diff --git a/unet_model/jbacon_unet_modeling-Copy1.ipynb b/unet_model/jbacon_unet_modeling-Copy1.ipynb
deleted file mode 100644
index cc5bc3e..0000000
--- a/unet_model/jbacon_unet_modeling-Copy1.ipynb
+++ /dev/null
@@ -1,1199 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "JYjXEHVOctJT"
-   },
-   "source": [
-    "# Check to see if we're running in Colab (versus local server)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/"
-    },
-    "id": "1EWdW30icnzz",
-    "outputId": "2e03f665-5e0d-4e17-8de1-51e3686a0672"
-   },
-   "outputs": [],
-   "source": [
-    "try:\n",
-    "    from google.colab import drive\n",
-    "\n",
-    "    IN_COLAB = True\n",
-    "except:\n",
-    "    IN_COLAB = False\n",
-    "\n",
-    "if IN_COLAB:\n",
-    "    print(\"We're running Colab\")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "vQ6hZkV_cpBr"
-   },
-   "source": [
-    "# Mount the Google Drive (if we're in Colab), switch current directory to a directory on the Google Drive\n",
-    "- we will (optionally) create the specified directory on the Google Drive if it doesn't exist\n",
-    "\n",
-    "- navigate to our Harvard Capstone shared folder -> right-click -> organize -> add shortcut -> all locations -> add \"My Drive\""
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/"
-    },
-    "id": "Wyo0Xe4vbX6n",
-    "outputId": "c96ca0ee-7e29-4316-e97b-79b12d977606"
-   },
-   "outputs": [],
-   "source": [
-    "if IN_COLAB:\n",
-    "    # Mount the Google Drive at mount\n",
-    "    mount = \"/content/gdrive\"\n",
-    "    print(\"Colab: mounting Google drive on \", mount)\n",
-    "\n",
-    "    drive.mount(mount)\n",
-    "\n",
-    "    # Switch to the directory on the Google Drive that you want to use\n",
-    "    import os\n",
-    "\n",
-    "    drive_root = mount + \"/My Drive/Harvard Capstone/Modeling/UNet\"\n",
-    "\n",
-    "    # Create drive_root if it doesn't exist\n",
-    "    # create_drive_root = True\n",
-    "    # if create_drive_root:\n",
-    "    #  print(\"\\nColab: making sure \", drive_root, \" exists.\")\n",
-    "    #  os.makedirs(drive_root, exist_ok=True)\n",
-    "\n",
-    "    # Change to the directory\n",
-    "    print(\"\\nColab: Changing directory to \", drive_root)\n",
-    "    %cd $drive_root"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "OKqlpKMnlCnh"
-   },
-   "source": [
-    "# Work with files on the Google Drive\n",
-    "- existing files\n",
-    "- upload files to Google Drive (as per normal)\n",
-    "- load files from external source"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/",
-     "height": 35
-    },
-    "id": "9gpzZZtUh30m",
-    "outputId": "9728f5a5-934c-4332-b5e4-e19e15de1e00"
-   },
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "'/home/bacon/code/personal/icedyno/unet_model'"
-      ]
-     },
-     "execution_count": 3,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "# Verify we're in the correct working directory\n",
-    "%pwd"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "Rz8POlq3lHSq"
-   },
-   "source": [
-    "## Verify that imports (of modules on the Google Drive) work"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {
-    "id": "f2Hsz8qQlGvo"
-   },
-   "outputs": [],
-   "source": [
-    "import glob, json, os\n",
-    "import datetime as dt\n",
-    "from IPython.display import HTML\n",
-    "\n",
-    "import pandas as pd\n",
-    "import numpy as np\n",
-    "import xarray as xr\n",
-    "import matplotlib.pyplot as plt\n",
-    "from sklearn.model_selection import train_test_split\n",
-    "from sklearn.ensemble import RandomForestClassifier\n",
-    "from sklearn.metrics import accuracy_score\n",
-    "\n",
-    "import tensorflow as tf\n",
-    "from tensorflow.keras.models import Model\n",
-    "from tensorflow.keras.layers import (\n",
-    "    Input,\n",
-    "    Conv2D,\n",
-    "    MaxPooling2D,\n",
-    "    Dropout,\n",
-    "    Conv2DTranspose,\n",
-    "    concatenate,\n",
-    "    Input,\n",
-    "    Lambda,\n",
-    "    Activation,\n",
-    ")\n",
-    "from tensorflow.keras.callbacks import ModelCheckpoint, EarlyStopping\n",
-    "from tensorflow.keras.optimizers import Adam\n",
-    "from tensorflow.keras.layers import (\n",
-    "    Conv2D,\n",
-    "    MaxPooling2D,\n",
-    "    concatenate,\n",
-    "    Conv2DTranspose,\n",
-    "    Input,\n",
-    "    Lambda,\n",
-    "    Activation,\n",
-    "    Add,\n",
-    "    Reshape,\n",
-    ")\n",
-    "from tensorflow.keras.models import Model"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "# Set configuration constants"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {
-    "id": "dzcSg_26piy6"
-   },
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "2024-03-23 17:15:12.856473: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2024-03-23 17:15:12.976760: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2024-03-23 17:15:12.977538: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n"
-     ]
-    }
-   ],
-   "source": [
-    "data_root = \"ims_netcdf_1km_cropped_2_000km_window_74lat_-170lon/\"\n",
-    "if not IN_COLAB:\n",
