxI/VkRJxv7JEtNVlxNLHRr9mPsmyzw4W+X5s01afkSU5nSClN2jBcZLLr/2yAyfvGeMvzy5yonFLiu9kKJj0+rHGLqGlucY5HTDjCBV5DkkCpqDdcqOwSOzNf72YpOWn7B31CPNc9p+wjc/vW/rua48rdncFm7oGq8tdumGMWfrAw7OVXlotnob/wSFEOLG7Zgw8+STT/Lkk09utXYJcTttbuGu9yImyg67RzwOTJX51YeneO5MgyhW7BrxWG6HOKbOdLnAdM2l5tlMVwvMT5Q4v97nL95YwyBnueUTqeGpTJjk5ECUgqmBpUOSQT9SnFzpkmQ5hgZBojB1nZVOSLVgEcSKjJwRz966SvJsA8fSeeFsA1MfzrTJ8pzjix3mJ0pbqxyknkYIcSfbMWFGiDvJ5hbuowsNmoOYsbLD4fkxTMOgH6YbSyYtABqDmEphWKtyZXiYqRbwLINjl9p4jkmzH5GkORqwuW9yc0iUvvFBN0hIMrBMcCITU1c0B4ojPz5NtWBx33SF3SMeico4fM8ojmlw32SZowtNNE2jWjDZO1YkSjKCWNEYxG+rpxkr2hJuhBB3FAkzQnxA5moFvnJw5m0LK691YvPVR2epevZVxbiOpdMJEgxdY36ihMpz6t0IzzEwdFjtxMQqZ3PspcoBDWwT8lwjSlOCHBxDR2U5tmGw2AqIE8XPLrX4qzfXKFgm90+XuW+6jFI542Wb9V7EeNkhUYq/fGOVIFHMVgvUexFPH1+m5JpESSbFwkKIO4ZMABbiQ3a97qEwUfzw+DIdP2Gi7HChMeBsfcAv3TOKbRm0/Ij/8Opl4jTHtYfBpO3H5DnoOtiGjq4Nr4ls0yDLcxxDx7EMLENntOgQJorVXoiuaUyWHSZKDoahce9kiaJtcLbuA7Br1EMj5/WlLhNlh0d3j2AbGn/15hq7Rj32T5To+AlVz+IrB2fkhEYI8b7JBGAhtpFrndjAL4pxJ8oOrmUwWytwtj6g3o/ZO+ZBrvH5+6dYqPe51PLZPVrg8w9MEKUZfpByqt6jNYhRuUaqclQ23Puk/ATb0ukGCUGakWcZtmWw2gu53A4wNZ0LDZ+9YwXuGS8zP1HkRydWWe0F2LrOYjugE8SMeBbn1300hl1Qs7UCi02f8+t99o2XbirQSB2OEOJWkjAjxG3gWm9/E39r0XDbT3h4tkLJNWkOYqqexa8+Mk3JNljphLiWTs2zeebkGiudAF3X+PliB12DVGV0AgUaGLpGmmZ00wxT07BMHT9K6QXDqylDB00b1t+MeA4XG7BQ7xGrnFHPou0nvHiuiWsZ6JrGmXrGSifkb840sG2dN9d63D9d4e98ZIb58RIwDCstP4Z8uDTzytcqc22EELeahBkh7hBvLRquetZWwe1bT3Gqnr31+w7Pj/HMG6vkwK6RAhNlh9cWu2hhiqlreI5BmmbEKsezTfSNTqc0HxYOWzpoZHSCmBfPN3ENjfV+jOcYrPQi4liRaZCkwy6qej8kU8OrrHsnS6x0Qi42fC43fb71mXtQec5fn65zaqUPwMNzVX7tkemtqzSZayOEuNV2TJiROTNiJ7jeFdQ7vdHP1Qp8/dFZlMp49VIbpYZXSZunLUrlJKlC0w1cW8ePFCobdkQ5loahabT8BJVB30wxjGHdTS/IMA0DXcvR8pwk14b1OZpGpHKyXHGpFTBVcXBNg4X6gCM/Pk0OLLVDpiounm1w/HKbkmPw9UNzb5tr0/IjXl/ucv9UmUN7Rj6UP2MhxM4jBcBC7BCL7YCnjy/z88sdFtb79MOUMFUMwpQcsE2dOM3I8+GyyiwHfSPwZNnwOso2NNIcDA1ilWGZOlXHJAe6QYqmgcoz4vQXv7dg6TjGMCgVHZOxkkM/TBgrOewZ81jvxczVXH7jY7uYqRZ4+sQKL5xtEKeKtV5EmCjunSjxP3z+Xh6Yeef/n73Tam3utOcRYjt7P+/fEmaE2EHCRNH2Y5Y7Ia8vdVhu+byx2iNOctDgtaUOKssZKVhoukajH1EwDSxr2O7d6CcYWk6a5ahs2Os9O1LAj1KiRA0XZUYKQweV5SQqR9OgVrBA13AMDdPQINewTR1b12n4Mbaps3fU46N7R3h4rsIPf7bCcicg2QhMvTDl4ekK/+Ov3Mv8ROmar+2dusBuR6CQ2h8hbi3pZhJCAMPrqOlqgelqgQdnKiy1A35wfImLDR/L0Fluh4RKMVlxGfUcLjs+o55NP0xpBTFZnpFlGkXHQDd0olhRdkx2j3gYmkaQpLyx0iXPcmzboKRpJCqn5lksd0MyQ6dkm+RaTqsfYZo6ChhEKa8tdVhYH3B6tcb8eJFOkJAoRT9WVAsmrTDm+TPrzNYKbwslV9baTJQd6r2IowsNDu2ucexS+0MPFNd6nudO1/nlAxNvK3gWQnzwJMwIsUNttndPll3Org1IlcJzDYK+Gl47WTpffHCaRCmeO7VOmuaUHJM815iqOPQjxZ6ax67RAh/fN0qUZBzaVeN/evYMFxoBKssYRCmWrrHej0jTnDxTqDxntupSLlj0w5T1XoTKMnRNw9Q1zq0P2D9epFYweW3ZR9cgSTWyLGdhvcfPLraZqjqMeDYqH3Z5tQYxa92QyYqLaxlMlB1WuyHPn1knUflVAedWz7251snPW9voDV3j6EKT1iBmsuLKKY0QHzIJM0LsYK5l8PiBCfphyomlLrPVAntHi+weLbB3rMgn7hnlpXNN3IcMygWT840Bbyx2KdgG5UKObehUChbdIKHkWoyVbT4yV2Oq7FLvRSy2A1Y6AWG8UXCcQ5blrLQDbNOgHyuidFifk5OT+gm6ptH0I752aJbFVkAjiAG43Ao4U/f561PDPVGzIwUe3T1C0TZIs5w3lru8erHNwV1lTHTQNdIsY67mbQWc5iAmiNU1T3bey1XU9a6SNtvol9sBrm3w0vkmhq4xWXHp+MkHEqqEENcnYUaIHW6uVuB3fmnvcFow4FkGWQ4F2/jFdu+NpZKTJZeaa1N0TAZxymonJMtgoT5gouzw3Jl1iq7FSJoRJhmJ8jF0jbGixSDJSFRGlGT4OdgqH7Z957/YIZVl0B7E/ORUHc8yuGeiRL7eJ1SK1M8JkpTGICPNcpY6IVqeoxs6KhsWHJ9c7vLaYpvxssPDM1WCNOVyO+TgTIVBrKh6FgX7FwEiTBRn631eW+xccwXDtULO5ucMjWtebW2GlL1jRV4+36Tei2j5Mb983yQV18I29OuGKiHEB0PCjBB3gc1ammu5clBfvRexb7zIFx+YZLEd8OK5Bs+eXh++OSeK5U7IbK1Aox/RDhIqroltarT8GI2cdKNLytQ1NCBDw7U1ojQb7pDSQNM1/Fjx9GsrFF0LU9eYKDmstgJ0NFQ23DcVpSkvXmgxUXJQeU7JtZisOFxuh6x1I6KkRS9MyHM4frnNJ/aO8ql7x/Dj4XiGxiDmudN1ji40yfKcj+0duerU5FpLNIGtz6FB20+YHy9i6BqFjeuuYOP7X2gM2Dde5KGZCkfPNzm92mPEswgSxXjJuSpUCSE+WBJmhLiLXW9Qn2MZnFzp0fET+mFKloMfp9QKFhcaA+6dLHFwrsrptT71XoimaSRJBig8x8A2DFSWAZBmOa5poBsQJTkaGVkGcZZTIidKc15fbjOIU8Jk2BaOtrE4M1HU+9Fw51QGI54FeY6pQ3MQY2galgH9MOWnZ9dZ6QbsHvUYKzkkKkepbDjhONe40Bjw8GyVfpjS9uO3nbr8+OQqSTo8HRotOrT6EYstn7YfEyaKRi+m6Bp85t5xJisu/TBltjosVj4wUeInp+r8+OQaNc/itz6+Bxg+41uvtqSdW4hbb8eEGRmaJ8R7c61Bfc1BTD9MGS056LpGwdKJ05w0A1MfrlFIVD5cOBkkzI+X+OjuKn9zrjm8otEhywxSpTgwVWGu6vKT03WiJMa1DHKGKxeCVOEYBrGCasEmzWJiBWxMJwaIU0WeawRJSJgMi5ejFDJyio5BkGQ45Kx0QvphwsnlHrWChZ8ovvTQNBXXphPEdIOUc2t9PNek3gu3CooNXSNKFUfPtelHKXmeM152MTSodyN6UUKYZBRdk0iZ/OsXzvP3P7Vv60Sr5lksrA+oeTa1gkXHT/jfnl/g1Qs1yq511dWWtHML8cGQOTNCiLfZ3ODd6EW8sdLl1FofzzKYqroc2lXjCw9MbrVEO5bOwV015qou//tLl3jxbINOlBAlit2jRf7RFw9QLVj8y78+y0vnW7T9GD9KCdOM6YqDbei0g5SCrRNGKav9BBiuWYiH66UwDTB1MA0DcnAsfXjykucYGiQbc3EmSg6DSKFpEKcZlYLFZ+8dp58oWoMYP06xDJ08h4Jl4lgaBcvk/HofRT6cbhxnjJdt2n6KocOuWoHlToBpGDw4W6bei/nkvhEO7R7h1FqPZj/m5EoPQ9dIVEZrEHN2vc9kyeFrH51FR6fqWXzxgUmeObl21WmQbB0X4hdkzowQ4pa68vppX1LEMnRGiw77xj0ePzDBXK3AbK3wtrULX310lhHPot4N8RyTzz8wyfx4iTBRPDJXw7NMztb7HL/cIUhjBnFGag1/5iDK0DWDgqXQGHY/xSrfqLMZ1t9EaUbRMdk7WmS1F7E+CEnRyLMcXRuuURguztSoFEzaQcJfvbnGAzMV1nsxigyN4YqHaiGjHeQkaYZh6GQqpx0lFEyD1iDCNAyqBQvD1MjRMHSN9iBF13JevdgiSjNGijafunccQ4fnzzZJlKIbphQsg0yDM6sDHpyt0BrENDZOuzbbua/svgLk6kmI90HCjBDimq68ftI1tjqgrtwX9dY33rlaga8fmrvmbqnD82MkKuN8Y8BIyaLiWcNgkeWESYZraeiahqHbmLqGH6WkKkXXczRNJ1UZhq5j6hrrg4h7xooUHYMozej6MfHGvqhcgaHldAYJmqbhazlr3YBOkDA74tHqx1iGRpgoZmsFLEOn5ccMohzHHHZ6qVyjO4ipdyNGyzZRkhEmilRltIOEom1w/1SFTpDwyvkmscpoDSLq/YjZagHTtVB5zqm1Hmu9iKmKw2fuHX9bsXXVs2j5MT9+88Mf/CfETiLXTEKID81yO+B/f/ECl1shrmmw3A1oDmLyPOcju2pMVRyOXerQHAwH7Q3iDJVlrHYiNHI818LSwU8y5mrDAXqpygniYaFwmqYMkl+0gjvGcCdVwRrOqnEMnTTLGcSKUc+k6rmMFW0cS6fei0ADU9NYbAeYuoZhaASxwjI0pisu/VjRC4c7qibLLgcmhzVDB+dqWDr84LVlVJaze8Tj3LqPacBHdtWYKLvsHfPeNrH40J4axy625epJCOSaSQixTYwUbeZGPC42A2KlqBYsCpbBQ7NlJsouUZLxqw9Pcf90hTzPOXquwc8utklVB8vQKBUs1johtqGTZVB2LZqDCD9OKbsmUaqjiAkTNq6qwAC6YYpnm+g6dP1ko5sK+nGCyhS6rpPlcM+Yh27A0kaYMQ2dsVGHlh/jJxm7ah6XWz7dIKHjx1xq+YRJhudoTFU8vv7oHM+eqtOPhtvH940VOTBdomCatAYxI57NVw7ObHVIBbGi5cdMld13Hfz3fkgH1c2RP6/tR8KMEOJD89aJxLqm8di+UX7tkWnGivbbrqcenKnQPhTz88sdfvjzJU6v9ZmfKLFr1ONS02e9FzFZcig7Fq6tc27Nx9Cgrw+XYiYqI1IZtqEzWrSIkmy4FJPhlZYGNIMEzzQYLdpcavloGpQKJoMwRdc11rohI55FmoGuaXiOQZAo0gxKjkkvCHnudJO5EZ8cCFKFpWu4lsFKL+RHJ1ZxTJ2pskvLj2n58IOfL3Gm3odcAw32jaUcmCxtnczcyhk10kF1cxbbAc+drg/DZ9HeqhF7p4Aj4ef2kzAjhPhQvXUi8YhnX1Vbc6UrF2ce2l3lP/ztElGsWB9ElG2TPaMeu6oFFjsBE2WHUc/h9aUuZTcniFNavsIxdQq2iR8p0iyjYJt0woQRxyZRGXGS4Rg6909XaAwiljshj987xvNnWsPlmyrHLOsYOvTChFY/JohTJisuicqZrLhUCxZLzZBWOJx9o23U/Kx2QwxDY3fNw9B1/s3fnKc9SDi73sfSNWoFGzSNlU6AZw+f89Ce2i17Q7zegk65xrq2MBkOczx+uY2+sVi17Sf8nx6auu5CUwmLdwYJM0KID907TSS+nslKgb/76CzPvLHKm2s9ZkZcHp6tYRsaQaIYKw3bvBOVDfdBZXCxNSBV+XCPUiegFyoGUUqWazSyaBikTH1jd1SGYxroGjimwRcfnOCZN9bIgQOTZWxL5/jlDoahU3RMwjRjqe3zwEwFy9ApujpBrJPmOc1BQqyGizg9y2Cs6LDaDXh9sUPJNbFNAz9KaQfJcM+To5OqnERlHLvYxjGN674h3swpwFsXYt7INdZbv//ddOrQGsScWOwQpYowGbbZr3SWNwZG2m8LhPDOKy/Eh0fCjBBi25irFfj6o7PogJ8ookTxwtkWuqYxN1rgiYen+WbVJUiGtSj/v1cX+dmlNoamUXFNsmw4v0blECYZhp6TaxpK5ZxY6pEDRcvk1YttsiwnSBS7xzwsQ6dkG6RpzvxkkZmqy0o75I2VHm0/Ye+YxWIroRcNO51SldEPUyxTJwd+tthBy3NilVFyDdZ6IZauMYgVmcpYaof0wpTPHJhgeeOa4+uH5t62L6rtx9c9IbiWzYWYb+2gutY11rX2WO0dK3KhMbiqYLlWsG8q2GyrMKQNu+vWexGmoeOYOr0w4fRqnycemnpbIMzhpsOi+GBImBFCbCtVz+YLD05t7V0COLSnRpLmnFzuMj9epOrZ1DybeydLrHRCojSjE0QULIM9ox7jZZs3V/q0/ZiyY1J0TfwwxbF1RksmQZTRj1OKtkXXT3gj6jCIUoIkI1MOeT6cO1O0TTzb4Phih46fkOU5ocpIVc5MzWWlE9LZKB6uugYFy0DTNHRyekGKyqHsmihgsRXyoxMrTFddztYH3D9VZs9YcSvA1Lshb672mRlxeXCqckOnANdbV/HWr9+sE7lyj1WjF/Hy+Sb7xovMVgucWevz8vkm+ydLjHj2DV2nbF7BtAYxlqnz+IFx5sdLt+TvwQdhxLO5d7LIufU+hq6BpnFgqgwaLLdD7DH9bYHQsXTON3xma8ON6beq5mlbhcA7wI4JM7LOQIi7x1ytwC8fmKA1iJmsuFRcizBRV/2v4iuLjX9+uYNjmehGTnmjANjUoOpZPLZnhHaQcGath5blLLcjXFsnShXzE0XeWOlRsHRKjkXBzof7qPoxmgazNZc8AzKwdI1YgcqGYWYQKkqOxSBMSJKcTqCoFXVa/ZiMnIzh0s14Ywnnpcin6ev0ggTXNPh//eUp9o0VaQxiKp5FGCneWOpyueUzUXSoeRYrnZDVbkDZta+7+btgGXzxgcm3zQnatFlXs96LNgYKZpxd67N/o+286loYukZzENHyE0q2edXCzne6rnrudJ2zawP6UUJ7kHByucvvf3ae+Yk7M9C4lsHfeXSWy62A9X7MVMVhouxScU1Krvm2QLjYDuiHKQv1PqdXe9w7VeJzD0y87/AhdTg3T+bMCCG2pc2VC+82oyVMhtczy52Qo+canFrpE6WKOM0puyaJymj0Ik6sdPFMndnRAt1QQZYzVSkQpcNrhNGiQy9MOFsfsHukgKZpjBUtVnsRq92IOFUUHBNbH9btBEmKqWsstQOidLhR3NRB1zWqBYssg4afkGUZlmkQpwrT0NAZBhw0cEwdQ9epFiz2jRZZ6YW0/ITpskOtaGMZGp49rD+6d7LE4wcmgGEdx/n1AZdbAbtGC+wbK3J4foyxov22/7XfHMT8+c+WSFTGs6fW6QQRpq5z/0yZVOXsnyxRcS1+/MYaxYLJR+aqZHnOIEr5rY/vZrp67U6fE0sd/pefLLDcCdA12D1WpB+mfGb/GL/58d139GnDQr3P82fWiVS2dQr11m67K//+5XnOyxeaWKbOp/ePb3VAvRc3+vd6J5I5M0KIu86NXqFc2RH14ExlOOMlVRxdaHKp4fO3l1rUBxFV16JSMMlzGCva6MBEySZIDWzTYKrqsNwJeHi2wvxkiXNrfRbbIa6lD1u08xxT08iBeydLtPyYXpgyVclZ7cbY5nDtApqGyjUOTJXQ6n26wXBoYNE2SNIcRUacDRdt5tnw2GcQpSiVUXAMHAPON32cdoBh6mR5jmcbnFwust6LKDomp9Z6nF3t0w1TljdODy42BpiGhmUMF4U+Mldl/0SJVCmag4gTyx0cS0P5sN4NSfOcj+6pEacZZ9d6rPZDeuspL59rkKic8bKNpmk8PFdhrRtt1Qg9fmCckm3w45Nr+FFKnGbYps7Fhs/+iSKRyj7UmpIrT6lUzg1d28xPlJitFYYdd/lwPtJbJ15vFlfXPIvXFrt4tommaaz34vdVBPxeiraFhBkhxDZ2rY3f7+TKLirLMHgmXaVct5iuFtA0SFVOnufsGfWYrLh86eEp1vsxz56u88bSsNgXTaNzocXldoBnGWi6hWPq+BsrEu6dKjNdLTA/XuT0Wp9elOA5On4MaT7cOzWIUi5vzLSxTR3bNEgV5KTE0TDIaICuQzo8JGKpE1FyjI0i1Ywo14gjha5lRIlBECvONbrY+rDTKk4VUxWXpU5AvR/R9GNqBYtRz6bgmDx/psFoyeJiI+Bic0CjHzFedqgWbPaOFfHjlNcud8nyDM+2CJOMME4ZxMNFnmGS8xcnVnjm5CqjnkO1YBHGir8+VUfTNBZbPo453H0epwrbNCi7JiOefUvn6LyTzVqg8/UBTT9mbsRj10jhhq5tGoOY507Xr9ozNlstbJ1AbRZXX2wMqPdDDGC07DBTc+mH6XsOH5vfd7kdUC5Y9IKEsbLzof2ZbVcSZoQQ29q1dkTdiLd2RrmmwasXWxi6zq7R4ULNyUqBSsHm+GKH+6bKfGRXlaPnGpxY7DJRcbh3skycKhKV8zuH93Kx5ZPl+dbVRJQofvjzJY6ea3JiqUvRsZiuuvSjhPVuzGjJplq16YcJfhgTpxkqH04uzgClfjHJOAcGkcK1NNIsJ85yNB2iFCKlGEQKlYNlxMNFl3nOQn1AzTNp+zFZBoansdINyfKcdtHi6EJMmGTMjrh4tklnkOCaOmEM/ThFQ2MQpYx4KboGlqkzahloQJoqBuTYSqOZRQyihH3jRZ47s07JNhkr2awPEoqWzljRxbF1piru+5qjczNFsWGiePr4MscutVntBvhxRqIyXEu/oXqfp19b4ejCOm0/IUwULyw0eGzvKJ5lbNWx7B0rcnShwYXGAEPXuWey/L6LgF3LYO9YkZfPN4c1SwWLj+0blVOZdyFhRghx19rsjNostvzU/jEO7qoxP17cevPwY0WUZOwd83Ct4RyaxVbAdMWh7Jqs9xWGrvHQXIXPHBh/2ynR7//yfj573wRHfnwWU4eZWoGldoCh+fzaI1PU+zGNfsz59T7rvRgjTYlVRpwOA4yuDZ9VA/J8eFWSZ8NfG57mDD+nAYY+PNVJspw0zUgz6PgxuaYxXnLx45QwyRjECWGakSpFoqDlJ8xUXeq9iH40bHkfKdoMNoJDpDJKjokKFUmeDcOUyvBsg8w0yADHMljvxegM63hmax6GHtLoR8xPFCkVLHJ41zk613OzRbEtP+bEUhfIsXSDoq2x3An52N6Rdz05aQ1ifn6pTXOjC8vOdN5Y6qBUzq89PE2jF/H0a8u4ps7+yRL3jBc5sdzl/PqAw/Oj17zuvFFhorjQGLBvvEjVteiECRcaAx6erUigeQcSZoQQd7V3u6p666yWVOXcO1kiy4fFsyrLObirtjXJ+Fo1O/dPVzg8P8bxy226QYqp69w3XSYHokRh6FB0TEqOSZ5rzNVcLrd8zjcGw3k4hobKcvIsI9dyLEtH0zTIM/w4R9OG83M0XSdTGZnKSDIwNfAcgzTTCOIU8px+nA53UyUJUQKGoRPEitVOxFjZ5qO7q/zsUheV57imTo6GpWsULIPYzdHynF6YomvDYYBlx+RCyyfLwTIgyTJWOiEF20QDDkyWmKkVyHOoeRZL15ijs+l6Jy8dP+Yv31glSBSz1cKNDafbaG0xdB3b0un1E2xD51IzYH6i+M4nJxqkG11p1YLBepgQJBmr3ZAXFhpomkYvTHBMnU/OjzFbLTBVcVnthnzuvombHgh5pc2amdlqAdcyKDqm1MzcAAkzQoi73jtdVb210His7PD3P30PZ9Z6V+3veac3Gtcy+LVHpik5xtbvuXeyzJm1HhcbARoaczWPs+t9CqaOrsPsiEetaPPmSo84HZ6GRMmwgHiqYhOnkKgMjZQkyxj2QQ1bvsnB27jWGe6oUnT84ZwcbeMUJ1E6up4B+fD/tJxHd9X47w/vZaZW5/ilNoqMy82QWGUYKcxUXPaMeqRZzmTZ5lIr5Pz6gDzL6foxbT9BI2fEc2j0IwqOwd/dP8MgVli6zmuLXbphzNn6gINzVR6arW79Gb315GVzQF/bj3nuzDrPn1mnXLDwnBsrih0p2jw8V+X45TaebbCaDTvKFts+kxWXxiC+7snOiGfzkbkqy52Qei+k7Sfouoamw8J6nzTLeWS2ip8oXr3QonTfsF19suJS8+wb/Ft3bTcz6FD8goQZIYR4F9c6vXl4tnLDhceb3+Prh+be9j0OzlU5vthhtRMCOVGaoWk6h/ZUObS7xn97Y4UXzqwTKYWGS9k1mSy7nFsfFu3uHivS8uNhPYg5XILpGBqjRZeia9AcDK9+bFND1w0MzWCQZESxouiYFGyd8ZLDro3i2NeXujy6u4pj6qz3QnQ0BtHw2ilKFT9f7KBp0PYLpFlGN0yGs3IAx9apucMVERXXot6P+cmbdQxDJ04zKq45rAfKc1692GLEsxkpDt/8N9cC1DyLk8tdji40mJ8ocqHhYxjDk5ALDZ+lVsChPTX2T5Qo2FevWwCuOtnZDJBn1nqsdUOcjefqBu88J8e1DP7uoTlyDX56ap2GP5y30w8VYZziWAbVgsX8RJGfX+qw1A6YrQ0Li2F4Yvdeh93daJcevLdOrZ1KwowQQtyAt57evJfC42t9j4dmq8xPlAhiha5BkKirFnA+PFvhv/vEblY6EccutShYJm+u9LAMjblagcPzoyy2A1KVc/9UiVOrfVoboeD0ao/2IKboWOi6zmjBpD2I0TVtGGLKDkXHIFEw6g0HxF1s+Lx8vknNs7jY8GkHCanKmTYNOkECOYx6Nq1BzIXmgKpr4tgmupYTpcOi5JVuQCdMGfEsXEun5Sd0w5SSazDi2VQLFq9eGK5lmKy43D9dph+mZHnO0XNNTq/2SFTGeNFmqR0QphmjRQtD16n3I1bbIX/vY7toDOKt05xEZaANr8SurKn5tYen+beDmJGiw66RYTdSy4+oFMx3PNmZqxX47z+xB0PTsBcMWn5MmKhhwXea8cJCA1PXGC85eLbBoT01AH54fPk9Dbu7MpRdGZ51bXP9xtXPunmSdbnps9INma4WbrhTayeSMCOEELfZlSGneo1f2ztWYqpSYLEd0PETHt1dpeVHGLrOnrEii+2QWCmWuxH9UJHmObM1F5XnjPoJe8c8XjrfohcmWDr4CZi6zkzFoROktP2Ik6tdBhtv1mGiUConTDPyLGcQKd5c6ZDlOTk6nSAhSNKN2TIaUZoTJQpyxeU4RQGmYTBbKzBZKaBpOgXLYKrkUnYNXrrQxrMMJivDFQB/e7HFaifg+bMNEpXRDlNcS+eViy0GkaLeC5kojjJechjxTO6Z8DB0tupoagWLv1lYx9B1vvDA5FUTilUOpqEzXXG2wsJyJ+S+yfLW1c31anVUPlw6+vG9Izx7uo6pwSDS8exhsCvZJuWCRZrlvHRuuFrDjxQTZYfldsAzb6zy9UdnqW5cPV3v51yvuPnKsHbl5zenNjd6Eev9iMYgxjS0G+rU2qkkzAghxDZw5fVDP0x5bO8oaBpRkvHgdJnXV3q0/ITd4wXKjkXZNakWrOF0YpVzYLLE0XMNHMNg/0SRXMs51/DJsuG1VstPWOs2SZRibsSjYBvM1lzWuhojHlxq+8N1B3lGnGloG8s7O36KberkWUaQ5hiaYsSz0bXh9GNdgxyNfRNFTi33eGOlix8rpioOe9eKjBdtXjrXptmPqA9iys7w6qgfJAyiFNfUCdKMo+dbzNVcqgWLU6t9vvfSZU6v9vDsYav4cifEMnQavRDXNlls+ax2A2xzWIMyWnRoDiJW2iEjRYtP3zu+tZLgel1Sm/UrUaKYHx9eeY2VHKquQSdUPDhdJkej6lo0N1Zc7BrxGEQpl1o+jUGMDnzhwSkAnjtdp9kftuR/4p5RagUbQ7v25u0vPjB53Y3cm0XC5YJF2mAj4GQbV2HvfcbNW22n/VA7JszIbiYhxE731todgCAenqSYry1vnRREieJy06dUsBgtQnMQkaY5e0eLfPrecebHi7xwdp1jYYdqwSXPc1Y2anZs08DYWMNwZrVPyTUZKzn80vwYH9td4z/+fHmj88lktuZxYqlNwdJxDB0nUVRcC9s0CJOUIM5IVM7BXVVSldEcxMM6l1hxvuHzvRcvMFZ0qBUtRoo2IwWTtp9QK1jUeyGDSJGojMmyQ70b0Q0T7psqs9QJudBcJ89hrTsMJwVreBX0n362jMpzDF3jzdU+902VKdoG5cIw4O0bL/LRPSPMXnHCsVmr89ZOqysDZLCx+6vpx4RJSsE2udQK2Dfu0QkTRkvD05fldsD5xoDlTsh01cXf2FHV8mNOLg+vB//2UsLzZ9b5yO4alqHT9pOtcQBFx+Bcvc/5ieLbJgGvdkOW2gFjRZuSa9LoRZj68GRnquLQGESUXItEqfdVtwPbbz/UjgkzTz75JE8++eTWbgchhNiJrlV3EybD05COnxClGa9ebKFrGvdNl656E49VjmPopFlOP1RMlmzSPEdlgAbjnss9E0WCRHF8scNwcbRGmA47rvaOJTw4VSZVGeQa6/2Q4RgcjYJtYJk6lq4zUbGp93I+c+8YXzk4Q9Wz+PevXCYjxzJ0dEPHTIct3v04o+GH6A1juE08UhQsHcsYzkH2LJ2ibRIUMkqOxVzN4c2V7nAGjqUTJorVTsaBiRJBnLLaiagVDFzH5Mxan6JtsH+yRNU1uW+qzKnVHscvdzi3Ptiq1TF07bqdVpsB8uRyl4tNn4prkagMQ9OIVUbNsxkvORyeH6MXxPzbly7zs8ttPNtE5cP/ftY6IW+u9nAtg7JrcX59wEqa8Yl7RgkixUonwLV0wljxV6fqKJWz3AmZrRVQWc5E2eHMWp+1XgjasJ5q71gRGNZYRWlGlsGFhk/ZtfjOfzv9vmporgx5bz0VulNPaHZMmBFCiLvV5gnCc6frHF0Y1m4c2lNDqRzPMfjl+yYY8eytGozWIGasbDM7WqDjx5xe7Q07kDwLTdO41PTRNTi0q8ZqLyIHPEfnUiNgpRuQ5bCw1sO1TR7ZVSOMM86v97FMnWrRYr0X45gGpqHz8sUWjqmT5RlZltPoRQRJRqKGE4v1LKMXAihUBoYGuakzO+JyuRWSAyXHJGNYl/Py+TZn1waofBgkvI2OpsVOgJ9k2KaGs7HZ27F0Fup95seHpxwnFjsEsaJcsFhsDLiw3idRGW+s9PE2BiKiaRxf7DA/Ubrqjfvnix0avRjb1AlThQF8ZK7KNw7NMVMrcGKpy7976SKvLXXo+AmjRYdUZbx6ocWBySJZtrFRPctJVI5rDXd+Fe3hHBmlMn5yqk6cZjw4W6EbDIflTVWGJzLLnYCqZ22F1gsM+OIDkyy2A1481+Dl8y2yHBKVM4jT91VDsx33Q0mYEUKIHWCuVuCXD0zQGsRMVlwqrkW4cTXimsPTnCuvqVp+zLFLbVqDmEfmqkxVXZ59s856P6bkmNimzmovxNB1ojTDMQx6UUKscg7vGyXPckZKNp87MEmW57xyoUU/TgljRc0dnkosrA9Y7gb0w2FXTq1gc6npk6jhfidtY7BdPhyfg6ZBrWjjGAbGRoHyaNGh5tnEqWIQpZxe65OTEat847VpWIaJH6dUChZBPCwYzrJhkbPK4bWlLo/MVVhphAw2Tm9Or3XoBhkqy1B5jm1ozI54HJgqc37dp+3HW8Pv/FgRRClxpvCTjJpnsdKJaPYTRos2UaL4dy9dZLEVUrRNVAar3ZCqazKIFb0gIUwVS92Y6UqGrmvUija9INk4AdOwLQ3PMXh4rkLFtSk7JqvdkI/M1bjYHHB0oUmSZZxY6rB3YwN5O4h55UKLdpBQsIdt+cudgANTJaJkuPi09Q4h5Ho1Mdtx1o2EGSGE2CFGivZWh5Bt6Nd8E9q8phop2szWClv1N36sWOmEHDR0Tq/1qffC4TbsSOFaOrMjDm8s9pmo2EzXCuwe9aj34603xI/tHeHxe8doDRJ+cmqNYxc7jJdtTi73iJVi32iRfbNVoo3amXo3IEgz8hxKrkE/yvAsDVPTma25FByDR2bG6IUpR883aQ8i+rHC1HUs0yROk61TjkSlOJZOpnIildELFbYBRddlrGgTJopuEPPyhSZNP6YfDFvFc9ia0ZJkGittnyBJudT02TXi8vkHppirFfBsg4JjYhk6JV2jF6SMFCzKnslyJ2C1E3FyuYdj6vSjFFOHREGzH7MexDQGEWNFBy2HJFV8ev8ocZrzyoUmORofv6eG76f4keJyM2TfuM5iO8A1df7y5Ap/dWqdei9itGgzXXG51PS5d7LEf31thZ9d7lDzLHRNw9J1VJZzZnWApuW0/Jjxks1qd3jCdWVoubImxrH0rS3qV9YKPXe6zqWmz2jJfl8rGj4MEmaEEGKHuJmBa5tff+WvbV5h7J8o0QkS5kY89o17FGwTS9eGhbqeg6Fpw2uUbLgeYXOx5mSlgLlxqjJetlnpRDT6MUGqcA2DuRGHRGXMVFyKrsm5tT6RyoiSDMfUKLkWRdvAMnRcU0cDjl1ubxTo2vhxwCBKsQydoq3Ti4b7pwq2jmfrNPwYgKpnYho6SZIRxoo4y6j3Y1KV0RpEBHE2DBwZWydEhj6s11F5QsmxuNj0r7qiefzAOCeXu6z3IyquRSeM+fmlDvVehMpA5Tm9MGEQp1sTgxOlGMQZOhquabBr1KPRj7jQDGj0Ii63Au6ZKPL6UpcwztA0aA1CVJ4xXRkGsZOrffIcxkoOLT+mNYjI8pzLLZ+ZmkfVtVjvxxQdgzTPqRYsumFCwTKZrjoYus6/fuE8+ydLW/89jRXtrZoYQ9d44WyD5880eHR3dWs7eNsf/nlp2ofxN/f9kzAjhBA7yLvtmrqet7Z+X7l0E7jqaqo5iNkz5vG1j84yUrCv+jmePTz1aQ2GtTh+/P9v795jrKrOxo9/9+3sc86cOefM/QaDDCCooFVRilrN+8pba321t7f51VBDL782tphq21htje0fjZW0SZO2ofaS1P5RK6mJ2tZYG35otfZFBAQUkIsCzghznzn3276s3x975pRBUKgDw6HPJyHRsxfbdR5k9pO113oeF1PXyFU8ntszgq4H+2pyFY+mWAhNU4QMk2Q06CheH7bIFl2yZYftb2coOB4hU8fQdRpjNkfGiyilcPygqrGmacRCBpoGrg9hE1piNmMFl9FChTmNdUFPpolKwnHbwnGDnlqGBhVv4kHoE6y82Ca2oZMtuYwX/vmKpqc5xv/9UA/P7x1ia+84hbJPWzwMQN9YnoaoRX+6jKcgZht4CvITG5yV7/Om52EZwSkxDYL2FJ7Pm4NZDg4HG5nrIxZtsaDf1c0Xd/La4TQRyyAZDVF0XHQg7ypaYiFMXQuSvIhFU12IdNHhA7MSdDREef1ImvZkhHjYYuuhMUZyFea3wOBE7ZtrFzSTK7kkoxY7+lJkSy6pQoVc2WVHX4q5zXXsH8yh6xpXzG1kLFvmT9sP86nLumiNn50nmiSZEUKIc8y/Up0Y3j0ROt6rqRO1AljW00Su7JKIhrgkGmI0X8ZxfUwzODpecRVdyTDdyTCxaIgL2+sZL1T4YE8Trx3J0DdW4Ei6SKHikik4QVNNXExDJxaxqLcNlIKC44EPrgLd84mYGr7SKFZ8dBT1tkljfXBk+vWBLPOa68iWPOq9oG2E8hVoPiFDx9Q1QqZB2NLpbo4wXnCwDR1d+2d7gp6WWPUIeMgwaI7ZpItOsKem4lNnBzFyPY8Dw4WJqr06ZR9yJYe3U0U0PWjGOZpziJg6o/kKYVMHTcM0dAxTo+L77B/KUh8xcbzgmHmp7FJ0ghYUF3YmGM1XGM8HqyfdDVE6EmEqns/6XYO8PV6gqd5mSVecN4dzeApe6RtnPF/BNnU8z8dHY09/lp2H0wxly1iGxqzGMHsHsrwxlMXUg9Npz+8ZIu8E+3729We55dIuls87+145STIjhBCi6r0SoZNJlLqSET52SSc6MJIv0zdqUnE9GmM23Y1RXn07jef7aIbOwvZ6hnMVhrIlNh4c5Y3BPAC6plFnm5Qcn3zFxfN9mmI2V/U0EYtYvDmYxTJ1yq7H/sEcSuks6gxzeLxE2fVQaERtk7fHi3Q3RohaBiXHZ15LHb1jRTTNpysRZWF7PWXXo2+sSNHxqbN18kWfZJ1Fa32YZ3YNUHb8aq2VproQXQ1ReseKDGdL9I0XMHWdZNQiV3ZJZ0q0xW0UCtdXlJyg9pmm6Xi+j0JjOFsGDYZyZZTno1sGuq7RWh8iXw46g/vArIYImzWCI/CGxvltceyQhusr8IP7e0rRNPH7Xu/PYBkarXGbvrEi/ekC6byDaQbNRnPloEVEtujiKp+hbImC4wIQC1scHC6QKzlEbZNE1CBdqPDWaJ54xKSxLsS+oRxrn9vPkVShup/obCHJjBBCiGmXiIb4jwvaeHH/ML2jRQxDZ15rDM9TXDW/qVrzJVdyGcqWaKkPMysZZfeRDIfHi8xvi3FkvERjXYgFbTFuvriDizoTtCUilB2PP+84QmGilYGhDaI0uG5+C3/fP0zveIn2uE2m5EwccVb0tNZRqgQ1WRa0xpjVGGHfQI6XDozRkQzTWGeTLbu4no+uB3tPntk1gK8Ul81pmNIi4ZoFLYwXKmw9OE7Z9Vg8K8Fl3Q0MZYps2DNIeaKppjvRqdvUteCIcywMmsZQuki27OH7waadkutiaDq9o3na4hGSUYuYbTKYKXNRV4IrzmtkJF8hbAWtJF7cN0Kq4BCPmnQkIlzUmeClN0exDI2GqE0iajGUCfbWxMJWkJQUXLobI5iGRu9YgXTJwdI1rpnfwu4jGcbyZVKFCtmSR7HiEbWCVaGK69MQCWEbJoRcSo5H3zH7ic4GkswIIYQ4LSY7hU92Bi87fnVTclcywgUdcY6kiqBBW32YsGWwdE4jQ5l+0gWXea111NkW81rq+M8L2qsPzrBl8B8XtFX391w5twk0DU/BRV0J0HTqwyadDWH6U6WgpYPjU3YVnufzwQUt7BvKkSkGe2d8D/YOZumIh7lsTpIth8Y5NFogZpsYusZbo3ku6kxUWwVAsFn6os566sYMOpJhkhGLYtllQUs9qZLDrMYI23vTFCrBykd3Y4Q626RQdoPihaZGXcim4is0TcNxXLJFl9Y4pAoOEctgOFemMxEhPLFv5vB4gbCh05awmd0YxTaCVZ5dR9LVV1LD2eD4edFxidkWpqERsYKaPYdTRaKFCiO5MvNbYrieT994kSvmJPl/e4fQNEhELXxfMZYvE4uYzG6MErZM8o7DUKaMoWkcHi8SMvSzqu6MJDNCCCFOm2M7gx+91yZsBc0oG6Khak0TXdP4r4vasE0DX6nqCZxjH5onau2ga/DMrgFGshXCIZ3X+tJErKBD+FiuQu9Ymb7xIuOFMo6naKizaIhZ9GdKVDyffQM5hrPBQzsa0jE1nUzRpT9VoiMZRp/opVQoe8xvi4OmM5QtMRgt0VRv83+unMOfdxxmJBdsknZdn5LrU3IUmVIRgJCp4/lQdl0SUYui4xOrCwWbnCNBf6UNrw8RtoL/9oLWGG8M5egdy5MqOOTKLt2NNroOZccnV3K59vwWhjJl/vHGCBXXx9B1dF2jMxnh4HAey9BwfbBNnYhlVhtwHhrN43k+TdEwHfEoDXUWA+kSY4Uys5JRVlzYysY3x/jHm1nyJZf6sMmW3nHeThX56MUdNNSFzuz/UCcgyYwQQojT7kR7bY5/nLydprrQe57IOl5rB4BrFrSw6cAoh8cLWKZOc8zGnHi4t9TbNMVC9KdL6LoWrDqUPRSKkWyZsBX0pio5QQ2csudjTRw1X9bThKeYUh13fmuM+rDJfy5spSMZrKK0xW3+940RhnJlBtMlBtLliS7nBvbE/SOmwcGRHJmSRyJsYVs6DVGLVMmhtT5MyNRJRkIMZ0tEQ0HC1JGIELEM/vfAKAeHc1imhmUYdDaE6UpG6GqIcO35LTTGQrw9VuDlQ2PsG8xhACHTYGFjhIZ6m7LjM5gpEbMtFnclaIiGGEiXKHs+nq9oqAsRtgw+0J1kxQXtLO5K0p8q0pcqYmgaugZFx2fjgTHOa46dFaszkswIIYSYUSc6RfWvPiQn75cqVLBMgz39maBlgK+4Ym4T/3NZF1fPL/LC/mH2DeQAuOK8JvrGCoQtg3ktMRxPUXY9ls5pYNm85mojyJLjvaM6bixsTplrT0useuorXazwzM4B6m2T1weypIsORcfD0DQWtNWTKlaIWiYKha+CasimDvGIycKOesbzFS7vbsA0dNrqwwxnS/zjjRHKnk9dOIRtmRxJlRgvVCg7PrMbo7i+ojUepjEaQkND0xWpgRyHM2UitslovkKm6NCVCDO/LUbENPF9hdI0+sbyZItBIb18xWPDniHmNtcFFYvLDsmoTcX18BXkytPXofv9kmRGCCHEjPtXj5O/2/3aExFuvqSThqjFeL5CQ12Iaxa0kIiGSERD9LTESBUqQXVcy6i+nupIhhnNlgmHDG65pJNENDTlvkevJPlKkSt7rN89OKW79OT3iYSMalXm1nqb3f0ZPE+xoC3Gf1/SSXvcZsuh8eoqTrrooNCY01RHuuDQUBdibksdB0bzDGfL+Aoilsmi9jAL2+vRNBjNVcgWXdKlCs/vGyZbcsgUK2i6zo0XtjOUK+P7GgPpInv6MxQqLgqNbX1pDozmsc3gNNWC1uBV4OBEN+5Y2MLzgkrCPc1R3h4P2jyETIPGWIjWevusaXGgKaXUTE9iOk12zU6n08Tj8ZmejhBCiBlWcryTKiJ4dIn/oxOTE90zVQiSh0LZm9LD6NhTPodTxWoTUMfzWdyVQNOgOWbz0SUdAKQKQWuI8YLDriNpskWXxliQfDXVhXi9P8P23hSZYpkX9o8SsQ3Ob62nd6yAqcMls5K89OYYB0fz2JZOxfWpuD5NsRBNdTbZssN4vkLF9XB9CBkaYwUHDYjaJrMbotXPihWPtriN4wcNMW1Dpy5kkK24pAsVGqJhrprfxH9f0jmtx7Pfz/NbVmaEEEKc00521edEr7uO15Bxsut12fHfs7v0uzUBLVY8Co5XTaIcz8fxFZoGjuezbzDD3oEs+wZyZEsOpqHR01pH31iwyuL6Cg2Nv+wcoOx6eCo4Dt5aH8b1ffIVj4MjOSKhYL6j+QqO62MYOr4PYSvoV6BpkK941NkGvh+88sqWXLIll+ZYiJCp01QXwjZ1LF0nZp8dKzKTzplkZu3ataxduxbP82Z6KkIIIWrUsYnPu63WnEp36RM1AZ08HRX0n7J4ft8wAEu6Emw6MMbb4wVQ0BIPM16oUHJ8Lp6V4Mo5SQ6OFmiJhdjWFxQhBPB9RarskoiEsAydnuYwY/kKDRGTfcM5dBQKKDseug6+b6ChkSo4NEQtChWfSMjA9TyKFZfOZISWejsomjeYoyUWJG6pgntW1ZrRZ3oC02X16tXs3r2bzZs3z/RUhBBC1LiS49GfDl4PpQsOjXWhauG8yaq+k/tnElHrPRt7nmjs0aejACxDQ9c0Do7k8XyfkuNSdn3yZafaO6rs+iSjNvmyx3C2wnjBoeT4hAyDZMRCQyNXdvFRjOXKjOYrpIoOUcukqT5M2NIJW3owvi5EJKTTELFY1BFndmOE9kSYzmQdizrifKA7SWM0xKGRAmXXQ0MRj5h0JMNT6u7MtHNmZUYIIYSYDpOrMYPpEvuHcnygO3nC10in0tjzeGOPPh2VjFrVU1SFSlAZueQqfN/FV4qIZeD4CtvUyZWdoDcV0Bq36R0toIDzW+tZ2tNIKu9QcFxKFY+muhC6BtpEi4h5LTHG8xV8IG5bJCIm1y1sJVV0mNUQoVDxQAtOVo1ky/h+kNyVKz5vp0rMb4uTLjgnXIWaCZLMCCGEEBNKE/tX0gWH9kSYvYNZXnlrnNj55gkf4KdyEut4tXEmT0dlSy5LuhKUHI9n9wxRcRUXdsTZP5Sj7Hp0NUSIhS3aE0HLg8WdCSqeR7rgYugahqazbF4TV85t5KUDo2gK9g5lqQsFtXTa4zaHRvN0JiOkYw7NMRvLCDqP/33/CL5SLJmV4K3RLADXnt9CxNB5YzjPhxY0Yxk6u/ozHBrJs6yn8YSrUDNBkhkhhBBiQqHiTSmKd1l3A9v7UgxlSrTGw6flAX7sis14vkK+7JIqOoDGFec1EA2Z/M9ls2hPhPEV6Bps2DPESK7MRR0Wo/ky9WGLWy7pxLYM9g5kGc2WsQ2dw6kibXGburDJVfObufb8FqKWQdHxeH7fMCPZMoYOmtI4MJxD1zQMPXjdFYtYZErBEfH2eIS2eJjBTInrzm+hPSGNJoUQQoizzrGbej1fsaynkWvPb6EhGjptKxFHr9g01IWY31bPSK5MImyRLgWrKHMmCvdNmlzRyZVc2hORiT05oSnXio6H6yuaY2GaYzbLepromEhCvHxQaK8zGWEs75AuVihWfEqOR8jUGc2V2d6XYixfYdOBMa6Y24jnBQX5ktGzo43BJKkzI4QQQhzlVOrNzPQc3q2GzuQ1XSMotnfMmJLj8fRr/aQLDoau8UrvOLqmcX57DA2NvQNZfKXoaanjwHAeXdNY1tPINQtaTks83s/zW5IZIYQQ4hgnW2iv1udwdNJkWzpLZiXpaa5jPF/hj9sPV+viZEsOg5kSH7+067S9XpKieUIIIcQ0mu72CmfrHE50GuvYujipgnNWvl6adM7UmRFCCCHEqQtbRrVT9tGfnWwNnbOBrMwIIYQQ4h1OpYbOTJNkRgghhBDHdTa8bjsZ8ppJCCGEEDVNkhkhhBBC1DRJZoQQQghR0ySZEUIIIURNk2RGCCGEEDVNkhkhhBBC1DRJZoQQQghR0ySZEUIIIURNk2RGCCGEEDVNkhkhhBBC1DRJZoQQQghR08653kxKKQAymcwMz0QIIYQQJ2vyuT35HD8V50wys3btWtauXUulUgFg9uzZMzwjIYQQQpyqbDZLIpE4pd+jqX8lBTqL+b7PkSNHqK+vR9O0abtvJpNh9uzZ9PX1EY/Hp+2+4vgk3meWxPvMk5ifWRLvM+tfibdSimw2S2dnJ7p+artgzpmVmUm6rjNr1qzTdv94PC5/Ec4gifeZJfE+8yTmZ5bE+8w61Xif6orMJNkALIQQQoiaJsmMEEIIIWqaJDMnybZtvve972Hb9kxP5d+CxPvMknifeRLzM0vifWad6XifcxuAhRBCCPHvRVZmhBBCCFHTJJkRQgghRE2TZEYIIYQQNU2SGSGEEELUNElmTtLatWs577zzCIfDLFu2jJdffnmmp1RzHnzwQa644grq6+tpbW3l4x//OHv37p0yplQqsXr1apqamojFYnzqU59icHBwypje3l5uuukmotEora2t3H333biueya/Sk1as2YNmqZx1113VT+TeE+vw4cP89nPfpampiYikQhLlixhy5Yt1etKKb773e/S0dFBJBJhxYoV7N+/f8o9xsbGWLlyJfF4nGQyyRe/+EVyudyZ/ipnPc/zuP/++5k7dy6RSIR58+bx/e9/f0pfH4n3+/PCCy9w880309nZiaZpPPnkk1OuT1d8X331VT70oQ8RDoeZPXs2P/zhD099skq8p3Xr1qlQKKR+85vfqF27dqkvfelLKplMqsHBwZmeWk254YYb1MMPP6x27typtm/frj760Y+q7u5ulcvlqmNuv/12NXv2bLVhwwa1ZcsW9cEPflBdddVV1euu66rFixerFStWqG3btqmnn35aNTc3q29/+9sz8ZVqxssvv6zOO+88dfHFF6s777yz+rnEe/qMjY2pOXPmqM997nNq06ZN6sCBA+qvf/2reuONN6pj1qxZoxKJhHryySfVjh071C233KLmzp2risVidcxHPvIRdckll6iXXnpJ/f3vf1fz589Xt95660x8pbPaAw88oJqamtRTTz2lDh48qB577DEVi8XUT37yk+oYiff78/TTT6v77rtPPf744wpQTzzxxJTr0xHfdDqt2tra1MqVK9XOnTvVo48+qiKRiPrlL395SnOVZOYkXHnllWr16tXVf/c8T3V2dqoHH3xwBmdV+4aGhhSgnn/+eaWUUqlUSlmWpR577LHqmNdff10BauPGjUqp4C+XrutqYGCgOuahhx5S8XhclcvlM/sFakQ2m1ULFixQ69evV9ddd101mZF4T6977rlHXXPNNSe87vu+am9vVz/60Y+qn6VSKWXbtnr00UeVUkrt3r1bAWrz5s3VMX/5y1+Upmnq8OHDp2/yNeimm25SX/jCF6Z89slPflKtXLlSKSXxnm7HJjPTFd+f//znqqGhYcrPk3vuuUctXLjwlOYnr5neQ6VSYevWraxYsaL6ma7rrFixgo0bN87gzGpfOp0GoLGxEYCtW7fiOM6UWC9atIju7u5qrDdu3MiSJUtoa2urjrnhhhvIZDLs2rXrDM6+dqxevZqbbrppSlxB4j3d/vSnP7F06VI+/elP09rayqWXXsqvf/3r6vWDBw8yMDAwJd6JRIJly5ZNiXcymWTp0qXVMStWrEDXdTZt2nTmvkwNuOqqq9iwYQP79u0DYMeOHbz44ovceOONgMT7dJuu+G7cuJFrr72WUChUHXPDDTewd+9exsfHT3o+51yjyek2MjKC53lTfpgDtLW1sWfPnhmaVe3zfZ+77rqLq6++msWLFwMwMDBAKBQimUxOGdvW1sbAwEB1zPH+LCavianWrVvHK6+8wubNm99xTeI9vQ4cOMBDDz3EN77xDb7zne+wefNmvva1rxEKhVi1alU1XseL59Hxbm1tnXLdNE0aGxsl3se49957yWQyLFq0CMMw8DyPBx54gJUrVwJIvE+z6YrvwMAAc+fOfcc9Jq81NDSc1HwkmREzYvXq1ezcuZMXX3xxpqdyzurr6+POO+9k/fr1hMPhmZ7OOc/3fZYuXcoPfvADAC699FJ27tzJL37xC1atWjXDszv3/OEPf+CRRx7h97//PRdddBHbt2/nrrvuorOzU+L9b0heM72H5uZmDMN4xwmPwcFB2tvbZ2hWte2OO+7gqaee4rnnnmPWrFnVz9vb26lUKqRSqSnjj451e3v7cf8sJq+Jf9q6dStDQ0NcdtllmKaJaZo8//zz/PSnP8U0Tdra2iTe06ijo4MLL7xwymcXXHABvb29wD/j9W4/S9rb2xkaGppy3XVdxsbGJN7HuPvuu7n33nv5zGc+w5IlS7jtttv4+te/zoMPPghIvE+36YrvdP2MkWTmPYRCIS6//HI2bNhQ/cz3fTZs2MDy5ctncGa1RynFHXfcwRNPPMGzzz77jqXFyy+/HMuypsR679699Pb2VmO9fPlyXnvttSl/QdavX088Hn/Hg+Tf3fXXX89rr73G9u3bq7+WLl3KypUrq/8s8Z4+V1999TtKDezbt485c+YAMHfuXNrb26fEO5PJsGnTpinxTqVSbN26tTrm2Wefxfd9li1bdga+Re0oFAro+tRHmGEY+L4PSLxPt+mK7/Lly3nhhRdwHKc6Zv369SxcuPCkXzEBcjT7ZKxbt07Ztq1++9vfqt27d6svf/nLKplMTjnhId7bV77yFZVIJNTf/vY31d/fX/1VKBSqY26//XbV3d2tnn32WbVlyxa1fPlytXz58ur1yaPCH/7wh9X27dvVM888o1paWuSo8Ek6+jSTUhLv6fTyyy8r0zTVAw88oPbv368eeeQRFY1G1e9+97vqmDVr1qhkMqn++Mc/qldffVV97GMfO+5R1ksvvVRt2rRJvfjii2rBggVyVPg4Vq1apbq6uqpHsx9//HHV3NysvvWtb1XHSLzfn2w2q7Zt26a2bdumAPXjH/9Ybdu2Tb311ltKqemJbyqVUm1tbeq2225TO3fuVOvWrVPRaFSOZp8uP/vZz1R3d7cKhULqyiuvVC+99NJMT6nmAMf99fDDD1fHFItF9dWvflU1NDSoaDSqPvGJT6j+/v4p9zl06JC68cYbVSQSUc3Nzeqb3/ymchznDH+b2nRsMiPxnl5//vOf1eLFi5Vt22rRokXqV7/61ZTrvu+r+++/X7W1tSnbttX111+v9u7dO2XM6OiouvXWW1UsFlPxeFx9/vOfV9ls9kx+jZqQyWTUnXfeqbq7u1U4HFY9PT3qvvvum3LEV+L9/jz33HPH/Zm9atUqpdT0xXfHjh3qmmuuUbZtq66uLrVmzZpTnqum1FHlEoUQQgghaozsmRFCCCFETZNkRgghhBA1TZIZIYQQQtQ0SWaEEEIIUdMkmRFCCCFETZNkRgghhBA1TZIZIYQQQtQ0SWaEEEIIUdMkmRFCCCFETZNkRgghhBA1TZIZIYQQQtQ0SWaEEEIIUdP+Pzy6/DbHgpwLAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "980 0.0029531973\n"
          ]
        }
      ],
      "source": [
        " \n",
        "#@title Training loop {vertical-output:true}\n",
        "TRAINING_ITTIRATION=\"1000\"#@param[500,1000,3000,5000,10000]\n",
        " \n",
        "for i in range(int(TRAINING_ITTIRATION)):\n",
        "  x=seed\n",
        "  loss,x=training_step()   \n",
        "  loss_log.append(loss.numpy())\n",
        "  if i%20==0:\n",
        "    clear_output(True)\n",
        "  \n",
        "    for n in range (1):\n",
        "      print(\"Batch :\",n,\"refrence layer 0,7,9\") \n",
        "      if flag:\n",
        "        temp=temp = x[n][...,:4]\n",
        "        imshow(zoom(np.hstack((temp[0], temp[7], temp[9]))))\n",
        "      else:\n",
        "        temp=temp = x[n][...,3]\n",
        "        imshow(zoom(np.hstack((temp[0], temp[7], temp[9]))))\n",
        "      \n",
        "    print()\n",
        "    pl.plot(loss_log,'.',alpha=0.3)\n",
        "    pl.yscale('log')\n",
        "    pl.show()\n",
        "    print(i,loss.numpy(),flush=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 76
        },
        "id": "yYAEUaCBQQvo",
        "outputId": "1c0d399d-8301-4e12-d497-201a5b18fc2a"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "step N :99\n"
          ]
        }
      ],
      "source": [
        "#@title Run the trained model\n",
        "x=seed\n",
        "ittiration=\"100\"#@param[100,200,300]\n",
        "for i in range(int(ittiration)):\n",
        "  x=ca(x)\n",
        "  clear_output(True)\n",
        "  temp = x[0][...,:4]\n",
        "  imshow(zoom(temp[6][...,:4][...,:4]))\n",
        "  print() \n",
        "  print('step N :'+str(i))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 450
        },
        "id": "JxpvUqZCbJ1v",
        "outputId": "f4520f8b-7620-4652-e94a-9075804713c3",
        "cellView": "form"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 500x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(19, 6, 19, 4)\n"
          ]
        }
      ],
      "source": [
        "#@title Final structure generated {vertical-output:true}\n",
        "fig = plt.figure(figsize=(5,5))\n",
        "ax=fig.add_subplot(projection='3d')\n",
        "temp = x[0][...,:4]\n",
        "max_v=np.max(temp)\n",
        "temp = np.where(temp < 0, 0, temp)\n",
        "temp=temp/max_v\n",
        "z_resized = np.pad(temp[...,3], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "r_resized = (np.pad(temp[...,0], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "g_resized = (np.pad(temp[...,1], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "b_resized = (np.pad(temp[...,2], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "\n",
        "obj=np.stack((r_resized,g_resized,b_resized,z_resized),axis=-1)\n",
        "\n",
        "ax=fig.add_subplot(projection='3d')\n",
        "xx = np.where(z_resized < 0.1, 0, z_resized)\n",
        "if flag:\n",
        "  ax.voxels(xx, facecolors=obj[...,:4])\n",
        "else:\n",
        "  ax.voxels(xx, facecolors=FACE_COLOR)\n",
        "\n",
        "\n",
        "plt.show()\n",
        "plt.close()\n",
        "print(target.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 447
        },
        "id": "0aRaZwakZddQ",
        "outputId": "610949ec-9088-411a-cb61-c747fa27fdd3"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f42c93e6770>]"
            ]
          },
          "execution_count": 20,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAi8AAAGdCAYAAADaPpOnAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABOx0lEQVR4nO3deVhU9eIG8PfMAMMOsoMsgii4ICoquWOaYpaaZqW2WGZqtpuZ97bfulZ227yW9rul2X4tW2yx3FfcUMQNBASRfZNhh2Hm/P4YmOKKCDjDmTPzfp6H59HhDLzzHGFez/kugiiKIoiIiIhkQiF1ACIiIqKOYHkhIiIiWWF5ISIiIllheSEiIiJZYXkhIiIiWWF5ISIiIllheSEiIiJZYXkhIiIiWbGROoCx6XQ65OXlwcXFBYIgSB2HiIiI2kEURVRWViIgIAAKRdvXViyuvOTl5SEoKEjqGERERNQJly5dQmBgYJvHWFx5cXFxAaB/8a6urhKnISIiovaoqKhAUFCQ4X28LRZXXppvFbm6urK8EBERyUx7hnxwwC4RERHJCssLERERyQrLCxEREckKywsRERHJCssLERERyQrLCxEREckKywsRERHJCssLERERyQrLCxEREckKywsRERHJCssLERERyQrLCxEREckKywsRERG1i7pGg9d+OYv1BzIlzWFxu0oTERGRcdU3avFZwkWs3pkOda0GrvY2mBkTCFd7W0nysLwQERFRq0RRxK+nCvDG1hRkl9UAACJ8XfC3KX0kKy4AywsRERG1IvFiGV775RyOZ5cDALxdVHh6Ym/cHhMEpUKQNBvLCxERERlcLK3GG1tT8OupAgCAg60SC8eGYcHoMDipzKM2mEcKIiIiktTl6gas3pmOzw5lQaMVoRCAO4YE4ambesPH1V7qeC2wvBAREVmx+kYtNh68iNU701BR1wgAiIvwxorJfRDh5yJxutaxvBAREVkhURSxJTkfb25NQc7lWgBApJ8L/j6lD0b38pY4XdtYXoiIiKzM0awyvPrLOZy8VA4A8HVV4emJEZgxOFDywbjtwfJCRERkJTJLqvHGbynYekY/GNfRTonFY3ti/uhQONrJpxLIJykRERF1Sll1A97fkYbPD11Eo04/GPeuYcF4YkIv+LiY12Dc9mB5ISIislB1Gi02HMzCmp3pqKzXD8a9MdIHKyZHopeveQ7GbQ+WFyIiIgu0K7UIL/10BhdL9Svj9gtwxd9v7oMR4V4SJ7t+LC9EREQWJOdyDV7ZchZ/nC0EoB+M+8ykSNw2qDsUMhiM2x4sL0RERBagvlGL/+zLxOqdaajT6GCjEPDAqFA8Nr4XnM1kZVxjsaxXQ0REZIX2pRXjxR/P4EJJNQAgNtQD/5jeH71lPK6lLSwvREREMpWvrsWrP5/DL6fyAQBezio8N6UPpg0MgCBYxi2i1rC8EBERyUxDow7rD2TivR1pqGnQQiEA943ogSdv6g1Xe1up45kcywsREZGMHMwowQs/nkF6URUAYEhIN7wyrT/6BrhKnKzrsLwQERHJQGFFHV775Rx+OpkHAPB0ssOKm/tghgXNImovlhciIiIzptHq8OnBLLy7PQ1V9Y1QCMDdN4Rg6U0RcHO0/FtErWF5ISIiMlNHMsvwwo+nkVJQCQAYGOSOV6f3R//ubhInkxbLCxERkZkprqzHyl/PYfOJXABAN0dbLI+PxB1DgqzuFlFrWF6IiIjMhCiK2JSYg9d+OQd1rQaCANw1NBjPTIpANyc7qeOZDZYXIiIiM5BdWoO/fX8K+9NLAOj3InrttigMDHKXNpgZYnnpgOr6RjhZ2BLLREQkrUatDusPZOFf21JRp9FBZaPAkzf1xoOjQmGjVEgdzyzxnbidKuo0GPjyHwj1ckJ0oDsGBLohOsgdffxdYW+rlDoeERHJ0Ln8Ciz/LhnJOWoAwPAwT6ycEYUeXk4SJzNvJi0vZWVlePTRR7FlyxYoFArMnDkT7733Hpydna/6nLi4OOzZs6fFYwsXLsTatWtNGfWaUvIroROBjOJqZBRXGwZR2SkVGNKjG+IivDG2tw96+zpb9JLMRER0/eo0WqzemYZ1ey6gUSfCxd4Gf7+5D+4cGsT3kHYQRFEUTfXFJ0+ejPz8fKxbtw4ajQb3338/hg4dii+//PKqz4mLi0Pv3r3xyiuvGB5zdHSEq2v7Vg6sqKiAm5sb1Gp1u5/TXqVV9UjOUeNkTjlOXipHco4apdUNLY7xd7PHjZE+uGVAAIaFekDJUeFERPQXR7PKsPy7ZFwo1m+iGN/PD69M6wcfV3uJk0mrI+/fJisv586dQ9++fXH06FEMGTIEALB161bcfPPNyMnJQUBAQKvPi4uLw8CBA/Huu+926vuasrz8L1EUkVlSjT3ni7E7tRiHLpSivlFn+Lyvqwp3x4ZgTmwwPJ1VJs1CRETmrU6jxb/+SMV/9mdCFAFvFxX+Ma0f4vv7Sx3NLJhFefnkk0+wdOlSXL582fBYY2Mj7O3tsWnTJtx2222tPi8uLg5nzpyBKIrw8/PDrbfeiueffx6Ojo6tHl9fX4/6+nrD3ysqKhAUFNQl5eV/1Wm0SLhQit9O5WPr6QJU1DUCAOxsFJgWHYD7R4Za1d4TRESkd/JSOZ76bxIymq62zIoJxHO39IWbg3WukNuajpQXk415KSgogI+PT8tvZmMDDw8PFBQUXPV5c+bMQUhICAICApCcnIzly5cjNTUVmzdvbvX4lStX4uWXXzZq9s6yt1ViXIQPxkX44NXpUfjtdD4+2Z+JkzlqbErMwabEHIzu5YVlkyIwINBd6rhERGRiDY06rN6Zhg92Z0CrE+HtosLrM6Iwvo+v1NFkrcPl5dlnn8Ubb7zR5jHnzp3rdKCHHnrI8OeoqCj4+/tj/PjxyMjIQM+ePa84fsWKFXjqqacMf2++8iI1OxsFpg3sjqnRATieXY71BzLx2+kC7Esrwb60Ekzq54ulEyPQ29dF6qhERGQC5/Ir8NR/T+JcfgUAYGp0AF6e2o+LzRlBh8vL0qVLMW/evDaPCQsLg5+fH4qKilo83tjYiLKyMvj5+bX7+8XGxgIA0tPTWy0vKpUKKpX5jicRBAExId0QE9IN2aU1eHfHefxwIhe/nynEH2cLcdvA7lgWHwF/NwepoxIRkRFodSI+2nsBb29LhUYropujLV6dHoUpAzi2xVg6XF68vb3h7e19zeOGDx+O8vJyJCYmIiYmBgCwc+dO6HQ6QyFpj6SkJACAv7/8T3qwpyPevmMgFo/tibe3ncdvpwuw+UQufjtdgEfHh2P+qFCobLhmDBGRXOWW1+Kpb5JwOLMMADCxry9euy0K3i7m+59sOTL5VOnCwkKsXbvWMFV6yJAhhqnSubm5GD9+PDZu3Ihhw4YhIyMDX375JW6++WZ4enoiOTkZTz75JAIDA69Y++VqunK20fVKzinHK1vO4thF/aDmUC8nvHhrX8RF+FzjmUREZG5+TMrFcz+cRmVdI5zslHhxaj/Mignkui3t1JH3b5OuO/zFF18gMjIS48ePx80334xRo0bho48+Mnxeo9EgNTUVNTU1AAA7Ozts374dEydORGRkJJYuXYqZM2diy5YtpowpmQGB7ti0aDjeviMa3i4qZJZUY976o1j42TEUVdRJHY+IiNqhok6DJ74+gce/TkJlXSMGBbvj18dH444hXHDOVEx65UUKcrry8leVdRq8vyMN6w9koVEnwtXeBs/d0petnYjIjB3JLMOT3yQht7wWCgF49MZeePTGcO5J1Almsc6LVORaXpqlFFTgmW//3OdiVLgXVs6IQpBH6+vcEBFR12to1OG9Hefx4e4M6EQg2MMR79w5EDEh3aSOJlssLzIuL4B+h9FPDmTiX3+cR32jDg62SiybFIH7RvTgdgNERBK7UFyFx79Owqlc/X8yb48JxEtT+8FZxb2OrwfLi8zLS7Oskmos/y7ZMGp9ULA7Vt0ejXCfq29sSUREprP5eA6e++E0ahq0cHOwxcoZUbg5Sv6zYc0By4uFlBcA0OlEfH30Ev756zlU1TfC3laB56b0xdzYYI6FISLqItX1jXjhxzP47ngOACA21APv3jWQa3QZEcuLBZWXZvnqWjzzbTL2pZUAACb08cEbMwdww0ciIhM7m1eBR746jgvF1VAIwOPje+ORG8N5G9/IWF4ssLwA+qsw6w9m4Y3fUtCg1cHLWYW3Zg3gujBERCYgiiI+P3QR//jlHBoadfBztce7dw3EDWGeUkezSCwvFlpemp3Lr8DjX5/A+cIqAMC8ET3w7ORI2NtydV4iImNQ12rw7HfJ+O20fiPh8ZE+WDUrGh7cl8hkWF4svLwAQJ1Gi9d/S8GGg1kAgN6+znh/9iBE+lnuayYi6gpn8tR4+IvjuFhaA1ulgGcn98EDI3twnKGJsbxYQXlptiu1CMs2JaOkqh4qGwWeu6Uv7uZgXiKiTtl07BKe++E06ht1COzmgA/mDsaAQHepY1kFs9kegExvXIQPtj4xGuMivFHfqMPzP5zGos8TUV7TIHU0IiLZqNNosWJzMpZ9m4z6Rh3GRXjj50dHsbiYKZYXC+DlrMLH9w3Fc1P6wFYp4Pczhbj5vX040rQ+DBERXd2lshrMWpuAr45cgiAAS2/qjY/vGwp3R45vMVcsLxZCoRDw4OgwbF48Ej08HZGnrsNdHyXgve1p0Oos6s4gEZHR7Eotwi2r9+NUrhrdHG3x6f3D8Oj4XlBwGrRZY3mxMFGBbvj5sdGYMbg7dCLwzvbzmP1/h5CvrpU6GhGR2dDpRLyz7Twe2HAU6loNopt+d47p7S11NGoHlhcL5Kyywdt3DMQ7d0bDyU6JI5llmPzePvxxpkDqaEREkiuvacD9G47ivR1pEEXg7huC8d9Fw9HdnavlygVnG1m4zJJqPPbVCcMGYg+MDMWzkyNhZ8PeSkTWJ72oEvM/PYaLpTWwt1Xgn7dFYcbgQKljETjbiP4i1MsJ3y0egQdHhQIAPjmQibs+SuBtJCKyOrtSi3DbmoO4WFqDwG4O2Lx4JIuLTLG8WAG7pvVfPronBi72NjieXY4p7+/HvrRiqaMREZmcKIr4v70XMH/DUVTWN2JYqAd+XDISfQN4dV6uWF6syMR+fvjl0dHoF+CKsuoG3PvJEby3PQ06zkYiIgtV36jFsm+T8dqv56ATgbuGBuHz+bHc1FbmWF6sTLCnI75bPAKzhwVBbJqNNG/DUZRVc1E7IrIsxZX1mPN/h/FtYg4UAvDirX2xckYUx/xZAJ5BK2Rvq8TKGQPw1qxo2NsqsPd8MW55fx9OZF+WOhoRkVGczavA9DUHkHjxMlzsbbDh/mG4f2Qot06xECwvVuz2mED8sGQkQr2ckKeuwx3rErDhQCYsbAIaEVmZracLMPPDg8gtr0WYlxN+WDKS67dYGJYXKxfp54qfHhmJm6P8oNGKeGnLWTz61QlU1TdKHY2IqENEUcSaXelY9HkiajVajO7lhe8fHome3s5SRyMjY3khuNjbYs2cwXj+lr6wUQj4OTkfU/+9H+cLK6WORkTULg2NOiz7Nhmrfk8FAMwb0QPr5w2Fm6OtxMnIFFheCAAgCALmjwrFNwtvgJ+rPS4UV2P6mgP49VS+1NGIiNpUXtOAez7WD8xVKgT8Y3p/vDS1H2yUfIuzVDyz1EJMiAd+eWwURoZ7oqZBi4e/OI43t6Zwc0ciMktZJdWY8cFBHM4sg7PKBh/fNwT33BAidSwyMZYXuoKnswqf3j8MC0brV+X9YHcG5n96FOoajcTJiIj+dCSzDNM/OIALJdXo7u6AbxcPR1yEj9SxqAuwvFCrbJQK/H1KX7x310DY2yqwO7UYU9fsR2oBx8EQkfS+P5GDu/9zGOU1+h2hv18yApF+XDHXWrC8UJumDeyObxeNQHd3B1wsrcFtHxzAbxwHQ0QSaZ5R9OQ3J9Gg1WFyfz98/dBw+LjYSx2NuhDLC11T/+5u2PLoKIzoqR8Hs/iL41j1O8fBEFHX0ulEvPrLOcOMooVjwrBmzmA42CklTkZdjeWF2sXDyQ4bHxhm2J16za6mcTC1HAdDRKan0erw9KaT+Hh/JgDguSl9sOLmPlAouGKuNWJ5oXazUep3p373zoFQ2ejHwUzjejBEZGK1DVos/CwRm0/kQqkQ8K9Z0XhwdJjUsUhCLC/UYdMHdcd3i/XjYLJKazB9zQFsPc1xMERkfOoaDe75+DB2phRBZaPAR/fEYGZMoNSxSGIsL9QpzeNghofpx8Es+vw43vo9leNgiMhoiirqcOdHCTh28TJc7W3w+YOxGN/HV+pYZAZYXqjTPJzs8Nn8YZjfNA7m37vS8eCnR1FRx3EwRHR9skqqMXPtQaQUVMLbRYVvFg7H0B4eUsciM8HyQtfFRqnA87f0xTt3RkNlo8Cu1GLc3rSbKxFRZ5zJU+P2tQm4VFaLEE9HfLdoBPr4cw0X+hPLCxnFbYMC8e2iEfBxUeF8YRVuW3MAp3PVUsciIpk5klmGu9YdQklVPfr4u2LTouEI9nSUOhaZGZYXMpqoQDd8v2Qkevs6o6iyHneuS8Cu1CKpYxGRTOxOLcK9nxxGZX0jhoV64JuFN3DxOWoVywsZVXd3B2xaNAIjenqiukGLBz89hi8PZ0sdi4jM3G+n8rFg4zHUaXQYF+GNjQ8Mg6u9rdSxyEyxvJDRuTnYYsP9wzBjcHdodSL+9v0prPo9BaLImUhEdKVvE3Ow5Mvj0GhFTInyx7p7hsDelqvm0tWxvJBJ2Nko8K9Z0Xh8fC8A+hV5n/gmCQ2NOomTEZE52ZiQhac3nYROBO4YEoj3Zw+CnQ3fmqht/BdCJiMIAp68qTdW3T4ANgoBPybl4cGNx1DT0Ch1NCIyA//ZdwEv/HgGAHD/yB54fcYAKLncP7UDywuZ3KwhQfhk3lA42Cqx93wx7v34CPdEIrJy/7f3Al795RwA4OG4nnjhlr7cp4jajeWFusSY3t74/MFhcLW3wbGLlzH7o0MorqyXOhYRSWDtngy89qu+uDx2YziWTYqAILC4UPuxvFCXiQnxwDcLh8PLWYWz+RW4Y10CF7MjsjJrdqXj9d9SAABPTOiFpyayuFDHsbxQl2pedKq7uwMyS6px+4cHkV5UJXUsIuoC/96ZhlW/pwIAnrqpN56Y0FviRCRXLC/U5UK9nPDt4uHo6e2EfHUd7liXwNV4iSzce9vT8NYf5wEAyyZF4LGmmYhEncHyQpLwd3PAfxcOR1R3N5RVN2D2R4dwJLNM6lhEZALvbj+Pd7bri8vy+EgsGRcucSKSO5YXkoynswpfLojFsFAPVNY34p6PD2NXCrcTILIka3al493taQCAFZMjsTiup8SJyBKwvJCkXOxtsfGBYbgx0gf1jTos2HgMv58pkDoWERnBR3szDGNcnp0ciYVjWVzIOFheSHL2tkqsuycGU6MD0KgT8ciXx7H9bKHUsYjoOqw/kIl//qqfVbT0pt5YxOJCRsTyQmbBVqnA23dE49boAGi0Ih7+4jhvIRHJ1GeHLuLlLWcBAI/eGI5HOTiXjIzlhcyGjVKBd+6IxpQofzRodVj4eSL2nC+WOhYRdcD3J3Lw/A+nAQALx4bhqZs4HZqMj+WFzIqNUoF37xqISf180dCow0Mbj2F/WonUsYioHXalFmHZpmQAwLwRPfBsfCQXoCOTYHkhs2OrVGD17MG4qa8v6ht1eHDjURzMYIEhMmfHsy/j4c+Po1En4rZB3fHCLX1ZXMhkWF7ILNnZKLBmzmCMj/RBnUaH+RuO4fCFUqljEVEr0osq8cCGo6jVaDG2tzfevH0AN1kkk2J5IbNlZ6PAB3cPRlyEN2o1Wty/4ShOZF+WOhYR/UVeeS3u/fgIyms0iA5yx4d3D4atkm8tZFr8F0ZmTWWjxNq7YzC6lxdqGrSYt/4oUgoqpI5FRADKaxpw3ydHkKeuQ5i3E9bPGwpHOxupY5EVYHkhs9e8DkxMSDeoazW45+MjyCqpljoWkVWrbdDigQ1HkVZUBT9Xe3w2PxYeTnZSxyIrwfJCsuBoZ4NP5g1FH39XFFfWY+5/DiNfXSt1LCKrpNHqsOTL4zieXQ43B1tsnD8M3d0dpI5FVoTlhWTDzUG/lUColxNyy2tx938Oo7SqXupYRFZFFEU8+90p7Ewpgr2tAp/MG4Levi5SxyIrw/JCsuLtosLnD8YiwM0eGcXVuG/9EVTUaaSORWQ1Xt+agu+O50CpELBmzmDEhHhIHYmsEMsLyU53dwd89mAsPJ3scDq3AvM3HEVtg1bqWEQW7z/7LmDdngsAgNdnRGF8H1+JE5G1YnkhWerp7YyN84fBxd4GR7MuY/EXidBodVLHIrJYm4/n4NVfzgHQ7xA9a0iQxInImpmsvLz22msYMWIEHB0d4e7u3q7niKKIF154Af7+/nBwcMCECROQlpZmqogkc/0C3LDh/qFwsFVid2oxVmw+BVEUpY5FZHF2pxbhmW/1y/4/OCoUC8eESZyIrJ3JyktDQwNmzZqFxYsXt/s5b775Jt5//32sXbsWhw8fhpOTEyZNmoS6ujpTxSSZiwnxwJq5g6BUCPg2MQf/+uO81JGILEpqQSUe+fKEYdn/v93ch8v+k+RMVl5efvllPPnkk4iKimrX8aIo4t1338Vzzz2HadOmYcCAAdi4cSPy8vLwww8/mComWYAbI33x2vT+AIB/70rHZ4cuSpyIyDKUVNXjgQ1HUVXfiBvCPPDGTC77T+bBbMa8ZGZmoqCgABMmTDA85ubmhtjYWCQkJFz1efX19aioqGjxQdbnrmHBeHJCbwDACz+extbTBRInIpK3Oo0WD208htzyWoR6OWHt3TGwszGbtwyycmbzL7GgQP9m4+vbcvS6r6+v4XOtWblyJdzc3AwfQUEcRGatHhsfjtnDgiGKwONfn0DixTKpIxHJkiiKWP5dsmERuo/vGwJ3R66eS+ajQ+Xl2WefhSAIbX6kpKSYKmurVqxYAbVabfi4dOlSl35/Mh+CIOAf0/phQh9f1DfqsGBjIrJLa6SORSQ7q3em48ekPNgoBHw4dzDCvJ2ljkTUQod20Fq6dCnmzZvX5jFhYZ0bhe7n5wcAKCwshL+/v+HxwsJCDBw48KrPU6lUUKlUnfqeZHlslAq8P3sg7lx3CKdy1bh/wxFsfngk3BxspY5GJAs/J+fh7W36ge+vTu+PEeFeEiciulKHyou3tze8vb1NEiQ0NBR+fn7YsWOHoaxUVFTg8OHDHZqxRORoZ4P/3DcE09ccQEZxNR7+IhEb7h8GW6XZ3CUlMktJl8qx9L8nAeinRN81LFjiREStM9lv8+zsbCQlJSE7OxtarRZJSUlISkpCVVWV4ZjIyEh8//33APSX/J944gm8+uqr+Omnn3Dq1Cnce++9CAgIwPTp000VkyyUr6s9Pr5vKJzslDiQXornfzjNNWCI2pBbXosHPz2G+kYdxkf6YMXNfaSORHRVHbry0hEvvPACPv30U8PfBw0aBADYtWsX4uLiAACpqalQq9WGY5555hlUV1fjoYceQnl5OUaNGoWtW7fC3t7eVDHJgvUNcMXqOYPw4KfH8PXRSwjzdsJDY3pKHYvI7FTVN2L+hqMoqapHpJ8L3putXzuJyFwJooX9d7SiogJubm5Qq9VwdXWVOg6ZgfUHMvHylrMQBODDuTGI7+8ndSQis6HViVj42TFsP1cEL2cVfnxkJLq7O0gdi6xQR96/OQiALN68ET1w7/AQiCLwxDcnkJxTLnUkIrPx5u8p2H6uCCobBf7v3hgWF5IFlheyeIIg4IVb+mJsb2/UaXSY/+kxFKi55QTRb6fyDbtEvzUrGoOCu0mciKh9WF7IKtgoFfj3nEGI8HVBcWU9Fn2eiPpGrdSxiCSTXlSFZU2bLS4cE4ZbowMkTkTUfiwvZDVc7G3x0b0xcLW3QdKlcrz001mpIxFJorq+EYs+TzTsWbRsUoTUkYg6hOWFrEqIpxPenz0IggB8dSQbXx/JljoSUZdqXvo/vagKvq4qrJ49GDZcA4lkhv9iyerERfhg6U3NmziewYnsyxInIuo6nxzIws/J+bBRCPhg7mB4u3CFcpIflheySg/HhWNiX180aHVY/PlxFFfWSx2JyOQSL5Zh5a/nAADPTemDmBAPiRMRdQ7LC1klhULAv+6IRk9vJxRU1OGRL4+jUauTOhaRyZTXNODRL0+gUSfi1ugA3Deih9SRiDqN5YWslou9LdbdMwROdkoczizDu9vTpI5EZBKiKOLpTcnIU9ehh6cjVs6IgiBwBV2SL5YXsmrhPs54feYAAMC/d6VjV2qRxImIjG/DwSxsP1cIO6UC/54zGM4qk+0MQ9QlWF7I6t0aHYB7bggBADz1TRLyymslTkRkPMk55fhn0ziXv0/pg/7d3SRORHT9WF6IADx3Sx9EdXfD5RoNHvnyODQc/0IWoLJOg0e/OgGNVsSkfr64d3iI1JGIjILlhQiAykaJNXMGw8XeBsezy/Hm1hSpIxFdF1EUsWLzKVwsrUF3dwe8OTOa41zIYrC8EDUJ9nTEqtujAQD/ty8TO84VSpyIqPO+PnrJsJ7L6jmD4OZoK3UkIqNheSH6i/j+frh/ZA8AwLJvk1FUyQ0cSX5SCirw0k9nAADLJkVgMDdcJAvD8kL0P5bHRyLSzwVl1Q14elMydDpR6khE7Van0eKRL0+gvlGHuAhvLBgdJnUkIqNjeSH6H/a2SqyePQgqGwX2ni/G+oNZUkciardVv6civagKPi4q/GtWNBQKjnMhy8PyQtSKXr4ueO6WvgCAN35Lwdm8CokTEV3boQul+ORAJgDgjdsHwNOZ+xaRZWJ5IbqKu2ODMaGPfv+jx78+gTqNVupIRFdVVd+IpzedhCgCs4cFYVyEj9SRiEyG5YXoKgRBwBszo+DtokJaURVe++Wc1JGIrur1384h53Iturs74O9T+kodh8ikWF6I2uDprMLbd+inT3926CK2neX0aTI/BzNK8PmhbADAqtsHcPl/sngsL0TXMLqXNx4cFQoAWP5dMooqOH2azEdNQyOe/e4UAGBObDBGhHtJnIjI9FheiNphWXwE+vq7oqy6AUs3neT0aTIbb25NRXZZDQLc7LFicqTUcYi6BMsLUTuobJR4f/ZA2NsqsC+tBF8cyZY6EhGOZpXh04QsAMDKmQPgYs9VdMk6sLwQtVO4jwuWx+v/Z7vy13PILq2ROBFZs9oGLZ75NhmiCNwxJBBje3tLHYmoy7C8EHXAfcN7YFioB2oatFj2LW8fkXTe2X4emSXV8HVVcXYRWR2WF6IOUCgEvHV7NBztlDicWYaNTZfsibrSiezL+M++CwCAf94WBTcH3i4i68LyQtRBwZ6OhoGRr29NQVZJtcSJyJrUabRY9m0ydCJw26DuGN/HV+pIRF2O5YWoE+bGhmBET0/UaXR4etNJaHn7iLrI+zvSkF5UBS9nFV68lbeLyDqxvBB1gkIh4I2ZA+Bkp8Sxi5exgZs3Uhc4m1eBdXv1t4tend4f7o52EicikgbLC1EnBXk44m9T+gAA3vo9FZfKOPuITEenE/HcD6eg1YmY3N8P8f39pI5EJBmWF6LrMHtoMIaFeqBWo8XffzgNUeTtIzKNb45dwvHscjjZKfHirf2kjkMkKZYXouugUAhYOSMKdjYK7D1fjB+T8qSORBaotKoer/+WAgB4amIE/NzsJU5EJC2WF6Lr1NPbGY/dGA4AeOXnsyirbpA4EVmaf/6aAnWtBn39XXHf8BCp4xBJjuWFyAgeGtMTkX4uKKtuwKs/n5U6DlmQwxdK8d3xHAgC8Opt/WGj5K9tIv4UEBmBnY0CK2dEQRCAzSdysfd8sdSRyAI0NOrw3A+nAQCzhwVjcHA3iRMRmQeWFyIjGRTcDfNG9AAA/P2HU6ht0EobiGTvkwOZSCuqgqeTHZZP4o7RRM1YXoiM6OmJEeju7oBLZbX4YHe61HFIxgor6vD+jjQAwIqb+8DNkVsAEDVjeSEyIieVDZ6/Rb/q6bo9F7h1AHXaG7+loKZBi0HB7pgxqLvUcYjMCssLkZFN6ueLMb290aDV4aUtZ7j2C3XY8ezL2HwiF4IAvHRrPygUgtSRiMwKywuRkQmCgJdu7QtbpYDdqcXYdrZQ6kgkIzqdiJd/OgMAmBUTiOggd2kDEZkhlhciEwjzdsaC0WEA9Gu/1Gk4eJfa59vjOTiZo4aLygbLOEiXqFUsL0Qm8siN4Qhws0fO5Vp8sDtD6jgkAxV1Gry5Vb+S7mPje8HbRSVxIiLzxPJCZCKOdjZ4rmnw7to9GbhYysG71LbVO9JQUtWAMG8n3Nc07Z6IrsTyQmRCk/v7YVS4FxoadfjHz+ekjkNmLKO4CusPZAEAnr+lL+xs+OuZ6Gr400FkQoIg4KWp/WCjELD9XCEOppdIHYnM1D9+PotGnYgbI30wLsJH6jhEZo3lhcjEwn2ccfcN+s30Xvn5LLQ6Tp2mlnamFGJ3ajFslYJhnSAiujqWF6Iu8Pj4XnC1t0FKQSU2HbskdRwyI/WNWsMtxQdGhSLUy0niRETmj+WFqAt0c7LDY+N7AQDe+uM8quobJU5E5uKzhIvILKmGt4sKj97YS+o4RLLA8kLURe4d3gOhXk4oqarHh9z3iACoazX49y79v4VlEyPgrLKROBGRPLC8EHUROxsFVkzWLzr2f/sykXO5RuJEJLV1ezJQXqNBLx9nzIwJlDoOkWywvBB1oZv6+mJ4mCcaGnV4Y2uq1HFIQgXqOnxyIBMA8Ex8JJTcv4io3VheiLqQIAh47pY+EARgy8k8JF68LHUkksh7O86jTqPDkJBumNCHU6OJOoLlhaiL9Qtww6ymWwT/+PksdJw6bXXSi6rwzVH9rLNnJ0dCEHjVhagjWF6IJPD0xAg42imRdKkcW5LzpI5DXWzV7ynQifrbiEN6eEgdh0h2WF6IJODjao+H43oCAN74LYW7TluRxIuX8fuZQigE4JlJEVLHIZIllhciiTw4OgwBbvbI+8vATbJsoijijaZdo2fFBKGXr4vEiYjkieWFSCL2tko83fQ/73V7LkBdq5E4EZnawYxSHMksg52NAk/cxAXpiDqL5YVIQtMGdkcvH2eoazX4eD+vvlgyURTx7vbzAIA5w4Lh7+YgcSIi+WJ5IZKQUiHgqZt6AwA+2Z+JsuoGiRORqSRklOJo1mXY2SiwaGxPqeMQyRrLC5HEJvXzQ78AV1TVN2Ldngyp45AJ6K+6pAEAZg8Ngp+bvcSJiOSN5YVIYgqFgKUT9VdfPk3IQlFFncSJyNgSLpTiSFYZ7JQKLIrjVRei62Wy8vLaa69hxIgRcHR0hLu7e7ueM2/ePAiC0OIjPj7eVBGJzMa4CB8MDnZHnUaHD3bz6oulea/pqstdw4I41oXICExWXhoaGjBr1iwsXry4Q8+Lj49Hfn6+4eOrr74yUUIi8yEIAp6eqJ959OXhbOSW10qciIwlIaMUhzP1V10W86oLkVGYbP/1l19+GQCwYcOGDj1PpVLBz8/PBImIzNuIcC+M6OmJgxmlWL0jDa/PHCB1JDKC93boZxjdOZRXXYiMxezGvOzevRs+Pj6IiIjA4sWLUVpa2ubx9fX1qKioaPFBJFfNY182JeYgu7RG4jR0vQ5dKMWhC2WwVQq86kJkRGZVXuLj47Fx40bs2LEDb7zxBvbs2YPJkydDq7360ukrV66Em5ub4SMoKKgLExMZV0yIB8b09oZWJ+LDPelSx6Hr1DzW5c6hQQhw51UXImPpUHl59tlnrxhQ+78fKSkpnQ5z1113YerUqYiKisL06dPx888/4+jRo9i9e/dVn7NixQqo1WrDx6VLlzr9/YnMwePjwwEA3ybmIOcyr77I1eELpUi4UNp01SVc6jhEFqVDY16WLl2KefPmtXlMWFjY9eS54mt5eXkhPT0d48ePb/UYlUoFlUpltO9JJLWYEA+MDPfEgfRSrN2TgVenR0kdiTrhvR36qy6zhgShO6+6EBlVh8qLt7c3vL29TZXlCjk5OSgtLYW/v3+XfU8ic/Dojb1wIL0U/z2agyXjwjnQU2aOZJbhYIb+qsvDHOtCZHQmG/OSnZ2NpKQkZGdnQ6vVIikpCUlJSaiqqjIcExkZie+//x4AUFVVhWXLluHQoUPIysrCjh07MG3aNISHh2PSpEmmiklklm4I88SwUA80aHVYt+eC1HGog1bv1F91uT0mCIHdHCVOQ2R5TFZeXnjhBQwaNAgvvvgiqqqqMGjQIAwaNAjHjh0zHJOamgq1Wg0AUCqVSE5OxtSpU9G7d2/Mnz8fMTEx2LdvH28LkVV6fLx+1+Evj2Rz1V0ZOZ2rxr60EigVvOpCZCqCKIqi1CGMqaKiAm5ublCr1XB1dZU6DlGniaKI29cmIPHiZcwfFYrnb+krdSRqh8e+OoGfTuZh2sAAvHfXIKnjEMlGR96/zWqqNBH9SRAEPHqjfpbKF4cvoqSqXuJEdC2Xymrwy6l8AMBDY4w3eYGIWmJ5ITJjY3t7IzrQDXUaHf6zL1PqOHQNH+/PhFYnYnQvL/QLcJM6DpHFYnkhMmOCIGDJuKarL4cuorJOI3EiuprymgZ8c1S/ztSisRzrQmRKLC9EZm5CH1/09HZCZX0jvjqSLXUcuoqvjlxCrUaLPv6uGNHTU+o4RBaN5YXIzCkUAhaO0f9P/uP9mWho1EmciP6XRqvDxoQsAMADI3tAEARpAxFZOJYXIhmYNigAPi4qFFbU48ekXKnj0P/4/UwB8tV18HK2w63RAVLHIbJ4LC9EMqCyUeKBUaEAgI/2XoBOZ1ErHMje+gNZAIA5sSGwt1VKG4bICrC8EMnEnNhguKhskFZUhZ0pRVLHoSZJl8qRePEybJUC7r4hWOo4RFaB5YVIJlztbTGn6c1x3d4MidNQs/UH9FPYbx0QAB8Xe4nTEFkHlhciGXlgZChslQKOZl1G4sUyqeNYvcKKOvySrF+U7v6RoRKnIbIeLC9EMuLrao/bBnUHAG7YaAY+S7iIRp2IoT26ISqQi9IRdRWWFyKZaV52/o+zhcgqqZY4jfWq02jxxeGLAPRXxIio67C8EMlMuI8LxkV4AwA2JlyUOI31+jEpF5drNOju7oCb+vpKHYfIqrC8EMnQvSN6AAA2JV5CdX2jtGGskCiKhuJ47/AQ2Cj5q5SoK/EnjkiGxvbyRg9PR1TWNeL7E1y0rqsl56hxJq8CdjYKzBoSJHUcIqvD8kIkQwqFgHuG9wAAbEzIgihy0bqu9Pkh/VWXKVH+8HCykzgNkfVheSGSqdtjAuFop8T5wiokXCiVOo7VUNdosCU5DwAwN5aL0hFJgeWFSKbcHGwN06Y/PZglbRgrsvlEDuo0OkT6uSAmpJvUcYisEssLkYzd1zRwd9vZQuSW10obxgqIoogvDmcD0F914e7RRNJgeSGSsd6+LhjR0xM68c9xGGQ6hzPLkF5UBUc7JaY3XfUioq7H8kIkc/c2Ddz9+kg26jRaacNYuOarLtMGdoeLva3EaYisF8sLkcxN6OOD7u4OuFyjwZaTeVLHsVglVfXYelq/jxEH6hJJi+WFSOZslArMbdptuvnKABnfpmM50GhFRAe5o3937mNEJCWWFyILMCsmCDYKAUmXypFSUCF1HIuj04n48oh+TBGvuhBJj+WFyAJ4u6gwoY9+f52vj1ySOI3lSbhQiktltXCxt8GtAwKkjkNk9VheiCzEXcP0y9R/fyKXA3eN7NvEHADA1OgAONgpJU5DRCwvRBZidC9vdHd3gLpWg62nC6SOYzEq6jT4rWmgLvcxIjIPLC9EFkKpEDBrSCAA4KsjHLhrLL8m56NOo0O4jzOiAzlQl8gcsLwQWZA7hgRBIegXU7tQXCV1HIvQfMvo9phArqhLZCZYXogsSIC7A8b29gYAfHOUA3ev14XiKhy7eBkKAYZ9pIhIeiwvRBbmrmH6qbzfJuagoVEncRp5++64/qrL2N7e8HW1lzgNETVjeSGyMDdG+sDbRYXS6gbsOFcodRzZ0upEbD6eCwC4PYYDdYnMCcsLkYWxVSowK6Zp4C5vHXXawYwS5Kvr4OZgi/F9fKSOQ0R/wfJCZIHuHKq/UrAvrRg5l2skTiNPm47pbxlNGxgAe1uu7UJkTlheiCxQiKcTbgjzgCgCPyZxs8aOUtdq8PsZ/Vo5tzddxSIi88HyQmShZgzWv+l+dzwHoihKnEZefknOR32jDr19nRHFTRiJzA7LC5GFmtzfD/a2ClworkZyjlrqOLLywwn9QN2Zg7m2C5E5YnkhslAu9raY2NcPALC5acovXVu+uhZHL5YBAG6N5iaMROaI5YXIgs0YrF9Y7aeTeVzzpZ1+Sc6HKAJDe3RDgLuD1HGIqBUsL0QWbFS4F7xdVLhco8Hu1CKp48jClmT9Joy86kJkvlheiCyYjVKBqU1vwj+e5Kyja7lUVoOTl8qhEIDJ/f2ljkNEV8HyQmThpg3Ul5cd5wpRXd8ocRrztiVZX/CG9/SEt4tK4jREdDUsL0QWLqq7G3p4OqJOo8O2s9wuoC0/n9TfMrplAG8ZEZkzlhciCycIguHW0U+8dXRV6UVVOJtfARuFgPh+flLHIaI2sLwQWYGpTbeO9p4vxuXqBonTmKefm24ZjerlhW5OdhKnIaK2sLwQWYFwHxf08XdFo07Eb6cLpI5jdkRRxJamq1K38pYRkdljeSGyEn/eOsqVOIn5SSmoREZxNeyUCtzUz1fqOER0DSwvRFbi1mj91N/DmWUoUNdJnMa8NN8yiovwhqu9rcRpiOhaWF6IrERgN0cMCekGUfzzzZqabxk1zTLiwnREssDyQmRFmgfuctbRn07nViC7rAb2tgpM6OMjdRwiageWFyIrcnOUP5QKAck5amSWVEsdxyz8fkY/gHlchA8c7WwkTkNE7cHyQmRFvJxVGNHTEwAMs2usXXN5mcS1XYhkg+WFyMpwwbo/ZRRXIa2oCjYKAeMiecuISC5YXoiszKT+frBVCkgvqkJ6UaXUcSTVfNVleE9PuDlwlhGRXLC8EFkZV3tbjAz3AgD8fsa69zpqfv28ZUQkLywvRFaoee+erVa82m6Bug4nL5VDEICJfbkwHZGcsLwQWaEJfX2hEIBTuWrkXK6ROo4k/jirL26Dgtzh42ovcRoi6giWFyIr5OWswpAeHgCAP6z01hFnGRHJF8sLkZUy3Do6Y323jsprGnDoQhkAlhciOWJ5IbJSE5s2IDyWVYaSqnqJ03StHeeKoNWJiPRzQQ8vJ6njEFEHsbwQWanAbo6I6u4GnQhsP2tdt46arzZN5FUXIllieSGyYvH9re/WUU1DI/aeLwYATOrHWUZEcsTyQmTFmt+8D6aXoqJOI3GarrH3fDHqG3UI7OaAvv6uUschok4wWXnJysrC/PnzERoaCgcHB/Ts2RMvvvgiGhoa2nxeXV0dlixZAk9PTzg7O2PmzJkoLLSuS9pEXSXcxwU9vZ3QoNVhV0qR1HG6xF8XphMEQeI0RNQZJisvKSkp0Ol0WLduHc6cOYN33nkHa9euxd/+9rc2n/fkk09iy5Yt2LRpE/bs2YO8vDzMmDHDVDGJrF7zraPfreDWkUarw45zXFWXSO4EURTFrvpmq1atwocffogLFy60+nm1Wg1vb298+eWXuP322wHoS1CfPn2QkJCAG2644Zrfo6KiAm5ublCr1XB15SVhomtJzinH1H8fgKOdEsefvwn2tkqpI5nMvrRi3PPxEXg52+Hw3yZAqeCVFyJz0ZH37y4d86JWq+Hh4XHVzycmJkKj0WDChAmGxyIjIxEcHIyEhIRWn1NfX4+KiooWH0TUflHd3RDgZo+aBi32pZVIHcekmq8u3dTXl8WFSMa6rLykp6dj9erVWLhw4VWPKSgogJ2dHdzd3Vs87uvri4KC1i9pr1y5Em5uboaPoKAgY8YmsniCIGBSf8vf60inEw2rCXOKNJG8dbi8PPvssxAEoc2PlJSUFs/Jzc1FfHw8Zs2ahQULFhgtPACsWLECarXa8HHp0iWjfn0ia9A8/mNHSiE0Wp3EaUzjxKVyFFXWw1llgxE9PaWOQ0TXwaajT1i6dCnmzZvX5jFhYWGGP+fl5WHcuHEYMWIEPvroozaf5+fnh4aGBpSXl7e4+lJYWAg/v9b/p6RSqaBSqdqdn4iuNLSHBzyd7FBa3YAjmWUYGe4ldSSj+6PpltG4SB+obCx3XA+RNehwefH29oa3t3e7js3NzcW4ceMQExOD9evXQ6Fo+0JPTEwMbG1tsWPHDsycORMAkJqaiuzsbAwfPryjUYmonZQKATf19cXXRy9h6+kCiysvoijij7PNs4y4MB2R3JlszEtubi7i4uIQHByMt956C8XFxSgoKGgxdiU3NxeRkZE4cuQIAMDNzQ3z58/HU089hV27diExMRH3338/hg8f3q6ZRkTUec23jv44WwCdrssmIXaJzJJqZJZUw1YpIC7CR+o4RHSdOnzlpb22bduG9PR0pKenIzAwsMXnmmdnazQapKamoqamxvC5d955BwqFAjNnzkR9fT0mTZqEDz74wFQxiajJiHBPOKtsUFhRj6SccgwO7iZ1JKPZlarfDiA2VP8aiUjeTHblZd68eRBFsdWPZj169IAoioiLizM8Zm9vjzVr1qCsrAzV1dXYvHnzVce7EJHxqGyUuDFSf1XidwubdbQ7Vb96cFxE+255E5F5495GRGTQfOto+znL2ZKjur4Rhy+UAdAP1iUi+WN5ISKD0b29oFQIyCiuRnZpzbWfIAMHM0rRoNUh2MMRYV5OUschIiNgeSEiA1d7WwwJ0Y912ZVqGRs1Nr+OcRHe3IiRyEKwvBBRC83jXiyhvIiiiN1Nu2XH8ZYRkcVgeSGiFprHhSRklKK2QStxmutzvrAKeeo6qGwUGB7GVXWJLAXLCxG10MvHGd3dHVDfqMOhC6VSx7kue87rr7oM7+lp0btlE1kblhciakEQBMOU4p0p8r511LxL9phenCJNZElYXojoCuMi/hz38te1meSktkGLw5n6KdJjelvWdgdE1o7lhYiuMCLcE3ZKBXIu1yKjuErqOJ1yOLMUDY06BLjZo6e3s9RxiMiIWF6I6AqOdjaIDfMAAOxKKZY4TefsPd90y6g3p0gTWRqWFyJqldynTO9L05eu0RzvQmRxWF6IqFXN416OZpWhsk4jcZqOySuvRVpRFRQCMCqc412ILA3LCxG1qoeXE0K9nKDRijiQXiJ1nA5pvuoSHeQON0dbidMQkbGxvBDRVTVPmZbbuJe9nCJNZNFYXojoquQ4ZVqrE7G/ubxwijSRRWJ5IaKrGhbqAQdbJYoq63Emr0LqOO2SnFMOda0GLvY2iA50lzoOEZkAywsRXZW9rRIjeur3BGperdbcNU+RHhXuBRslf8URWSL+ZBNRm0b30t962Z8uj3EvzYN1x/TmeBciS8XyQkRtGtU06PVo5mWz32W6ok6DE5fKAfxZuojI8rC8EFGbeno7IcDNHg1aHY5klUkdp00H00ug1YkI83ZCYDdHqeMQkYmwvBBRmwRBwKimqxj7zpv3raM95zlFmsgasLwQ0TU1L7G/34wXqxNFEXvPN4934S0jIkvG8kJE1zQy3AuCAKQUVKKook7qOK3KLKlGbnkt7JQK3BDmKXUcIjIhlhciuiYPJzv0D3ADYL5XX5qvugzp0Q2OdjYSpyEiU2J5IaJ2aZ69Y67rvewzrKrL8S5Elo7lhYjaZdRfyou5bRXQ0KhDwoVSAJwiTWQNWF6IqF1iQrrBwVaJkqp6pBRUSh2nhWMXy1DToIWXswp9/FyljkNEJsbyQkTtorJRIjbMA8Cfq9iai72GKdJeUCgEidMQkamxvBBRuzVPmTa3cS/cEoDIurC8EFG7NY8nOZJZhjqNeWwVUPyXHa9HcbwLkVVgeSGiduvl4wxfVxXqG3U4lnVZ6jgA/twwsl+AK7ycVRKnIaKuwPJCRO0mCMJfbh2Zx7gXw3gX3jIishosL0TUIc23jvaawbgXnU40jL/hFGki68HyQkQdMjJcXxLO5VeguLJe0iznCipQUlUPRzslhoR4SJqFiLoOywsRdYiXswp9/fVrqRyQeKuA5ltGw8M8YWfDX2dE1oI/7UTUYaN7m8dWAZwiTWSdWF6IqMPG/GXQrlRbBdQ0NBpmPHG8C5F1YXkhog6LCekGlY0CRZX1SCuqkiTDoQulaNDqENjNAaFeTpJkICJpsLwQUYfZ2yoRG+YJANh7Xpop03+dIi0I3BKAyJqwvBBRp4wOl3bcy97m8S68ZURkdVheiKhTmgftHs4s7fKtAnIu1+BCcTWUCgEjwlleiKwNywsRdUqErwt8XFSo0+hwNKusS7938y2jQUHucLW37dLvTUTSY3khok4RBAFjm6Yo707t2nEvzeNsOEWayDqxvBBRp8VF+AAAdqcWddn3bNTqcCCDWwIQWTOWFyLqtFG9vKBUCMgorsalspou+Z4nc8pRWdcId0dbDAh075LvSUTmheWFiDrNzcEWMcHdAAC7u2jK9J6m8S4jw/XFiYisD8sLEV2XsRFN415SuubWUfOWAGN7cbwLkbVieSGi6xLXVF4OZph+yrS6RoOTl8oB/DlVm4isD8sLEV2Xvv6u8HFRoVajNfmU6f3pJdCJQC8fZ/i7OZj0exGR+WJ5IaLr0pVTpjlFmogAlhciMoKumDItiqJhvAunSBNZN5YXIrpuXTFlOqO4CnnqOtjZKBAb6mmS70FE8sDyQkTXrSumTDdPkY4N9YCDndIk34OI5IHlhYiMwtRTpg3jXThFmsjqsbwQkVGYcsp0TUMjEi6UAuBgXSJieSEiIzHllOl9aSVoaNQh2MMRvX2djfq1iUh+WF6IyCj+OmV6V4pxx71sP1sIAJjQxxeCwC0BiKwdywsRGc2Nkfop07uMOGVaqxOxs2kczYS+Pkb7ukQkXywvRGQ0o3p5wVYpILOkGheKq4zyNU9kX0ZpdQNc7W0wtIeHUb4mEckbywsRGY2Lva1hDZadRpp1tO2c/pbRuEgf2Cr5K4uIWF6IyMjG99Hf2tlxzjjl5a/jXYiIAJYXIjKy5nEvR7PKoK7VXNfXSi+qQkZxNWwUgmEdGSIik5WXrKwszJ8/H6GhoXBwcEDPnj3x4osvoqGhoc3nxcXFQRCEFh+LFi0yVUwiMrIQTyeE+zijUScaFpbrrJ9O5gHQr+3iam9rjHhEZAFsTPWFU1JSoNPpsG7dOoSHh+P06dNYsGABqqur8dZbb7X53AULFuCVV14x/N3R0dFUMYnIBMZH+iC9qAo7U4pwa3RAp76GKIrY0lRepnbyaxCRZTJZeYmPj0d8fLzh72FhYUhNTcWHH354zfLi6OgIPz8/U0UjIhMb38cX6/ZewPazhaht0HZqL6IzeRXILKmGykaBCX053oWI/tSlY17UajU8PK491fGLL76Al5cX+vfvjxUrVqCmxjS71BKRaQwJ6YYgDwdU1jfil1P5nfoazbeMJvTxhbPKZP/PIiIZ6rLfCOnp6Vi9evU1r7rMmTMHISEhCAgIQHJyMpYvX47U1FRs3ry51ePr6+tRX19v+HtFRYVRcxNRxykUAu4aGoxVv6fi6yPZuD0msEPP1+lE/NxUXm6N9jdFRCKSsQ5feXn22WevGFD7vx8pKSktnpObm4v4+HjMmjULCxYsaPPrP/TQQ5g0aRKioqIwd+5cbNy4Ed9//z0yMjJaPX7lypVwc3MzfAQFBXX0JRGRCcyKCYRSIeDYxctIK6zs0HOPXbyMPHUdnFU2iIvgqrpE1JIgiqLYkScUFxejtLS0zWPCwsJgZ2cHAMjLy0NcXBxuuOEGbNiwAQpFx/pSdXU1nJ2dsXXrVkyaNOmKz7d25SUoKAhqtRqurq4d+l5EZFwLNh7DtrOFmD8qFM/f0rfdz3v4i0T8eqoAdwwJxJu3R5swIRGZi4qKCri5ubXr/bvDt428vb3h7d2+9RZyc3Mxbtw4xMTEYP369R0uLgCQlJQEAPD3b/3SsUqlgkql6vDXJSLTmz0sCNvOFmLz8RwsmxQBe9trD9y9WFqNracLAADzR4WZOiIRyZDJBuzm5uYiLi4OwcHBeOutt1BcXIyCggIUFBS0OCYyMhJHjhwBAGRkZOAf//gHEhMTkZWVhZ9++gn33nsvxowZgwEDBpgqKhGZyNjePuju7oDLNRpsOnapXc/5ZH8mdCIwtrc3IvxcTJyQiOTIZOVl27ZtSE9Px44dOxAYGAh/f3/DRzONRoPU1FTDbCI7Ozts374dEydORGRkJJYuXYqZM2diy5YtpopJRCakVAhYNFZ/9eSD3Rmob9S2efzl6gb891gOAOChMbzqQkSt6/CYF3PXkXtmRGR6dRotxq7ahcKKevzztijMiQ2+6rHvbj+Pd7enoa+/K355bBQEQejCpEQkpY68f3NvIyIyKXtbJRaN7QkAWLMrHRqtrtXjLpXVYO0e/azCxXE9WVyI6KpYXojI5GYPC4aXswq55bVYt+fKZQ9EUcRLP51BnUaHG8I8cMsAru1CRFfH8kJEJmdvq8Qz8REAgH9tO4/dqUUtPv/H2ULsSCmCrVLAq9P786oLEbWJ5YWIusQdQ4Iwe1gQRBF4/OsknM5VAwB+P1OAp/97EoB+kG64D2cYEVHbuGEIEXWZl6b2w7n8SiRdKsctq/ejl48z0oqqAABDe3TDI+N6SZyQiOSAV16IqMuobJT46J4YxPfzg41CMBSXB0eF4ssFN3Rq92kisj688kJEXcrH1R5r74lBcWU9tp7OR7CnE8b2bt+q3UREAMsLEUnE20WFe4b3kDoGEckQbxsRERGRrLC8EBERkaywvBAREZGssLwQERGRrLC8EBERkaywvBAREZGssLwQERGRrLC8EBERkaywvBAREZGssLwQERGRrLC8EBERkaywvBAREZGssLwQERGRrFjcrtKiKAIAKioqJE5CRERE7dX8vt38Pt4WiysvlZWVAICgoCCJkxAREVFHVVZWws3Nrc1jBLE9FUdGdDod8vLy4OLiAkEQjPq1KyoqEBQUhEuXLsHV1dWoX9tcWPprtPTXB1j+a7T01wfwNVoCS399gPFfoyiKqKysREBAABSKtke1WNyVF4VCgcDAQJN+D1dXV4v9x9jM0l+jpb8+wPJfo6W/PoCv0RJY+usDjPsar3XFpRkH7BIREZGssLwQERGRrLC8dIBKpcKLL74IlUoldRSTsfTXaOmvD7D812jprw/ga7QElv76AGlfo8UN2CUiIiLLxisvREREJCssL0RERCQrLC9EREQkKywvREREJCssL+20Zs0a9OjRA/b29oiNjcWRI0ekjtRpK1euxNChQ+Hi4gIfHx9Mnz4dqampLY6Ji4uDIAgtPhYtWiRR4o556aWXrsgeGRlp+HxdXR2WLFkCT09PODs7Y+bMmSgsLJQwccf16NHjitcoCAKWLFkCQJ7nb+/evbj11lsREBAAQRDwww8/tPi8KIp44YUX4O/vDwcHB0yYMAFpaWktjikrK8PcuXPh6uoKd3d3zJ8/H1VVVV34KtrW1mvUaDRYvnw5oqKi4OTkhICAANx7773Iy8tr8TVaO/evv/56F7+S1l3rHM6bN++K7PHx8S2OkfM5BNDqz6UgCFi1apXhGHM+h+15f2jP79Ds7GxMmTIFjo6O8PHxwbJly9DY2Gi0nCwv7fDNN9/gqaeewosvvojjx48jOjoakyZNQlFRkdTROmXPnj1YsmQJDh06hG3btkGj0WDixImorq5ucdyCBQuQn59v+HjzzTclStxx/fr1a5F9//79hs89+eST2LJlCzZt2oQ9e/YgLy8PM2bMkDBtxx09erTF69u2bRsAYNasWYZj5Hb+qqurER0djTVr1rT6+TfffBPvv/8+1q5di8OHD8PJyQmTJk1CXV2d4Zi5c+fizJkz2LZtG37++Wfs3bsXDz30UFe9hGtq6zXW1NTg+PHjeP7553H8+HFs3rwZqampmDp16hXHvvLKKy3O7aOPPtoV8a/pWucQAOLj41tk/+qrr1p8Xs7nEECL15afn49PPvkEgiBg5syZLY4z13PYnveHa/0O1Wq1mDJlChoaGnDw4EF8+umn2LBhA1544QXjBRXpmoYNGyYuWbLE8HetVisGBASIK1eulDCV8RQVFYkAxD179hgeGzt2rPj4449LF+o6vPjii2J0dHSrnysvLxdtbW3FTZs2GR47d+6cCEBMSEjoooTG9/jjj4s9e/YUdTqdKIryPn+iKIoAxO+//97wd51OJ/r5+YmrVq0yPFZeXi6qVCrxq6++EkVRFM+ePSsCEI8ePWo45rfffhMFQRBzc3O7LHt7/e9rbM2RI0dEAOLFixcNj4WEhIjvvPOOacMZQWuv77777hOnTZt21edY4jmcNm2aeOONN7Z4TC7nUBSvfH9oz+/QX3/9VVQoFGJBQYHhmA8//FB0dXUV6+vrjZKLV16uoaGhAYmJiZgwYYLhMYVCgQkTJiAhIUHCZMajVqsBAB4eHi0e/+KLL+Dl5YX+/ftjxYoVqKmpkSJep6SlpSEgIABhYWGYO3cusrOzAQCJiYnQaDQtzmdkZCSCg4Nlez4bGhrw+eef44EHHmixGamcz9//yszMREFBQYvz5ubmhtjYWMN5S0hIgLu7O4YMGWI4ZsKECVAoFDh8+HCXZzYGtVoNQRDg7u7e4vHXX38dnp6eGDRoEFatWmXUy/Gmtnv3bvj4+CAiIgKLFy9GaWmp4XOWdg4LCwvxyy+/YP78+Vd8Ti7n8H/fH9rzOzQhIQFRUVHw9fU1HDNp0iRUVFTgzJkzRsllcRszGltJSQm0Wm2LkwAAvr6+SElJkSiV8eh0OjzxxBMYOXIk+vfvb3h8zpw5CAkJQUBAAJKTk7F8+XKkpqZi8+bNEqZtn9jYWGzYsAERERHIz8/Hyy+/jNGjR+P06dMoKCiAnZ3dFW8Gvr6+KCgokCbwdfrhhx9QXl6OefPmGR6T8/lrTfO5ae3nsPlzBQUF8PHxafF5GxsbeHh4yPLc1tXVYfny5Zg9e3aLTe8ee+wxDB48GB4eHjh48CBWrFiB/Px8vP322xKmbZ/4+HjMmDEDoaGhyMjIwN/+9jdMnjwZCQkJUCqVFncOP/30U7i4uFxxW1ou57C194f2/A4tKCho9We1+XPGwPJi5ZYsWYLTp0+3GBMCoMU95qioKPj7+2P8+PHIyMhAz549uzpmh0yePNnw5wEDBiA2NhYhISH473//CwcHBwmTmcbHH3+MyZMnIyAgwPCYnM8f6Qfv3nHHHRBFER9++GGLzz311FOGPw8YMAB2dnZYuHAhVq5cafZL0d91112GP0dFRWHAgAHo2bMndu/ejfHjx0uYzDQ++eQTzJ07F/b29i0el8s5vNr7gzngbaNr8PLyglKpvGIkdWFhIfz8/CRKZRyPPPIIfv75Z+zatQuBgYFtHhsbGwsASE9P74poRuXu7o7evXsjPT0dfn5+aGhoQHl5eYtj5Ho+L168iO3bt+PBBx9s8zg5nz8AhnPT1s+hn5/fFYPoGxsbUVZWJqtz21xcLl68iG3btrW46tKa2NhYNDY2Iisrq2sCGlFYWBi8vLwM/y4t5RwCwL59+5CamnrNn03APM/h1d4f2vM71M/Pr9Wf1ebPGQPLyzXY2dkhJiYGO3bsMDym0+mwY8cODB8+XMJknSeKIh555BF8//332LlzJ0JDQ6/5nKSkJACAv7+/idMZX1VVFTIyMuDv74+YmBjY2tq2OJ+pqanIzs6W5flcv349fHx8MGXKlDaPk/P5A4DQ0FD4+fm1OG8VFRU4fPiw4bwNHz4c5eXlSExMNByzc+dO6HQ6Q3kzd83FJS0tDdu3b4enp+c1n5OUlASFQnHF7RY5yMnJQWlpqeHfpSWcw2Yff/wxYmJiEB0dfc1jzekcXuv9oT2/Q4cPH45Tp061KKLNRbxv375GC0rX8PXXX4sqlUrcsGGDePbsWfGhhx4S3d3dW4yklpPFixeLbm5u4u7du8X8/HzDR01NjSiKopieni6+8sor4rFjx8TMzEzxxx9/FMPCwsQxY8ZInLx9li5dKu7evVvMzMwUDxw4IE6YMEH08vISi4qKRFEUxUWLFonBwcHizp07xWPHjonDhw8Xhw8fLnHqjtNqtWJwcLC4fPnyFo/L9fxVVlaKJ06cEE+cOCECEN9++23xxIkThpk2r7/+uuju7i7++OOPYnJysjht2jQxNDRUrK2tNXyN+Ph4cdCgQeLhw4fF/fv3i7169RJnz54t1Uu6QluvsaGhQZw6daoYGBgoJiUltfjZbJ6hcfDgQfGdd94Rk5KSxIyMDPHzzz8Xvb29xXvvvVfiV6bX1uurrKwUn376aTEhIUHMzMwUt2/fLg4ePFjs1auXWFdXZ/gacj6HzdRqtejo6Ch++OGHVzzf3M/htd4fRPHav0MbGxvF/v37ixMnThSTkpLErVu3it7e3uKKFSuMlpPlpZ1Wr14tBgcHi3Z2duKwYcPEQ4cOSR2p0wC0+rF+/XpRFEUxOztbHDNmjOjh4SGqVCoxPDxcXLZsmahWq6UN3k533nmn6O/vL9rZ2Yndu3cX77zzTjE9Pd3w+draWvHhhx8Wu3XrJjo6Ooq33XabmJ+fL2Hizvn9999FAGJqamqLx+V6/nbt2tXqv8v77rtPFEX9dOnnn39e9PX1FVUqlTh+/PgrXntpaak4e/Zs0dnZWXR1dRXvv/9+sbKyUoJX07q2XmNmZuZVfzZ37doliqIoJiYmirGxsaKbm5tob28v9unTR/znP//Z4s1fSm29vpqaGnHixImit7e3aGtrK4aEhIgLFiy44j+Bcj6HzdatWyc6ODiI5eXlVzzf3M/htd4fRLF9v0OzsrLEyZMniw4ODqKXl5e4dOlSUaPRGC2n0BSWiIiISBY45oWIiIhkheWFiIiIZIXlhYiIiGSF5YWIiIhkheWFiIiIZIXlhYiIiGSF5YWIiIhkheWFiIiIZIXlhYiIiGSF5YWIiIhkheWFiIiIZIXlhYiIiGTl/wH7Bb8VIigbfwAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "#@title loss over life time {vertical-output:true}\n",