-    "    data_root = os.path.join(\"..\", \"data\", data_root)\n",
-    "    tf.config.set_visible_devices([], \"GPU\")\n",
-    "WINDOW_SIZE = 2000  # km"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 100,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "batch_size = 1\n",
-    "test_batch_size = 3\n",
-    "dim = (WINDOW_SIZE, WINDOW_SIZE, 3)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "HCLz7nyLun6k"
-   },
-   "source": [
-    "## Data Processing: XArray and Numpy\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {
-    "id": "J_uYc9VMujWV"
-   },
-   "outputs": [],
-   "source": [
-    "# Define a function to load a single .nc file for a given year and day\n",
-    "def load_nc_file(year, day) -> xr.Dataset:\n",
-    "    \"\"\"Loads the cropped, grid-corrected netcdf files on the Beaufort Sea with 74,0lat_-170,0lon\"\"\"\n",
-    "    # Generate the file path based on the year and day\n",
-    "    file_path = os.path.join(\n",
-    "        data_root,\n",
-    "        str(year),\n",
-    "        f\"ims{year}{day:03d}_1km_v1.3_grid{WINDOW_SIZE}_74,0lat_-170,0lon.nc\",\n",
-    "    )\n",
-    "\n",
-    "    # Load the .nc file using xarray\n",
-    "    with xr.open_dataset(file_path) as dataset:\n",
-    "        return dataset"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {
-    "id": "ui-tRfxBjY6B"
-   },
-   "outputs": [],
-   "source": [
-    "def crop_to_beaufort_sea(ds: xr.Dataset, window_size: int) -> xr.Dataset:\n",
-    "    \"\"\"\n",
-    "    Center window on beaufort sea coordinates in **current** netcdf coordinate system (not quite polar stereographic) and\n",
-    "    crop to WINDOW_SIZE x WINDOW_SIZE (not 2*window size x 2*window size as was previously)\n",
-    "    \"\"\"\n",
-    "    # Beaufort Sea x, y in **current** IMS netcdf coordinate system\n",
-    "    x = -1652603.364653003  # meters\n",
-    "    y = -291398.56159791426  # meters\n",
-    "\n",
-    "    # These x, y in convert back to the below with current geolocation.py functions.\n",
-    "    ## longitude: -80.0, latitude: 74.0\n",
-    "    # Actual Beaufort Sea coordinates are closer to longitude: -140, latitude: 74.\n",
-    "\n",
-    "    beaufort_ds = ds.sel(\n",
-    "        x=slice(x - 1000 * window_size // 2, x + 1000 * window_size // 2),\n",
-    "        y=slice(y - 1000 * window_size // 2, y + 1000 * window_size // 2),\n",
-    "    )\n",
-    "    return beaufort_ds"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {
-    "id": "TbKOBHXaytLN"
-   },
-   "outputs": [],
-   "source": [
-    "def load_sie_data(year, day) -> np.array:\n",
-    "    \"\"\"Returns a 2D numpy array copy of the IMS surface values\"\"\"\n",
-    "    return load_nc_file(year, day).IMS_Surface_Values[0].values.copy()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {
-    "id": "l0SnpTY11oiw"
-   },
-   "outputs": [],
-   "source": [
-    "def load_target_sie_data(year, day) -> np.array:\n",
-    "    \"\"\"Returns a 2D numpy array copy of the IMS surface values\"\"\"\n",
-    "    ds = load_nc_file(year, day)\n",
-    "    sie = ds.IMS_Surface_Values[0].values.copy()\n",
-    "    binary_sie = sie.copy()\n",
-    "    binary_sie[sie != 3] = 0\n",
-    "\n",
-    "    # Sea and Lake Ice is treated as 1\n",
-    "    binary_sie[sie == 3] = 1\n",
-    "    return binary_sie"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "metadata": {
-    "id": "SR_qBhtuzRYP"
-   },
-   "outputs": [],
-   "source": [
-    "def binarize_data(sie: np.array) -> np.array:\n",
-    "    \"\"\"\n",
-    "    New SIE:\n",
-    "    0: Open water/out of bounds\n",
-    "    1: Sea ice or lake ice (lake mask not applied)\n",
-    "    2: Land\n",
-    "    \"\"\"\n",
-    "    binary_sie = sie.copy()\n",
-    "    binary_sie[sie != 3] = 0\n",
-    "\n",
-    "    # Sea and Lake Ice is treated as 1\n",
-    "    binary_sie[sie == 3] = 1\n",
-    "\n",
-    "    # Land and Snow-Covered Land is sent to 2.\n",
-    "    binary_sie[sie == 2] = 2\n",
-    "    binary_sie[sie == 4] = 2\n",
-    "    return binary_sie"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "metadata": {
-    "id": "XK74-F6I6uv5"
-   },
-   "outputs": [],
-   "source": [
-    "def load_n_day_chunk(year: int, day: int, n=4) -> np.array:\n",
-    "    \"\"\"\n",
-    "    Return np.array with shape (height, width, channels).\n",
-    "\n",
-    "    Does NOT wrap years or account for missing days.\n",
-    "    Starts n_day chunk at specified day, year.\n",
-    "\n",
-    "    Returns:\n",
-    "        np.array: shape (sie_y_shape, sie_x_shape, n_day)\n",
-    "    \"\"\"\n",
-    "    sie_chunk = []\n",
-    "    for day in range(day, day + n):\n",
-    "        sie = binarize_data(load_sie_data(year, day))\n",
-    "        sie_chunk.append(sie)\n",
-    "\n",
-    "    assert len(sie_chunk) == n\n",
-    "    # Use np.stack to stack the individual 2D arrays along a new third axis, resulting in (height, width, channels)\n",
-    "    return np.stack(sie_chunk, axis=-1)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "0jzLlnYC3NXT"
-   },
-   "source": [
-    "## Example usage"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/",
-     "height": 435