        "\n",
        "mse_loss = tf.keras.losses.MeanSquaredError()\n",
        "loss_log1=[]\n",
        "x=seed\n",
        "ittiration=\"200\"#@param[100,200,300,1000]\n",
        "for i in range(int(ittiration)):\n",
        "  x=ca(x)\n",
        "  loss=  mse_loss(x[..., :4], tf.cast(target, tf.float32))\n",
        "  loss_log1.append(loss.numpy())\n",
        "    \n",
        "  \n",
        "\n",
        "pl.plot(np.log10(loss_log1))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "OeOr7tEUb7kD",
        "cellView": "form"
      },
      "outputs": [],
      "source": [
        "#@title Visualize structure generation (plot image  of each stage) (loaded)\n",
        "x=seed\n",
        "ittiration=\"100\"#@param[100,120,300]\n",
        "if not os.path.exists('img'):\n",
        "  os.mkdir('img')\n",
        "else:\n",
        "  import shutil\n",
        "  shutil.rmtree('img')\n",
        "  os.mkdir('img')\n",
        "\n",
        "for i in range(int(ittiration)):\n",
        "  x=ca(x)\n",
        "  #if i==60:\n",
        "    #msk = np.ones([1,SIZE_X,SIZE_Y,SIZE_Z,Channel],np.float32)\n",
        "    #msk[ 0,:SIZE_X//2,:, :, :] = 0\n",
        "    #x=x*msk\n",
        "  \n",
        "  clear_output(True) \n",
        "   \n",
        "  fig = plt.figure(figsize=(8,8))\n",
        "  ax=fig.add_subplot(projection='3d')\n",
        "  temp = x[0][...,:4]\n",
        "  max_v=np.max(temp) \n",
        "  temp = np.where(temp < 0, 0, temp)\n",
        "  temp=temp/max_v\n",
        "  z_resized = np.pad(temp[...,3], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "  r_resized = (np.pad(temp[...,0], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "  g_resized = (np.pad(temp[...,1], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "  b_resized = (np.pad(temp[...,2], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "\n",
        "  ax=fig.add_subplot(projection='3d')\n",
        "  obj=np.stack((r_resized,g_resized,b_resized,z_resized),axis=-1)\n",
        "  \n",
        "  xx = np.where(z_resized < 0.1, 0, z_resized)\n",
        "  if flag:\n",
        "    ax.voxels(xx, facecolors=obj[...,:3])\n",
        "  else:\n",
        "    ax.voxels(xx, facecolors=FACE_COLOR)\n",
        "  \n",
        "  from mpl_toolkits.mplot3d import Axes3D\n",
        "  #ax.view_init(azim=i)\n",
        "  ax.view_init()\n",
        "\n",
        "  filename = \"img/voxel_plot_{}.png\".format(i)\n",
        "  plt.savefig(filename)\n",
        "\n",
        "  \n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G9I50r15cMK8",
        "outputId": "58f2c4a0-6c2e-41b0-c546-3dc3caccac62",
        "cellView": "form"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Moviepy - Building video growth.mp4.\n",
            "Moviepy - Writing video growth.mp4\n",
            "\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": []
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Moviepy - Done !\n",
            "Moviepy - video ready growth.mp4\n"
          ]
        }
      ],
      "source": [
        "#@title Convertingn image sequence to video \n",
        "from moviepy.editor import ImageSequenceClip\n",
        "\n",
        "file_list = [\"img/voxel_plot_{}.png\".format(i) for i in range(1,int(ittiration))]\n",
        "\n",
        "clip = ImageSequenceClip(file_list, fps=6)\n",
        "clip.write_videofile(\"growth.mp4\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "OggIMHz_8SqO",
        "cellView": "form"
      },
      "outputs": [],
      "source": [
        " \n",
        "#@title Training loop with pooling {vertical-output:true}\n",
        "#@title Defining loss & train step\n",
        "BATCH_SIZE = 1 #@param {type:\"slider\", min:1, max:16, step:1}\n",
        "#defining optimizer\n",
        "lr=tf.keras.optimizers.schedules.PiecewiseConstantDecay([1000],[1e-3,3e-4])\n",
        "trainer=tf.optimizers.Adam(lr)\n",
        "loss_log=[]\n",
        "mse_loss = tf.keras.losses.MeanSquaredError()\n",
        "\n",
        "pool=np.repeat(seed,BATCH_SIZE,0)\n",
        "\n",
        "@tf.function\n",
        "def training_step(x):\n",
        "  with tf.GradientTape() as g:\n",
        "    N=\"100\"#@param[50,100,200]\n",
        "    ittiration =int(N)\n",
        "    for i in range(ittiration):\n",
        "      x=ca(x)\n",
        "    if flag:\n",
        "      loss = mse_loss(x[..., :4], tf.cast(target, tf.float32))\n",
        "    else:\n",
        "      loss = mse_loss(x[..., :4][...,3], tf.cast(target[...,3], tf.float32)) \n",
        "    \n",
        "\n",
        "  params=ca.trainable_variables\n",
        "  grads=g.gradient(loss, params) \n",
        "  grads=[g/(tf.norm(g)+1e-8) for g in grads]\n",
        "  trainer.apply_gradients(zip(grads,params))\n",
        "  return loss,x\n",
        "\n",
        "TRAINING_ITTIRATION=\"500\"#@param[500,1000,3000,5000,10000]\n",
        " \n",
        "for i in range(int(TRAINING_ITTIRATION)):\n",
        "  batch_idx=np.random.choice(len(pool),8, replace=True)\n",
        "  x0=pool[batch_idx]\n",
        "  x0[:1]=seed\n",
        "  loss,x=training_step(x0)   \n",
        "  loss_log.append(loss.numpy())\n",
        "  if i%10==0:\n",
        "    print()\n",
        "    pl.plot(loss_log,'.',alpha=0.3)\n",
        "    pl.yscale('log')\n",
        "    pl.show()\n",
        "    print(i,loss.numpy(),flush=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "SXg98Qe3kv8y"
      },
      "outputs": [],
      "source": [
        "#@title Save mode\n",
        "to_drive='False'#@param['True','False']\n",
        "folder_path='/content/'\n",
        "if to_drive=='True':\n",
        "  folder_path = '/content/drive/MyDrive/voxel/models/'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "nfDZCD91de0b"
      },
      "outputs": [],
      "source": [
        "#@title Save the trained model\n",
        "import pickle\n",
        "\n",
        "\n",
        "print(folder_path)\n",
        "# Save the model to a file\n",
        "file_name = path.split('/')[-1].split('.')[0]\n",
        "model_name ='3dNCA-'+file_name+'x'+str(Channel)+'x'+str(flag)+'.pkl'\n",
        "with open(folder_path+model_name, 'wb') as f:\n",
        "    pickle.dump(ca, f)\n",
        "print(model_name)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hFKbouVzbkU7"
      },
      "source": [
        "The code below is for loding and testing already trained models"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1aNTaCEihITx",
        "outputId": "86228f61-1645-4a5f-a368-9fc82ccf5bcb"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "/content/3dNCA-lizard_44x19x8x16xTrue.pkl\n",
            "['44', '19', '8', '16', 'True']\n",
            "True\n"
          ]
        }
      ],
      "source": [
        "#@title  Load the saved model from the file\n",
        "\n",
        "#@title Imports and Notebook Utilities\n",
        "import os\n",
        "import io\n",
        "import PIL.Image, PIL.ImageDraw\n",
        "import zipfile\n",
        "import numpy as np\n",
        "import matplotlib.pylab as pl\n",
        "\n",
        "import tensorflow as tf\n",
        "\n",
        "from IPython.display import Image, HTML, clear_output\n",
        "\n",
        "import csv\n",
        "import numpy as np\n",
        "from IPython.display import Image, HTML, clear_output\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def imwrite(f, a, fmt=None):\n",
        "  a = np.asarray(a)\n",
        "  if isinstance(f, str):\n",
        "    fmt = f.rsplit('.', 1)[-1].lower()\n",
        "    if fmt == 'jpg':\n",
        "      fmt = 'jpeg'\n",
        "    f = open(f, 'wb')\n",
        "  np2pil(a).save(f, fmt, quality=95)\n",
        "\n",
        "def np2pil(a):\n",
        "  if a.dtype in [np.float32, np.float64]:\n",
        "    a = np.uint8(np.clip(a, 0, 1)*255)\n",
        "  return PIL.Image.fromarray(a)\n",
        "\n",
        "\n",
        "def imencode(a, fmt='jpeg'):\n",
        "  a = np.asarray(a)\n",
        "  if len(a.shape) == 3 and a.shape[-1] == 4:\n",
        "    fmt = 'png'\n",
        "  f = io.BytesIO()\n",
        "  imwrite(f, a, fmt)\n",
        "  return f.getvalue()\n",
        "\n",
        "\n",
        "def imshow(a, fmt='jpeg'):\n",
        "  display(Image(data=imencode(a, fmt)))\n",
        "\n",
        "\n",
        "\n",
        "def zoom(img, scale=4):\n",
        "  img = np.repeat(img, scale, 0)\n",
        "  img = np.repeat(img, scale, 1)\n",
        "  return img\n",
        "\n",
        "\n",
        "\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from keras.layers import Conv3D\n",
        "from keras.models import Sequential, load_model\n",
        "\n",
        "\n",
        "def get_living_mask(x):\n",
        "  alpha = x[..., 3:4]\n",
        "  return  tf.cast(tf.nn.max_pool3d(alpha,3,1,'SAME') > 0.1,tf.float32)\n",
        "\n",
        "class CA(tf.Module):\n",
        "  def __init__(self):\n",
        "    self.model=tf.keras.Sequential([\n",
        "        Conv3D(filters=Channel*3, kernel_size=3, padding='same',input_shape=(SIZE_X,SIZE_Y,SIZE_Z,Channel), activation=tf.nn.relu),\n",
        "        Conv3D(filters=Channel, kernel_size=3, padding='same',kernel_initializer=tf.zeros),\n",
        "    ])\n",
        "  \n",
        "  @tf.function\n",
        "  def __call__(self,x):\n",
        "    alive_mask= get_living_mask(x)\n",
        "    update_mask = tf.floor(tf.random.uniform(x.shape) + 0.5)\n",
        "    x= x+self.model(x)*update_mask\n",
        "    x *= alive_mask\n",
        "    return x\n",
        "\n",
        "#------------------------------------------------------------------------------------\n",
        "\n",
        "\n",
        "\n",
        "import pickle\n",
        "\n",
        "model_name = \"3dGNCA-torus_19x6x19x24xTrue.pkl\" #@param [\"3dGNCA-torus_19x6x19x24xTrue.pkl\",\"3dNCA-bird-c_15x28x28x24xTrue.pkl\"]\n",