-    },
-    "id": "gbF6ZpdDukLT",
-    "outputId": "210099b5-9e63-4eab-d66b-54dcf33c131f"
-   },
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "# Example usage:\n",
-    "example = True\n",
-    "if example:\n",
-    "    year = 2023\n",
-    "    day = 150\n",
-    "    data = load_nc_file(year, day)\n",
-    "    multiclass_sie = load_sie_data(year, day)\n",
-    "    sie = binarize_data(multiclass_sie)\n",
-    "    plt.imshow(sie)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "KtZScGB68EEj"
-   },
-   "source": [
-    "# Data Processing: Loading data for Tensorflow"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "LlkL2oECSHhf"
-   },
-   "source": [
-    "# Data generator for all data available\n",
-    "To be used as either train or test data generator, depending on slice of filenames passed in."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {
-    "id": "RVeGt-1c_zww"
-   },
-   "outputs": [],
-   "source": [
-    "class AllDataGenerator(tf.keras.utils.Sequence):\n",
-    "    \"\"\"\n",
-    "    Generator for Keras training to allow multiprocessing and training on batches with only the\n",
-    "    batch itself being loaded into memory.\n",
-    "\n",
-    "    \"\"\"\n",
-    "\n",
-    "    def __init__(\n",
-    "        self,\n",
-    "        filenames: list[str],\n",
-    "        batch_size: int = 2,\n",
-    "        dim: tuple = (8000, 8000, 5),\n",
-    "        shuffle: bool = True,\n",
-    "    ):\n",
-    "        self.filenames = sorted(filenames)\n",
-    "        self.years = self.years_from_filenames()\n",
-    "        self.days = self.days_from_filenames()\n",
-    "        self.batch_size = batch_size\n",
-    "        self.dim = dim  # (height, width, channel)\n",
-    "        self.shuffle = True\n",
-    "        self.data_IDs = self._get_data_ids()\n",
-    "        self.on_epoch_end()\n",
-    "\n",
-    "    def years_from_filenames(self):\n",
-    "        years = [\n",
-    "            int(file.split(\"/\")[-1].split(\"ims\")[1][:4]) for file in self.filenames\n",
-    "        ]\n",
-    "        return years\n",
-    "\n",
-    "    def days_from_filenames(self):\n",
-    "        days = [int(file.split(\"/\")[-1].split(\"_\")[0][-3:]) for file in self.filenames]\n",
-    "        return days\n",
-    "\n",
-    "    def _get_data_ids(self):\n",
-    "        return list(zip(self.years, self.days))\n",
-    "\n",
-    "    def get_years_days_of_batch(self, index: int):\n",
-    "        \"\"\"Given a batch index, return a list of the year and days for that batch\"\"\"\n",
-    "        years = self.years[index * self.batch_size : (index + 1) * self.batch_size]\n",
-    "        days = self.days[index * self.batch_size : (index + 1) * self.batch_size]\n",
-    "        return list(zip(years, days))\n",
-    "\n",
-    "    def __len__(self):\n",
-    "        \"\"\"Number of batches per epoch\"\"\"\n",
-    "        return len(self.data_IDs) // self.batch_size\n",
-    "\n",
-    "    def __getitem__(self, index):\n",
-    "        \"\"\"Generate one batch of data\"\"\"\n",
-    "        # Collect data IDs for this batch number\n",
-    "        batch_data_ids = self.data_IDs[\n",
-    "            index * self.batch_size : (index + 1) * self.batch_size\n",
-    "        ]\n",
-    "\n",
-    "        # Generate data\n",
-    "        X, y = self._data_generation(batch_data_ids)\n",
-    "\n",
-    "        return X.astype(\"float16\"), y.astype(\"int32\")\n",
-    "\n",
-    "    def on_epoch_end(self):\n",
-    "        \"\"\"Updates indexes after each epoch\"\"\"\n",
-    "        if self.shuffle:\n",
-    "            np.random.shuffle(self.data_IDs)\n",
-    "\n",
-    "    def load_n_day_chunk(self, i, n):\n",
-    "        \"\"\"Starts at year, day and returns the next n days of processed SIE.\"\"\"\n",
-    "        days = self.days[i : i + n]\n",
-    "        years = self.years[i : i + n]\n",
-    "\n",
-    "        sie_chunk = []\n",
-    "        for year, day in zip(years, days):\n",
-    "            sie = binarize_data(load_sie_data(year, day))\n",
-    "            sie_chunk.append(sie)\n",
-    "\n",
-    "        assert len(sie_chunk) == n\n",
-    "        # Use np.stack to stack the individual 2D arrays along a new third axis, resulting in (height, width, channels)\n",
-    "        return np.stack(sie_chunk, axis=-1)\n",
-    "\n",
-    "    def _data_generation(self, batch_data_ids):\n",
-    "        \"\"\"Generates data containing batch_size samples\"\"\"\n",
-    "        X = np.empty((self.batch_size, *self.dim), dtype=\"float16\")\n",
-    "        y = np.empty((self.batch_size, self.dim[0], self.dim[1], 1), dtype=\"int32\")\n",
-    "\n",
-    "        for i, (year, day) in enumerate(batch_data_ids):\n",
-    "            # Load a 5-day chunk as the input\n",
-    "            X[i,] = self.load_n_day_chunk(i, self.dim[2])\n",
-    "            # Load the next day as the target\n",
-    "            y[i,] = np.expand_dims(\n",
-    "                load_target_sie_data(\n",
-    "                    self.years[i + self.dim[2]], self.days[i + self.dim[2]]\n",
-    "                ),\n",
-    "                axis=-1,\n",
-    "            )\n",
-    "\n",
-    "        return X, y"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "1qnZICkoWQCG"
-   },
-   "source": [
-    "# Test/Train split"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/"
-    },
-    "id": "kJL6GG2AWPmr",
-    "outputId": "06dea8b2-2efd-4b49-ec88-cf5300ff20c7"
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Number of Train files is:  2344\n",
-      "Number of Test files is:  669\n",
-      "Number of Validation files is:  336\n"
-     ]
-    }
-   ],
-   "source": [
-    "test_frac = 0.2\n",
-    "validation_frac = 0.1\n",
-    "\n",
-    "train_frac = 1 - validation_frac - test_frac\n",
-    "all_netcdf_files = glob.glob(data_root + \"/**/*.nc\", recursive=True)\n",
-    "\n",
-    "train_idx = int(len(all_netcdf_files) * train_frac)\n",
-    "test_idx = train_idx + int(len(all_netcdf_files) * test_frac)\n",
-    "\n",
-    "train_files = all_netcdf_files[:train_idx]\n",
-    "test_files = all_netcdf_files[train_idx:test_idx]\n",
-    "validation_files = all_netcdf_files[test_idx:]\n",
-    "\n",
-    "print(\"Number of Train files is: \", len(train_files))\n",
-    "print(\"Number of Test files is: \", len(test_files))\n",
-    "print(\"Number of Validation files is: \", len(validation_files))"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "4-YSJYkJ3qUm"
-   },
-   "source": [
-    "### Model prototype\n",
-    "Very simple UNet model that takes the entire region of interest in and outputs an image of the same size."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {
-    "id": "xWORYNWN_xyC"
-   },
-   "outputs": [],
-   "source": [
-    "def simple_unet_with_softmax(input_shape=(4000 * 2, 4000 * 2, 5)):\n",
-    "    inputs = Input(input_shape)\n",
-    "    # Downsample\n",
-    "    c1 = Conv2D(16, (3, 3), activation=\"relu\", padding=\"same\")(inputs)\n",
-    "    p1 = MaxPooling2D((2, 2))(c1)\n",
-    "    c2 = Conv2D(32, (3, 3), activation=\"relu\", padding=\"same\")(p1)\n",
-    "    p2 = MaxPooling2D((2, 2))(c2)\n",
-    "    # Bottleneck\n",
-    "    b = Conv2D(64, (3, 3), activation=\"relu\", padding=\"same\")(p2)\n",
-    "    # Upsample\n",
-    "    u1 = Conv2DTranspose(32, (2, 2), strides=(2, 2), padding=\"same\")(b)\n",
-    "    u1 = concatenate([u1, c2])\n",
-    "    u2 = Conv2DTranspose(16, (2, 2), strides=(2, 2), padding=\"same\")(u1)\n",
-    "    u2 = concatenate([u2, c1])\n",
-    "    outputs = Conv2D(1, (1, 1), activation=\"sigmoid\")(u2)  # 0, 1, 2\n",
-    "    model = Model(inputs=[inputs], outputs=[outputs])\n",
-    "    return model\n",
-    "\n",
-    "\n",
-    "def simple_unet_with_skip(input_shape: tuple[int, int, int]):\n",
-    "    \"\"\"\n",
-    "    Unet to predict the change in sea ice concentration by adding last day's forecast\n",
-    "    to model output, then sigmoid activation for pixel-wise classification.\n",
-    "    \"\"\"\n",
-    "    inputs = Input(input_shape)\n",
-    "    # Slice the last channel of the input\n",
-    "    last_channel = Lambda(lambda x: x[:, :, :, -1:])(\n",
-    "        inputs\n",
-    "    )  # Assuming the last channel is what we want to add to the output\n",
-    "\n",
-    "    # Downsample\n",
-    "    c1 = Conv2D(16, (3, 3), activation=\"relu\", padding=\"same\")(inputs)\n",
-    "    p1 = MaxPooling2D((2, 2))(c1)\n",
-    "    c2 = Conv2D(32, (3, 3), activation=\"relu\", padding=\"same\")(p1)\n",
-    "    p2 = MaxPooling2D((2, 2))(c2)\n",
-    "\n",
-    "    # Bottleneck\n",
-    "    b = Conv2D(64, (3, 3), activation=\"relu\", padding=\"same\")(p2)\n",
-    "\n",
-    "    # Upsample\n",
-    "    u1 = Conv2DTranspose(32, (2, 2), strides=(2, 2), padding=\"same\")(b)\n",
-    "    u1 = concatenate([u1, c2])\n",
-    "    u2 = Conv2DTranspose(16, (2, 2), strides=(2, 2), padding=\"same\")(u1)\n",
-    "    u2 = concatenate([u2, c1])\n",
-    "\n",
-    "    # Concatenate the last channel of the input with the last upsampled features before the final convolution\n",
-    "    pre_output = concatenate([u2, last_channel], axis=-1)\n",
-    "\n",
-    "    # Final convolution without activation\n",
-    "    outputs = Conv2D(1, (1, 1), padding=\"same\", activation=\"softmax\")(pre_output)\n",
-    "\n",
-    "    model = Model(inputs=[inputs], outputs=[outputs])\n",
-    "    return model"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "Ap-87s6j324g"
-   },
-   "source": [
-    "# Model Training"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 45,
-   "metadata": {
-    "id": "TscI1oFMQo3y"
-   },
-   "outputs": [],
-   "source": [
-    "train_generator = AllDataGenerator(train_files, batch_size=batch_size, dim=dim)\n",
-    "test_generator = AllDataGenerator(test_files, batch_size=test_batch_size, dim=dim)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 32,
-   "metadata": {
-    "id": "HsQeOTpEbQPT"
-   },
-   "outputs": [],
-   "source": [
-    "# Setup model checkpointing\n",
-    "import datetime\n",
-    "\n",
-    "datetime_string = datetime.datetime.now().strftime(\"%I:%M%p_%B_%d_%Y\")\n",
-    "\n",
-    "# Model checkpoint foldernames now generated by datetime (won't overwrite previous runs)\n",