        "\n",
        "folder_path='/content/'\n",
        "if to_drive=='True':\n",
        "  from google.colab import drive\n",
        "  drive.mount('/content/drive')\n",
        "  folder_path = '/content/drive/MyDrive/voxel/models/'\n",
        "  file_path = folder_path+model_name\n",
        "else:\n",
        "  file_path='/content/3dNCA-lizard_44x19x8x16xTrue.pkl' #@param {type:\"string\"}\n",
        "\n",
        "print(file_path)\n",
        "with open(file_path ,'rb') as f:\n",
        "    GNCA = pickle.load(f)\n",
        "\n",
        "file_path = file_path.split('/')[-1].split('.')[0]\n",
        "param = [str(x) for x in file_path.split(\"_\")[-1].split(\"x\")]\n",
        "print(param)\n",
        "SIZE_X=int(param[0])\n",
        "SIZE_Y=int(param[1])\n",
        "SIZE_Z=int(param[2])\n",
        "Channel=int(param[3])\n",
        "flag=param[4]=='True'\n",
        "print(flag)\n",
        "max_dim=max(SIZE_X,SIZE_Y,SIZE_Z)\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 128
        },
        "id": "nn3RXkX5HpWc",
        "outputId": "5450c3a4-6bbc-496b-82f8-0142b569e0d9",
        "cellView": "form"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAACAAAABMCAYAAAAMTLXVAAAAa0lEQVR4nO3SwQkCMRAF0IlRsAxbWRuxtm1ksRprEN2g18zNS8jB946fT/gMiQAAAAD+XYkap5S08s6Vz9ABh6GvG/CDcr1d7n2wrY8lV15DP8H0CxhQzhHHPnhG3XOlDR0w/QIGAAAAAHwBoZ4MFt/D/ZwAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "step N :199\n"
          ]
        }
      ],
      "source": [
        "#@title Run the saved model\n",
        "import numpy as np\n",
        "seed = np.zeros([1,SIZE_X,SIZE_Y,SIZE_Z,Channel],np.float32)\n",
        "seed[:,SIZE_X//2,SIZE_Y//2,SIZE_Z//2,:4]=1\n",
        "print(\"Shape of seed:\")\n",
        "print(seed.shape)\n",
        "\n",
        "x=seed\n",
        "ittiration=\"200\"#@param[100,200,300]\n",
        "for i in range(int(ittiration)):\n",
        "  x=GNCA(x)\n",
        "  if  i==50:\n",
        "    msk = np.ones([1,SIZE_X,SIZE_Y,SIZE_Z,Channel],np.float32)\n",
        "    msk[ :,:SIZE_X//2, :, :, :] = 0\n",
        "    x=x*msk\n",
        "  clear_output(True)\n",
        "  temp = x[0][...,:4]\n",
        "  imshow(zoom(temp[7][...,:4]))\n",
        "  print() \n",
        "  print('step N :'+str(i))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "VDNbgp4UL9yu"
      },
      "outputs": [],
      "source": [
        "#@title Final structure generated  (loaded)\n",
        "fig = plt.figure(figsize=(5,5))\n",
        "ax=fig.add_subplot(projection='3d')\n",
        "temp = x[0][...,:4]\n",
        "max_v=np.max(temp)\n",
        "temp = np.where(temp < 0, 0, temp)\n",
        "temp=temp/max_v\n",
        "z_resized = np.pad(temp[...,3], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "r_resized = (np.pad(temp[...,0], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "g_resized = (np.pad(temp[...,1], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "b_resized = (np.pad(temp[...,2], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "\n",
        "obj=np.stack((r_resized,g_resized,b_resized,z_resized),axis=-1)\n",
        "\n",
        "ax=fig.add_subplot(projection='3d')\n",
        "xx = np.where(z_resized < 0.1, 0, z_resized)\n",
        "if flag:\n",
        "  ax.voxels(xx, facecolors=obj[...,:3])\n",
        "else:\n",
        "  ax.voxels(xx, facecolors=FACE_COLOR)\n",
        "\n",
        "\n",
        "plt.show()\n",
        "plt.close()\n",
        "print(xx.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "cKXvJRCseSRv"
      },
      "outputs": [],
      "source": [
        "#@title Visualize structure generation (plot image  of each stage) (loaded)\n",
        "x=seed\n",
        "ittiration=\"200\"#@param[100,200,300]\n",
        "if not os.path.exists('img'):\n",
        "  os.mkdir('img')\n",
        "else:\n",
        "  import shutil\n",
        "  shutil.rmtree('img')\n",
        "  os.mkdir('img')\n",
        "\n",
        "for i in range(int(ittiration)):\n",
        "  if  i==60:\n",
        "    \n",
        "  x=GNCA(x)\n",
        "   #if i==50:\n",
        "    #msk = np.ones([1,SIZE_X,SIZE_Y,SIZE_Z,Channel],np.float32)\n",
        "    #msk[ 0,:SIZE_X//2,:, :, :] = 0\n",
        "    #x=x*msk\n",
        "  \n",
        "  clear_output(True) \n",
        "   \n",
        "  fig = plt.figure(figsize=(5,5))\n",
        "  ax=fig.add_subplot(projection='3d')\n",
        "  temp = x[0][...,:4]\n",
        "  max_v=np.max(temp) \n",
        "  temp = np.where(temp < 0, 0, temp)\n",
        "  temp=temp/max_v\n",
        "  z_resized = np.pad(temp[...,3], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "  r_resized = (np.pad(temp[...,0], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "  g_resized = (np.pad(temp[...,1], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "  b_resized = (np.pad(temp[...,2], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "\n",
        "  ax=fig.add_subplot(projection='3d')\n",
        "  obj=np.stack((r_resized,g_resized,b_resized,z_resized),axis=-1)\n",
        "  \n",
        "  xx = np.where(z_resized < 0.1, 0, z_resized)\n",
        "  if flag:\n",
        "    ax.voxels(xx, facecolors=obj[...,:3])\n",
        "  else:\n",
        "    ax.voxels(xx, facecolors=FACE_COLOR)\n",
        "  \n",
        "  from mpl_toolkits.mplot3d import Axes3D\n",
        "  ax.view_init(azim=i)\n",
        "\n",
        "  filename = \"img/voxel_plot_{}.png\".format(i)\n",
        "  plt.savefig(filename)\n",
        "\n",
        "  \n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "IuydJ7eVHr7R"
      },
      "outputs": [],
      "source": [
        "#@title Visualize structure generation (plot image  of each stage)  (loaded)\n",
        "x=seed\n",
        "ittiration=\"100\"#@param[100,200,300]\n",
        "if not os.path.exists('img'):\n",
        "  import shutil\n",
        "  shutil.rmtree('img')\n",
        "  os.mkdir('img')\n",
        "\n",
        "for i in range(int(ittiration)):\n",
        "  x=GNCA(x)\n",
        "  clear_output(True) \n",
        "   \n",
        "  fig = plt.figure(figsize=(5,5))\n",
        "  ax=fig.add_subplot(projection='3d')\n",
        "  temp = x[0][...,:4]\n",
        "  max_v=np.max(temp) \n",
        "  temp = np.where(temp < 0, 0, temp)\n",
        "  temp=temp/max_v\n",
        "  z_resized = np.pad(temp[...,3], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "  r_resized = (np.pad(temp[...,0], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "  g_resized = (np.pad(temp[...,1], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "  b_resized = (np.pad(temp[...,2], (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant'))\n",
        "\n",
        "  obj=np.stack((r_resized,g_resized,b_resized,z_resized),axis=-1)\n",
        "  ax=fig.add_subplot(projection='3d')\n",
        "  obj=np.stack((r_resized,g_resized,b_resized,z_resized),axis=-1)\n",
        "  \n",
        "  xx = np.where(z_resized < 0.1, 0, z_resized)\n",
        "  if flag:\n",
        "    ax.voxels(xx, facecolors=obj[...,:3])\n",
        "  else:\n",
        "    ax.voxels(xx, facecolors=FACE_COLOR)\n",
        "  \n",
        "  from mpl_toolkits.mplot3d import Axes3D\n",
        "  ax.view_init(azim=i)\n",
        "\n",
        "  filename = \"img/voxel_plot_{}.png\".format(i)\n",
        "  plt.savefig(filename)\n",
        "\n",
        "  \n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Tm26mYZ5Jr0M",
        "outputId": "f30fabb7-1ddd-43c5-ba3b-bffed501828a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Moviepy - Building video growth.mp4.\n",
            "Moviepy - Writing video growth.mp4\n",
            "\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": []
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Moviepy - Done !\n",
            "Moviepy - video ready growth.mp4\n"
          ]
        }
      ],
      "source": [
        "#@title Convertingn image sequence to video (loaded)\n",
        "from moviepy.editor import ImageSequenceClip\n",
        "\n",
        "file_list = [\"img/voxel_plot_{}.png\".format(i) for i in range(int(ittiration))]\n",
        "\n",
        "clip = ImageSequenceClip(file_list, fps=6)\n",
        "clip.write_videofile(\"growth.mp4\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "8NalgjCKvoHQ"
      },
      "outputs": [],
      "source": [
        "!pip install gradio"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "PV7CklGlvq8u"
      },
      "outputs": [],
      "source": [
        "#@title web interface\n",
        "import matplotlib\n",
        "matplotlib.use('Agg')\n",
        "import gradio as gr\n",
        "import os\n",
        "import io\n",
        "import PIL.Image, PIL.ImageDraw\n",
        "import requests\n",
        "import numpy as np\n",
        "import matplotlib.pylab as pl\n",
        "import glob\n",
        "import tensorflow as tf\n",
        "from IPython.display import Image, HTML, clear_output\n",
        "import tqdm\n",
        "import os\n",
        "\n",
        "\n",
        "import csv\n",
        "import numpy as np\n",
        "from IPython.display import Image, HTML, clear_output\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from keras.layers import Conv3D\n",
        "from keras.models import Sequential, load_model\n",
        "\n",
        "\n",
        "def train_model(csv_file, training_type, channels, training_iterations, batch_size, color):\n",
        "  print(csv_file, training_type, channels, training_iterations, batch_size, color)\n",
        "\n",
        "  path=csv_file.name\n",
        "\n",
        "  def load_data(path):\n",
        "    \"\"\"Loads the data from a CSV file and returns it as a numpy array \n",
        "    along with its dimensions\"\"\"\n",
        "    file_name = path.split('/')[-1].split('.')[0]\n",
        "    dimension = [int(x) for x in file_name.split(\"_\")[-1].split(\"x\")]\n",
        "    with open(path, 'r') as f:\n",
        "        reader = csv.reader(f)\n",
        "        data = list(reader)\n",
        "    arr = np.array(data).astype(float)\n",
        "    return arr, dimension\n",
        "\n",
        "  def return_zLayer(arr, z):\n",
        "      \"\"\"Returns a single layer of voxels\"\"\"\n",
        "      z_slice = np.array([arr[arr[:, 2] == z]])\n",
        "      Zxy = z_slice[0][...,:6].astype(int)\n",
        "      R = np.zeros((SIZE_X, SIZE_Y))\n",
        "      G = np.zeros((SIZE_X, SIZE_Y))\n",
        "      B = np.zeros((SIZE_X, SIZE_Y))\n",
        "      Z = np.zeros((SIZE_X, SIZE_Y))\n",
        "      for i in Zxy:\n",
        "          Z[i[0]][i[1]] = 1\n",
        "          R[i[0]][i[1]] = i[3]\n",
        "          G[i[0]][i[1]] = i[4]\n",
        "          B[i[0]][i[1]] = i[5]\n",
        "      return R, G, B, Z\n",