-    "checkpoint_dir = f\"./model_checkpoints/jbacon/unet_{datetime_string}_{WINDOW_SIZE}km/\"\n",
-    "if not os.path.exists(checkpoint_dir):\n",
-    "    os.makedirs(checkpoint_dir)\n",
-    "checkpoint_path = os.path.join(checkpoint_dir, \"cp-{epoch:04d}.ckpt\")\n",
-    "\n",
-    "train = False\n",
-    "if train:\n",
-    "    checkpoint_callback = ModelCheckpoint(\n",
-    "        filepath=checkpoint_path,\n",
-    "        save_weights_only=False,\n",
-    "        monitor=\"loss\",\n",
-    "        mode=\"min\",\n",
-    "        save_best_only=True,\n",
-    "        verbose=1,\n",
-    "    )\n",
-    "\n",
-    "    early_stopping_callback = EarlyStopping(\n",
-    "        monitor=\"loss\", patience=10, verbose=1, mode=\"min\"\n",
-    "    )\n",
-    "\n",
-    "    model = simple_unet_with_softmax(input_shape=dim)  # skip(input_shape=dim)\n",
-    "    model.compile(\n",
-    "        optimizer=\"adam\", loss=\"binary_crossentropy\", metrics=[\"accuracy\"]\n",
-    "    )  #'binary_crossentropy', metrics=['accuracy']) 'sparse_categorical_crossentropy'\n",
-    "else:\n",
-    "    model = tf.keras.models.load_model(\n",
-    "        \"./model_checkpoints/jbacon/unet_12:35PM_March_23_2024_2000km/cp-0003.ckpt\"\n",
-    "    )"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 33,
-   "metadata": {
-    "id": "MsLENCJXfvkZ"
-   },
-   "outputs": [],
-   "source": [
-    "if train:\n",
-    "    # Log model parameters\n",
-    "    with open(os.path.join(checkpoint_dir, \"model_params.json\"), \"w\") as f:\n",
-    "        f.write(model.to_json())"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 34,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/"
-    },
-    "id": "X9LCjLY83Jkk",
-    "outputId": "b003fdfc-7934-456f-f55a-8dda5a51d7ab"
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Model: \"model\"\n",
-      "__________________________________________________________________________________________________\n",
-      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
-      "==================================================================================================\n",
-      "input_1 (InputLayer)            [(None, 2000, 2000,  0                                            \n",
-      "__________________________________________________________________________________________________\n",
-      "conv2d (Conv2D)                 (None, 2000, 2000, 1 448         input_1[0][0]                    \n",
-      "__________________________________________________________________________________________________\n",
-      "max_pooling2d (MaxPooling2D)    (None, 1000, 1000, 1 0           conv2d[0][0]                     \n",
-      "__________________________________________________________________________________________________\n",
-      "conv2d_1 (Conv2D)               (None, 1000, 1000, 3 4640        max_pooling2d[0][0]              \n",
-      "__________________________________________________________________________________________________\n",
-      "max_pooling2d_1 (MaxPooling2D)  (None, 500, 500, 32) 0           conv2d_1[0][0]                   \n",
-      "__________________________________________________________________________________________________\n",
-      "conv2d_2 (Conv2D)               (None, 500, 500, 64) 18496       max_pooling2d_1[0][0]            \n",
-      "__________________________________________________________________________________________________\n",
-      "conv2d_transpose (Conv2DTranspo (None, 1000, 1000, 3 8224        conv2d_2[0][0]                   \n",
-      "__________________________________________________________________________________________________\n",
-      "concatenate (Concatenate)       (None, 1000, 1000, 6 0           conv2d_transpose[0][0]           \n",
-      "                                                                 conv2d_1[0][0]                   \n",
-      "__________________________________________________________________________________________________\n",
-      "conv2d_transpose_1 (Conv2DTrans (None, 2000, 2000, 1 4112        concatenate[0][0]                \n",
-      "__________________________________________________________________________________________________\n",
-      "concatenate_1 (Concatenate)     (None, 2000, 2000, 3 0           conv2d_transpose_1[0][0]         \n",
-      "                                                                 conv2d[0][0]                     \n",
-      "__________________________________________________________________________________________________\n",
-      "conv2d_3 (Conv2D)               (None, 2000, 2000, 1 33          concatenate_1[0][0]              \n",
-      "==================================================================================================\n",
-      "Total params: 35,953\n",
-      "Trainable params: 35,953\n",
-      "Non-trainable params: 0\n",
-      "__________________________________________________________________________________________________\n"
-     ]
-    }
-   ],
-   "source": [
-    "model.summary()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 35,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/"