        "\n",
        "  def stack_layers(arr):\n",
        "      \"\"\"Stacks up all the layers of voxels\"\"\"\n",
        "      r_stack, g_stack, b_stack, z_stack = return_zLayer(arr, 0)\n",
        "      for i in range(1, SIZE_Z):\n",
        "          r, g, b, a = return_zLayer(arr, i)\n",
        "          r_stack = np.dstack((r_stack, r))\n",
        "          g_stack = np.dstack((g_stack, g))\n",
        "          b_stack = np.dstack((b_stack, b))\n",
        "          z_stack = np.dstack((z_stack, a))\n",
        "      return r_stack, g_stack, b_stack, z_stack\n",
        " \n",
        " #@title  Parameters\n",
        "  arr, dimention = load_data(path)\n",
        "  max_dim=int(max(dimention))\n",
        "  SIZE_X=dimention[0]\n",
        "  SIZE_Y=dimention[1]\n",
        "  SIZE_Z=dimention[2]\n",
        " \n",
        "  flag = True if training_type[1] == 1 else False\n",
        "  BATCH_SIZE = batch_size\n",
        "  FACE_COLOR=color\n",
        "  Channel=int(channels)\n",
        "  print(dimention)\n",
        "  print(arr)  \n",
        "  arr = np.array(arr).astype(float)\n",
        "\n",
        "    \n",
        "  r_stack,g_stack,b_stack,z_stack=stack_layers(arr)\n",
        "\n",
        "  #@title Define target  \n",
        "\n",
        "  target=np.stack((r_stack/255,g_stack/255,b_stack/255,z_stack),axis=-1)\n",
        "\n",
        "  def plot_target(r_stack, g_stack, b_stack, z_stack, max_dim, SIZE_X, SIZE_Y, SIZE_Z, flag=True, FACE_COLOR=(0, 0, 1, 0.2)):\n",
        "      z_resized = np.pad(z_stack, (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "      r_resized = np.pad(r_stack, (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "      g_resized = np.pad(g_stack, (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "      b_resized = np.pad(b_stack, (((max_dim-SIZE_X)//2,(max_dim-SIZE_X)//2), ((max_dim-SIZE_Y)//2,(max_dim-SIZE_Y)//2), ((max_dim-SIZE_Z)//2,(max_dim-SIZE_Z)//2)), mode='constant')\n",
        "      target_resized = np.stack((r_resized/255, g_resized/255, b_resized/255, z_resized), axis=-1)\n",
        "      fig_t = plt.figure()\n",
        "      ax = fig_t.add_subplot(projection='3d')\n",
        "      if flag:\n",
        "          ax.voxels(target_resized[...,3], facecolors=target_resized[..., :3])\n",
        "      else:\n",
        "          ax.voxels(target_resized[...,3], facecolors=FACE_COLOR)\n",
        "      ax.set_title('Target voxel')\n",
        "      return fig_t\n",
        "\n",
        "\n",
        "  fig_t = plot_target(r_stack, g_stack, b_stack, z_stack, max_dim, SIZE_X, SIZE_Y, SIZE_Z, flag, FACE_COLOR)\n",
        "\n",
        "  def create_seed(channels, size_x, size_y, size_z):\n",
        "    \"\"\"\n",
        "    This function creates a seed array of zeros with dimensions [1, size_x, size_y, size_z, channels] and sets\n",
        "    the center of the array to 1 for the first three channels.\n",
        "    \n",
        "    Args:\n",
        "    channels (int): The number of channels for the seed array.\n",
        "    size_x (int): The size of the array along the x-axis.\n",
        "    size_y (int): The size of the array along the y-axis.\n",
        "    size_z (int): The size of the array along the z-axis.\n",
        "    \n",
        "    Returns:\n",
        "    seed (numpy array): A seed array of zeros with dimensions [1, size_x, size_y, size_z, channels] and the center\n",
        "    of the array set to 1 for the first three channels.\n",
        "    \"\"\"\n",
        "    seed = np.zeros([1, size_x, size_y, size_z, channels], np.float32)\n",
        "    seed[:, size_x//2, size_y//2, size_z//2, :4] = 1\n",
        "    print(\"Shape of seed:\")\n",
        "    print(seed.shape)\n",
        "    return seed\n",
        "  #defien seed\n",
        "  seed= create_seed(channels=Channel, size_x=SIZE_X, size_y=SIZE_Y, size_z=SIZE_Z)\n",
        "\n",
        "  def get_living_mask(x):\n",
        "    \"\"\"Returns a mask that identifies living cells in the CA grid\"\"\"\n",
        "    alpha = x[..., 3:4]\n",
        "    return  tf.cast(tf.nn.max_pool3d(alpha,3,1,'SAME') > 0.1,tf.float32)\n",
        "\n",
        "  class CA(tf.Module):\n",
        "    \"\"\"Defines the 3D Cellular Automaton model\"\"\"\n",
        "    def __init__(self):\n",
        "      self.model=tf.keras.Sequential([\n",
        "          Conv3D(\n",
        "              filters=Channel*3, \n",
        "              kernel_size=3, \n",
        "              padding='same',\n",
        "              input_shape=(SIZE_X,SIZE_Y,SIZE_Z,Channel), \n",
        "              activation=tf.nn.relu\n",
        "              ),\n",
        "          Conv3D(\n",
        "              filters=Channel,\n",
        "              kernel_size=3, \n",
        "              padding='same',\n",
        "              kernel_initializer=tf.zeros),\n",
        "      ])\n",
        "    \n",
        "    @tf.function\n",
        "    def __call__(self,x):      \"\"\"Runs a forward pass of the model\"\"\"\n",
        "      alive_mask= get_living_mask(x)\n",
        "      update_mask = tf.floor(tf.random.uniform(x.shape) + 0.5)\n",
        "      x= x+self.model(x)*update_mask\n",
        "      x *= alive_mask\n",
        "      return x\n",
        "\n",
        "  ca=CA()\n",
        "  x=seed\n",
        "  ittiration=100\n",
        "  for i in range(ittiration):\n",
        "    x=ca(x)\n",
        "\n",
        "  #@title Defining loss & train step\n",
        "  #defining optimizerimport tensorflow as tf\n",
        "  \n",
        "  # Create a PiecewiseConstantDecay learning rate schedule\n",
        "  def create_lr_schedule():\n",
        "      \"\"\"Creates a PiecewiseConstantDecay learning rate schedule\"\"\"\n",
        "      boundaries = [1000] # Set boundaries for learning rate changes\n",
        "      values = [1e-3, 3e-4] # Set learning rate values for each boundary\n",
        "      return tf.keras.optimizers.schedules.PiecewiseConstantDecay(boundaries, values)\n",
        "\n",
        "  # Define the optimizer\n",
        "  def create_optimizer(lr):\n",
        "      \"\"\"Creates an Adam optimizer\"\"\"\n",
        "      return tf.optimizers.Adam(lr)\n",
        "\n",
        "  # Define the mean squared error loss function\n",
        "  def create_mse_loss():\n",
        "      \"\"\"Creates a mean squared error loss function\"\"\"\n",
        "      return tf.keras.losses.MeanSquaredError()\n",
        "\n",
        "  # Define the training step function\n",
        "  @tf.function\n",
        "  def training_step(seed, target, ca, BATCH_SIZE, mse_loss, trainer, flag):\n",
        "      \"\"\"Performs a single training step\"\"\"\n",
        "      with tf.GradientTape() as g:\n",
        "          # Repeat the seed BATCH_SIZE times to create a batch of seeds\n",
        "          x = tf.repeat(seed, BATCH_SIZE, 0)\n",
        "          N = 100 # Number of iterations\n",
        "          for i in range(N):\n",
        "              x = ca(x)\n",
        "              if flag:\n",
        "                  # Compute loss for RGBA channels\n",
        "                  loss = mse_loss(x[..., :4], tf.cast(target, tf.float32))\n",
        "              else:\n",
        "                  # Compute loss for only alive channel\n",
        "                  loss = mse_loss(x[..., :4][...,3],\n",
        "                  tf.cast(target[...,3], tf.float32))  \n",
        "\n",
        "          # Compute gradients and update the model parameters\n",
        "          params = ca.trainable_variables\n",
        "          grads = g.gradient(loss, params) \n",
        "          grads = [g / (tf.norm(g) + 1e-8) for g in grads]\n",
        "          trainer.apply_gradients(zip(grads, params))\n",
        "          return loss, x\n",
        "\n",
        "  # Define the training loop\n",
        "  def train_model(seed, target, ca, training_iterations, batch_size, flag):\n",
        "      \"\"\"Trains the 3D Cellular Automaton model\"\"\"\n",
        "      # Create the optimizer, loss function, and learning rate schedule\n",
        "      lr_schedule = create_lr_schedule()\n",
        "      optimizer = create_optimizer(lr_schedule)\n",
        "      mse_loss = create_mse_loss()\n",
        "\n",
        "      # Track the loss over time\n",
        "      loss_log = []\n",
        "\n",
        "      for i in range(int(training_iterations)):\n",
        "          # Take a training step\n",
        "          loss, x = training_step(seed, target, ca, int(batch_size), mse_loss, optimizer, flag)\n",
        "          \n",
        "          # Append the loss to the log\n",
        "          loss_log.append(loss.numpy())\n",
        "          \n",
        "          # Print the loss every 20 iterations\n",
        "          if i % 20 == 0:\n",
        "              print(i, loss.numpy(), flush=True)\n",
        "              \n",
        "      # Create a plot of the loss over time\n",
        "      fig_l, ax = plt.subplots()\n",
        "      ax.plot(loss_log, '.', alpha=0.3)\n",
        "      ax.set_yscale('log')\n",
        "      ax.set_title('Loss over Training Iterations')\n",
        "      ax.set_xlabel('Training Iterations')\n",
        "      ax.set_ylabel('Loss')\n",
        "      plt.close(fig_l)\n",
        "      \n",
        "      return fig_l\n",
        "  \n",
        "  fig_l=train_model(seed, target, ca, training_iterations, batch_size, flag)\n",
        "  \n",
        "  return fig_t,fig_l\n",
        "\n",
        "\n",
        "\n",
        "    \n",
        "\n",
        "input_csv = gr.inputs.File(label='CSV File')\n",
        "input_training_type = gr.inputs.Radio(choices=[('Structure only', 0), ('structure with color', 1)], label='Training Type',default=('Structure only', 0))\n",
        "input_channels = gr.inputs.Number(label='Channels',default=16)\n",
        "input_training_iterations = gr.inputs.Number(label='Training Iterations',default=1000)\n",
        "input_batch_size = gr.inputs.Number(label='Batch Size',default=1)\n",
        "input_color = gr.inputs.Dropdown(choices= ['blue', 'green', 'red', 'cyan', 'magenta', 'yellow', 'black', 'white', 'gray', 'silver', 'purple', 'orange', 'brown']\n",
        ", label='Color',default='silver')\n",
        "\n",
        "\n",
        "\n",
        "# Create the Gradio interface\n",
        "gr.Interface(fn=train_model,\n",
        "             inputs=[input_csv, input_training_type, input_channels, input_training_iterations, input_batch_size, input_color],\n",
        "             outputs=[\"plot\",\"plot\"],\n",
        "             title='Loss Plot and Images Generator'\n",
        "             ).launch()"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}