-    },
-    "id": "BIK7kxOlZTxw",
-    "outputId": "bcf637b9-a78b-4e35-9a71-d0ed8c55379f"
-   },
-   "outputs": [],
-   "source": [
-    "if train:\n",
-    "    # Train the model\n",
-    "    history = model.fit(\n",
-    "        train_generator,\n",
-    "        epochs=20,\n",
-    "        use_multiprocessing=True,\n",
-    "        callbacks=[checkpoint_callback, early_stopping_callback],\n",
-    "    )\n",
-    "\n",
-    "    # Save the final model\n",
-    "    model.save(os.path.join(checkpoint_path, \"unet_with_2d_output_model.h5\"))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 61,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/",
-     "height": 449
-    },
-    "id": "85HoU8we3HsR",
-    "outputId": "c631f942-da43-426d-e686-9ab76257bd23"
-   },
-   "outputs": [],
-   "source": [
-    "if train:\n",
-    "    # Plot training history\n",
-    "    plt.plot(history.history[\"loss\"], label=\"Loss\")\n",
-    "    plt.plot(history.history[\"accuracy\"], label=\"Accuracy\")\n",
-    "    plt.xlabel(\"Epochs\")\n",
-    "    plt.ylabel(\"Metric\")\n",
-    "    plt.legend()\n",
-    "    plt.show()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "Xwvbbr-CF3wr"
-   },
-   "source": [
-    "# Model Predictions"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 64,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/",
-     "height": 617
-    },
-    "id": "MXq734JLZAha",
-    "outputId": "08290386-35c7-4c6a-fc03-48060903152f"
-   },
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 2500x1000 with 6 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAACbkAAALRCAYAAAB82DzHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAACtmUlEQVR4nOzdd3gUVeP28XtJDwkt9JbQe29SE3oHUaSJEJpUKYIIiCSA9CIIKKIGRAEBG00p0pEiSK8qVXroCCQkcN4/eHd/LLshBZDF5/u5rr0gZ87MnJndnZndufccizHGCAAAAAAAAAAAAAAAAAAAF5TseTcAAAAAAAAAAAAAAAAAAIC4EHIDAAAAAAAAAAAAAAAAALgsQm4AAAAAAAAAAAAAAAAAAJdFyA0AAAAAAAAAAAAAAAAA4LIIuQEAAAAAAAAAAAAAAAAAXBYhNwAAAAAAAAAAAAAAAACAyyLkBgAAAAAAAAAAAAAAAABwWYTcAAAAAAAAAAAAAAAAAAAui5AbAAAAAAAAAAAAAAAAAMBlEXIDAAAAAACJMmvWLFksFs2aNeuJlmOxWBQSEvJU2vS8xLUvnuW2nThxQhaLRaGhoc9k+c/S3bt3NXjwYOXKlUuenp6yWCxat27d824WAAAAAAAAABdHyA0AAAAAABdnDTVZLBZlyZJF9+7dc1pv3759tnr58+f/l1v5bFnDZA8/fHx8lD9/fr399tu6dOnS827iUxUUFKSgoKDn3Yynbvz48RoxYoSyZ8+u/v37Kyws7F/dzsuXL2vGjBlq1KiRcubMKS8vL6VNm1Z169bVihUr4pwvOjpaw4YNU968eeXt7a1MmTKpY8eOOn/+vEPdP//8UyNHjlSVKlWUOXNmeXp6Klu2bGrTpo0OHz6coHZu3bpVbm5uslgsGj16dIK3b8+ePWrXrp2KFi2qgIAAeXt7K1euXGrWrJl27NjhdJ5nEZh8+P06fvx4p3XCw8NlsVj0zTffPNV1P05S31fOjj0ZM2ZUpUqV1K9fP+3Zs+fpN/YJhIaGOrT54Ud4ePjzbqLL+K8eawEAAAAA+C9yf94NAAAAAAAACePu7q6zZ89qxYoVqlevnsP0L774Qu7u7oqNjX0Orft3VK9eXZUqVZIkRUZGasWKFfrwww/1ww8/aMeOHQoICHjOLXzg0KFD8vX1fSbLzpIliw4dOqSUKVM+k+U/Sz/99JP8/Py0cuVKeXh4/OvrX7hwobp27aosWbKoWrVqypIli06fPq3vvvtOy5cv17hx49SvXz+7ee7fv6/GjRtrxYoVKleunF555RUdPXpUM2fO1KpVq7Rt2zZlzJjRVv/999/X/PnzVbhwYTVu3FgpUqTQvn379NVXX+nbb7/VihUrVLly5TjbeOfOHYWGhsrHx0e3bt1K1PZt375dP/30k8qXL6/g4GAlT55cx44d05IlS/Ttt99q9uzZat26deJ22hMaNWqUOnbsqFSpUv2r633aAgIC1KNHD0lSTEyMLl26pJ07d2rChAmaMGGC2rdvr48//lheXl7PuaX/p0OHDsqaNatD+YvegyYAAAAAAPjfRMgNAAAAAIAXRIUKFbRnzx5FREQ4hNzu3r2rOXPmqF69elq8ePFzauGzV6NGDQ0YMMD2d0xMjGrXrq21a9dq6tSpCgsLe46t+z/Psic9Dw+PF7anvrNnzyogIOC5BNwkKW/evFq6dKnq1q2rZMn+b4CDwYMHq1y5cho0aJBatWqlzJkz26Z9+eWXWrFihVq0aKG5c+fKYrFIkmbOnKn27dvr3Xff1ZdffmmrX6dOHQ0cOFDFihWzW/c333yjli1bqkuXLjpw4ECcbXzvvfd07tw5DRgwQO+//36itq9169bq2LGjQ/mBAwdUunRp9e3bV6+//rptG561XLly6ejRoxo9enSieqRzRWnTpnXaA9q+ffvUpk0bRURE6O7du/rqq6/+/cbFoWPHjnrppZeedzMAAAAAAACeCoYrBQAAAADgBeHj46PmzZtryZIlDsNzLl68WJcuXVK7du3inP/27dsKDw9X/vz55e3trTRp0qh+/fravHmz0/pXrlxRly5dlCFDBvn6+qpMmTL64YcfHtvGvXv3qkWLFsqUKZM8PT0VGBiot956S5cvX078BieAh4eHOnfuLOlBL1aStG7dOtuQfFu2bFHt2rWVKlUqu2CPMUYRERGqWLGiUqRIIV9fX5UuXVoRERFO15PYfWGxWJz2lnT37l1NnjxZZcuWlb+/v/z8/FSwYEG9/fbbunr1qm1o2pMnT+rkyZNOhxi01nE2xOSpU6fUoUMHZcmSRZ6ensqaNas6dOigv//+26FuSEiILBaLYmNjNXz4cOXIkUNeXl7KmzevPv74Y4f6UVFRmjBhgooVK6aUKVPKz89PuXLlUsuWLbVv374494X0f0NTHj9+3G67Ht5HsbGx+vDDD1WsWDH5+PgoZcqUqlq1qpYtW+awPOtwmLNmzdKyZctUuXJl+fv7xzvsYLVq1VS/fn27gJsk5cuXT82bN1dMTIzD++Gzzz6TJI0ePdruNdSuXTsVKFBA8+fP182bN23loaGhDgE3SWrRooXy5s2rgwcPxjm87q+//qrJkydr/PjxTnvgio+3t7fT8kKFCqlAgQK6ePGibty4Ee9yzp07p169eilPnjzy8fFRmjRpVKRIEXXr1i1B81uFhoYqd+7c+uijj3TmzJkEz3f8+HF17NhR2bNnl5eXlzJlyqTQ0FCdPHnSVicqKkpFihSRh4eHtm7dajf/nTt3VLBgQXl6emr79u0Jel8lVZEiRbRy5UqlT59eX3/9tX777TfbtLt372rKlCmqXbu2smXLJi8vL6VPn16vvPKKdu3aZbecmTNnymKxaNy4cU7X89NPP8lisahXr15P1N5HPe333d27dzVx4kSVLFlSyZMnl7+/vypXrhxn+Dq+Y6LV2rVr1b59e+XLl09+fn7y8/NT6dKlNWPGDKfL3blzp5o2bWp7DWXIkEHly5e3hS2f5WsCAAAAAAA8G/TkBgAAAADAC6R9+/aaMWOG5syZYxd2iIiIUPr06dWgQQOn80VHR6t69eraunWrSpYsqd69e+vixYuaP3++Vq5cqfnz5+uVV16x1b99+7ZCQkK0b98+29CHf//9t5o3b65atWo5XcfixYvVrFkzubm5qVGjRsqWLZsOHjyoqVOnasWKFdq2bZtSp079dHfIY2zevFkjR45U1apV9eabb+rUqVOSHgTcWrdurblz5ypv3rxq1aqVPD09tWrVKnXo0EEHDx7U+PHjbctJyr5wJioqSrVr19aGDRuUJ08etWvXTl5eXvrzzz81ffp0tWnTRkFBQQoLC9OkSZMkSb1797bNH98Qg3/++acqVaqkixcvqmHDhipUqJAOHDigiIgILV26VL/++qty587tMF/Lli21bds21a1bV25ublqwYIG6d+8uDw8PderUyVavbdu2WrBggYoWLWpr+6lTp7R27VrVrl1bRYoUibNt1rY/ul3WcIwxRs2bN9f333+vvHnzqnv37rp165YWLFigBg0aaPLkyerZs6fDchcuXKiVK1eqQYMG6tatm13YLLGsvcu5u//f12VRUVHatm2b8uXLp8DAQId5atWqpcmTJ2vr1q2qWbNmktZhdfv2bYWGhiokJESdOnXSrFmzkrgljo4ePaojR44oW7Zs8Q5ze/v2bVWsWFEnTpxQrVq11KRJE929e1fHjh3TrFmz1L9/f6VIkSJB63V3d9eIESPUvHlzhYWF6fPPP493nm3btql27dq6deuWGjZsqNy5c+vEiROaM2eOfv75Z23ZskU5c+aUt7e35s2bpzJlyqhVq1bavXu3rV19+vTRoUOHNGrUKJUpU0bXrl1L8vsqIdKlS6cuXbpo2LBhmj9/vsqWLSvpQTi2d+/eqly5surVq6fUqVPr2LFjWrx4sX7++Wdt2LBBZcqUkSQ1b95cffr00eeff6533nnHYR3Wfeesp76ketrvu+joaNWpU0fr1q1TiRIl1KFDB8XExGjZsmVq3LixpkyZYhvyVUrYMdF6zhgzZoz++usvvfTSS2rSpImuXbum5cuXq3Pnzjpy5IgmTJhgW+7u3btVoUIFubm5qXHjxgoMDNS1a9d04MABffbZZxowYIBSpUr1TF8TAAAAAADgGTAAAAAAAMClHT9+3EgytWvXNsYYU6hQIVO0aFHb9NOnTxs3NzfTt29fY4wxkky+fPnsljFs2DAjybz++uvm/v37tvI9e/YYLy8vkzp1anPjxg1beVhYmJFkOnXqZLecFStWGElGkpk5c6at/NKlSyZFihQma9as5uTJk3bzzJ0710gyPXr0sCuXZIKDgxO0D2bOnGkkmVGjRtmV371714SEhBhJJjw83BhjzNq1a21t/OKLLxyWNWPGDCPJdOjQwcTExNjKo6OjTcOGDY0ks2PHjiTvi7i27Z133jGSzBtvvGFiY2Ptpl27ds3cvHnT9ndgYKAJDAx0ui+sr4e2bdvalVerVs1IMp9++qld+aeffmokmerVq9uVBwcHG0mmXLly5vr167byw4cPG3d3d7vX0LVr14zFYjGlS5d2aHtsbKy5evWq07Y+Kq7tmj17tm2fRUdH28r//vtvkz59euPh4WGOHTtmK7e+HiwWi1m1alWC1v04N27cMBkyZDDe3t7m0qVLtvL9+/cbSaZBgwZO55s6daqRZKZNmxbvOrZt22YkmTJlyjid/tZbb5nkyZPbtjOu13xC7Nq1y4SFhZlBgwaZ119/3fj7+xtfX1+zbNmyeOddvHixkWT69OnjMO3GjRt2z09cHm77/fv3TenSpY2bm5s5ePCgrY71fTVv3jxb2d27d01QUJDx9/c3u3fvtlvmxo0bjZubm8NzYX0OWrVqZYwx5scffzSSTNWqVc29e/fs6j7uffU4zo6pj1q9erWRZCpXrmwri4qKMqdPn3aou3//fuPn52dq1KhhV969e3cjyaxfv96u/MKFC8bDw8OUK1cuQe1t27at7RgXFhZm93j49fS033eDBg2yHYsfPs/cuHHDlC5d2nh6epozZ87YyhNzTHy4HVYxMTGmZs2axs3Nze688/bbbxtJZtGiRQ7zPPz+NibprwkAAAAAAPDvY7hSAAAAAABeMO3atdPevXv1+++/S3owhNy9e/fUvn37OOeZNWuWPDw8HIZcLFq0qEJDQ3X16lUtWrTIVj579mx5enpq2LBhdsupVauWqlev7rD82bNn68aNGxo1apSyZ89uN61ly5YqWbKkvvnmmyRt78N++eUXhYeHKzw8XD169FCBAgW0bt065ciRQ2+99ZZd3RIlSjjdJ1OnTlXy5Mk1depUux61PD09NWLECEnSvHnz7LYtMfvCmXv37unTTz9VypQpNXnyZLm5udlNtw7/mVR///231qxZo4IFC9r1viZJnTp1UoECBbR69Wqnw5aOGjXKrmeufPnyqWLFijpy5IithyaLxSJjjLy8vBza7ubmplSpUiW57ZJsvZaNHTtWnp6etvKsWbOqT58+iomJ0Zw5cxzme/nll1WjRo0nWrckdenSRRcuXNCgQYMUEBBgK79+/bokxdn7mXW/WevF5fr162rbtq2SJUumsWPHOkxfv369pk6dqpEjRypHjhxJ3Qyb3bt3a+jQoRo5cqTmzJkjX19f/fDDD6pXr16Cl+Hj4+NQ5u/vb/f8JITFYtGYMWN07949DRw48LF1ly5dqhMnTqh///4OQ75WqlRJjRs31k8//WQ3ZGr37t3VsGFDzZ07V2PHjlWHDh2UJk0azZ4922FY2mcpc+bMkmQ3FK2Xl5eyZMniULdQoUKqWrWqNmzYoJiYGFu5dejlR3u8+/LLLxUTE+Pw3o7PF198oaFDh9o9rMN1Sk/3fXf//n198sknyp07t4YMGWJ3nvH399eQIUN09+5dff/995ISf0x09r5wd3dXly5ddO/ePa1du9ZhurPX8MPvbwAAAAAA8GJhuFIAAAAAAF4wb7zxhgYOHKiIiAiVKlVKs2bNUrly5VSwYEGn9W/cuKFjx46pQIECypo1q8P0kJAQffrpp9q9e7dat26tmzdv6vjx4ypYsKAyZszoUL9y5cpavXq1XdnWrVtt//71118O80RFRenSpUu6dOmS0qZNm5TNliStXr3atm4vLy8FBQXp7bff1sCBA5UmTRq7utYhAx92+/Zt7du3T5kzZ7YLe1hZAyeHDx+WpCTtC2cOHz6sGzduqEaNGs9kyNZdu3ZJkoKDg+3CJdKDkFGVKlV06NAh7dmzR9myZbObXrJkSYflWV8n165dk7+/v1KkSKE6depo+fLlKlmypJo2barKlSurXLlyiQ49xdV+Hx8fp8+ZdejA3bt3O0xzVj+xBg0apLlz56pOnToaNGjQEy/vUVFRUXrllVd0+PBhjRgxwmEoxFu3bql9+/YqX7683VCOTyI0NFShoaGKiorSn3/+qQkTJqhu3boaM2aM+vXr99h5q1SpoowZM2rUqFHavXu36tevr0qVKqlIkSIOr62EqlatmmrVqqVFixZp8+bNqlChgtN61uPI4cOHFR4e7jD9/Pnzun//vv744w+VLl3aVh4REaGiRYvq3XfflSR99913To91z5Ixxmn57t27NXbsWG3atEnnz5+3C7VJD0JxmTJlkiQVKVJE5cuX17fffqspU6bYwpURERHy8/NT8+bNE9WmLVu26KWXXopz+tN83x05ckRXr15V5syZNXToUIfpkZGRkv7v2JrYY+LNmzc1fvx4/fjjjzp69Khu3bplN/3s2bO2/zdt2lSTJk3Syy+/rGbNmqlmzZqqVKmSQwAbAAAAAAC8WAi5AQAAAADwgkmfPr3q1aunefPmqVGjRvrrr78eG1yx9nqUIUMGp9Ot4S1rb1TWf9OnT++0vrPlXLlyRZI0bdq0x7b91q1bTxRyGzVqlAYMGJCgus7aefXqVRljdObMGadBjIfbKSVtXzhz7do1SXLaq9PTkNjn+GHOeimz9nB37949W9m3336rkSNHat68eXrvvfckPeihqX379ho5cqR8fX2fqP2Phu8S0vaE7v+4DB06VKNGjVK1atX0/fffO+1NKq51S/+33+Pq6S06OlpNmjTRmjVrNHDgQKchuvfee09nz57VTz/99NR7HvP29laRIkU0a9YsRUZG6t1331WdOnVUuHDhOOdJmTKltmzZorCwMC1ZskQ//fSTpAfBx4EDB6pbt25JasuYMWO0atUqvfvuu9q4caPTOtbjiLPewx72aMApbdq0qly5shYsWKDAwEA1atQoSW18EufOnZMkpUuXzla2efNmVatWTdKDnh/z5MkjPz8/WSwW/fjjj9qzZ4+io6PtlvPmm2+qXbt2mjNnjrp166ZNmzbp8OHD6tSp0xP19ujM03zfWZ+7AwcO6MCBA3Gu0/rcJeaYePfuXYWEhGjnzp0qUaKE3njjDQUEBMjd3V0nTpzQl19+abcfy5cvrzVr1mjUqFGaN2+erce6UqVKady4capatWq86wQAAAAAAK6H4UoBAAAAAHgBtW/fXlevXlWHDh3k4+Ojli1bxlnXOqTihQsXnE63llvrWf+9ePHiY+s7W8e+fftkjInzERgYmMAtfHLOep2ytrNUqVKPbad16Luk7AtnrMN5njlzJrGbkSCJfY6TInny5BoxYoSOHTumY8eO6YsvvlD+/Pk1efJk9enTJ8nLtbYrKW1Pas9i0oOAW3h4uEJCQrRkyRKnQxvmypVLyZIl059//ul0GdbyPHnyOEyLiopS48aNtXz5cvXv318jR450uozdu3crKipK+fPnl8VisT3atWsnSRo4cKAsFot69+6dxC19oFatWrp//36cAbOHBQUF6csvv1RkZKR27dqlMWPGyBij7t272w3lmxjFixdXy5YttWnTJi1ZssRpHetzvGTJkse+P4ODg+3mW7hwoRYsWKCAgACdPHlSYWFhSWrjk1i3bp0kqUyZMrayESNGKDo6WqtXr9bixYs1YcIE2+vOWc+QktS8eXOlSpXKNmSp9d/EDlWaEE/zfWet9+qrrz72uZs5c6akxB0TFy1apJ07d6pjx47auXOnPvnkE33wwQcKDw9XnTp1nM4THBys5cuX6+rVq1q7dq3efvttHThwQPXr19fRo0fjXScAAAAAAHA9hNwAAAAAAHgB1atXTxkzZtSZM2f06quvPja8lCJFCuXMmVN//fWX00DB+vXrJT0IoVjr58iRQ3/99ZfOnz/vUN9ZSKZcuXKSHgyP58r8/f1VoEABHTp0yNaT0OMkZV84ky9fPqVIkULbt2/X1atX463v5uZm14tafKzP3YYNGxyGTTTG2NpprfekcuTIofbt22v9+vXy8/PT4sWLn2h5JUqU0J07d/Tbb785THv09fk0hIeHKzw8XMHBwVq2bFmcvdB5e3urbNmyOnLkiE6ePOkwfeXKlfLy8rK9/q2ioqL08ssva8WKFerXr5/GjBkTZ1vq16+vDh06ODyqVKki6UFoqkOHDipfvvwTbPH/Dedo7aUvIdzc3FS8eHH179/fFm57kuf6gw8+kKenpwYOHKj79+87TE/KceTUqVN68803lT59eu3evVulS5fW6NGjba+bhyX2fZVQkZGR+vTTTyVJLVq0sJUfPXpUadKkUcWKFe3q3759Wzt37nS6LB8fH7Vu3Vq7du3S+vXrtXDhQhUtWtQuPPe0PM33XYECBZQiRQrt2LHDYUhWZxJzTLSG0pz10BffMdjHx0chISGaMGGCBg0apDt37uiXX36xTX9WrwkAAAAAAPD0EXIDAAAAAOAF5O7ursWLF+uHH37QiBEj4q3ftm1bxcTEaODAgXYhqP3792vmzJlKmTKlXn75ZVv5G2+8obt372rIkCF2y1m5cqVWr17tsPx27drJ399f7733ntOh6m7fvq2tW7cmYgufnZ49e+r27dvq1KmTw7CHknT8+HGdOHHC9ndi94Uz7u7u6ty5s65fv65evXo5hCquX7+uf/75x/Z3mjRpdOnSJUVFRSVo+dmzZ1fVqlV14MABRURE2E2LiIjQgQMHVK1atTiHJoxPZGSk0yDM1atXFR0d7bQXtMRo27atpAe9lj0ckDlz5owmTpwod3d3vf7660+0DquwsDANHTpUlStXfmzAzerNN9+UJA0YMMDuvTNz5kwdOnRIzZs3twuZWntwW7Fihd5++22NGzfusct/55139Pnnnzs8rD25vfLKK/r888/VvHnzeLft119/VWxsrEP57t27NX36dLm7u6tmzZqPXcb+/fudBvqsPXs9yXOdI0cOdenSRQcOHNDcuXMdpjdu3FjZs2fXxIkTtWHDBofpMTEx2rRpk+3v+/fvq3Xr1rp27ZpmzZqlrFmzau7cufLx8dEbb7zhEJ5K7PsqIfbv369atWrp4sWLCg0NVenSpW3TAgMDdfXqVbtj4r1799SvXz9FRkbGuczOnTtLklq1amU7Vj0LT/N95+7urq5du+rkyZPq16+f06Db/v37bb1iJuaYaO0B9OHnXnoQxPvss88c1rNx40bbUMIPc/YafhavCQAAAAAA8Gwk/KebAAAAAADApZQpUybBvfv0799fy5Yt01dffaVDhw6pevXqioyM1Pz58xUTE6PZs2fL39/frv7333+vzz77TAcOHFCVKlX0999/a8GCBapfv76WLVtmt/x06dJp3rx5eu2111SsWDHVqVNH+fPnV1RUlE6ePKn169erQoUKWr58+VPdB0nRuXNnbd26VV9++aV+/fVX1ahRQ5kzZ9aFCxd0+PBhbdu2TXPnzlVQUJCkxO+LuAwbNkxbt27VV199pa1bt6pu3bry8vLSsWPHtHz5cm3atMnWa1K1atW0Y8cONWzYUJUrV5anp6cqVaqkSpUqxbn8Tz75RJUqVVKnTp20ZMkSFSxYUAcPHtTixYuVLl06ffLJJ0neZ2fOnFG5cuVUqFAhlSxZUlmyZNHly5e1aNEixcTEqH///kletvQgSPj9999r0aJFKlq0qBo0aKBbt25pwYIFunz5siZMmKCcOXM+0TokadasWRo2bJjc3d1VtmxZpwG0kJAQhYSE2P5u06aN5s+fr2+++UbHjx9XSEiIjh07pu+++07ZsmVz6KWtS5cuWrlypTJmzCh/f3+Fh4c7rCM0NNT2+nqaunfvrsjISFWsWFHZs2dXbGysjhw5opUrV8oYo4kTJ8a73l9++UV9+/ZVxYoVlT9/fgUEBOjYsWNavHixfHx81KNHjydq4/vvv6+ZM2c6HTLSy8tL3377rerWravg4GBVr15dhQsXlvSgx7aNGzcqICBAhw8flvRgONCNGzeqZ8+eqlu3rqQHQ8d+9NFH6tChg958800tXLjQtvykvK+sLl26ZHsuY2NjdfnyZf3+++/avn27JKljx46aNm2a3TxvvfWWVq5cqUqVKqlZs2by9vbWunXrdObMGYWEhNiGOH1U4cKFVaFCBW3evFne3t5q3bp1vO1Liqf9vhs6dKh27typjz76SMuWLVNwcLDSpUunM2fOaN++fdqzZ4+2bNmi9OnTS0r4MbFhw4YKCgrS2LFjtX//fhUuXFhHjhzR0qVL9fLLL+u7776za8eECRO0atUqVa1aVTlz5pS3t7d27typ1atXK3fu3GrSpImt7pO8JgAAAAAAwL/MAAAAAAAAl3b8+HEjydSuXTtB9SWZfPnyOZT/888/5v333zd58+Y1np6eJlWqVKZu3bpm48aNTpdz+fJl8+abb5p06dIZb29vU6pUKfP999+bmTNnGklm5syZDvMcPnzYdOjQwQQGBhpPT0+TOnVqU6RIEdOzZ0/z22+/ObQzODg4QdtkXeeoUaPirbt27VojyYSFhT223vz5802NGjVM6tSpjYeHh8mSJYsJCQkxEyZMMJGRkXZ1E7sv4tq2qKgoM378eFO8eHHj4+Nj/Pz8TMGCBU3fvn3N1atXbfVu3rxpOnXqZDJlymSSJUtmtz3W10Pbtm0dln/ixAnTrl07kylTJuPu7m4yZcpk2rVrZ06cOOFQNzg42MT11VDbtm2NJHP8+HFjjDFXr1414eHhpkqVKiZTpkzG09PTZM6c2dSpU8esWLEizn38qMDAQBMYGOh0WkxMjBk/frwpUqSI8fLyMv7+/iY4ONgsWrTIoe7jXoOPExYWZiQ99uHsdRMVFWWGDh1qcufObTw9PU2GDBlM+/btzdmzZx3qWvfr4x5r166Nt62Jec1bzZ4927z88ssmMDDQ+Pj4GE9PTxMYGGhatWplNm/enKBlHDx40PTq1cuUKFHCBAQEGC8vL5MzZ04TGhpqDh48mKBlxNf2YcOG2fbFvHnzHKafPn3a9OrVy+TJk8d4eXmZFClSmAIFCpiOHTua1atXG2OM2bJli3F3dzeFCxc2d+7ccVhG06ZNjSTz2Wef2coe9756nEefPy8vL5M+fXpTsWJF069fP7Nnz5445/32229NyZIlja+vr0mbNq1p1qyZOXr0qMN77FGffvqpkWRat24db/seZV32li1b4q37tN93sbGx5tNPPzUVK1Y0KVKkMF5eXiZ79uymTp065pNPPjH//POPXf2EHhOPHTtmXn31VZMuXTrj6+trypQpY7755hunx/vly5ebNm3amHz58hl/f3/bMgcPHmwuXbpkt/6kviYAAAAAAMC/z2LMQ+MsAAAAAAAAAACeq27duumTTz7R+vXrVaVKlefdHAAAAAAAgOeOkBsAAAAAAAAAuIjIyEjlyJFD2bNn18GDB593cwAAAAAAAFyC+/NuAAAAAAAAAAD8r1u2bJl27typb7/9Vrdu3VJYWNjzbhIAAAAAAIDLIOQGAAAAAAAAAM/ZwoUL9eWXXypz5swaOXKkmjdv/rybBAAAAAAA4DIYrhQAAAAAAAAAAAAAAAAA4LKSPe8GAAAAAAAAAAAAAAAAAAAQF0JuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAAAAAAAAAAAAAAAcFmE3AAAAAAAAAAAAAAAAAAALouQGwAAAAAAAAAAAAAAAADAZRFyAwAAAAAAAAAAAAAAAAC4LEJuAAAAAAAAAAAAAAAAAACXRcgNAAAAAAAAAAAAAAAAAOCyCLkBAAAAAAAAAAAAAAAAAFwWITcAAAAAAAAAAAAAAAAAgMsi5AYAAAA7FoslQY9169Y913aGhITIYrGoTp06DtNOnDghi8Wi8ePHP7P1b968WeHh4bp27VqC6oeHh9vtP19fX2XNmlW1a9fWlClTdPPmzWfW1ofbEBQU9FSWdevWLY0ZM0bFihVTihQp5O/vr1y5cqlZs2Zav369rd66detksVj07bff2spmzZrl0q8tAHjRPHpcdXd3V9asWdWuXTudOXPmX2lDUFCQQkNDbX9bj/+JPaYn9vyaGKGhoQk6D3KNkXhcYwDAf4P1OLpjx47n3ZTn7uDBgwoPD9eJEycSVP9/5XosJCREISEhT9ROAHAle/fuVbt27ZQjRw55e3vLz89PJUuW1NixY3XlyhVbvaCgIDVo0OA5tvTfMWXKFOXOnVuenp6yWCzP5LP507Zx40Y1a9ZMWbJkkaenp1KmTKkKFSrok08+0a1bt55385Lsp59+Unh4eILrG2P0zTffqHLlykqfPr28vb1t3w18/vnndnUtFot69Ohh+9v6XUdcj/ja8ay+y7FeX3l7e+vkyZMO00NCQlS4cOGnus6H3b59W+Hh4Qm+lnp0P3p4eCggIEBlypRRnz59dODAgWfW1v8l7s+7AQAAAHAtW7Zssft7+PDhWrt2rdasWWNXXrBgwX+zWXFasWKF1qxZo2rVqv2r6928ebOGDh2q0NBQpUqVKsHzLV++XClTptTdu3d19uxZrV69Wv3799e4ceO0ZMkSFStW7Nk1+im5d++eatWqpX379umdd95R2bJlJUl//vmnlixZoo0bNyo4ODje5cycOVP58+d3KHeV1xYAvGisx9U7d+5ow4YNGjVqlNavX699+/YpefLk/2pbSpYsqS1btiT6mJ7U8+uzwDXGv49rDACAqzh48KCGDh2qkJCQRAW5/+vXYx9//PFTbCEAPF+fffaZunXrpnz58umdd95RwYIFFRMTox07dmj69OnasmWLfvjhh+fdzH/N7t271bNnT3Xs2FFt27aVu7u7/P39n3ezHissLEzDhg1ThQoVNHz4cOXKlUu3b9+2ha7++OMPffjhh8+7mUny008/adq0aQkOug0cOFBjxoxRp06d9M4778jf318nT57UmjVrtGjRInXs2DHeZbz11ltq1aqVQ3nWrFkfO9+z/i4nOjpagwcP1ldfffXUl/04t2/f1tChQyUpUSF/6368f/++rl27pl27dikiIkJTpkzRqFGj9M477zyjFv9vIOQGAAAAOy+99JLd3+nSpVOyZMkcyh91+/Zt+fr6PsumOcibN69iY2PVv39/bd++XRaL5V9df1KUKlVKadOmtf3dokUL9ejRQ8HBwWrUqJH++OMPeXl5PccWxm/Dhg3avHmzIiIi1K5dO1t57dq11aNHD92/fz9ByylcuLBKly79rJoJAP9zHj6uVq1aVffu3dPw4cP1448/6vXXX3c6z7M6f6dIkSLeawdXxjXG88E1BgDgce7cuSMfHx+H8piYGFvvac/bf/16jMA4gP+KLVu2qGvXrqpZs6Z+/PFHu89KNWvWVN++fbV8+fLn2MJ/n7WXqU6dOtl+cBSX53Ev4FELFy7UsGHD1KFDB3322Wd2n9vr1q2r/v37O/ygP6ket71xXZ/8m+7cuaNJkyapTZs2mjFjht200NDQBH+Wzp49u0t+l1OnTh3NnTtX/fr1eyF+wPfofqxXr57efvttvfLKK+rfv78KFy6sunXrPscWvtgYrhQAAACJZu0GesOGDapQoYJ8fX3Vvn17SYqz++pHh82QpPPnz6tz587KmjWrPD09lSNHDg0dOlSxsbEJaoeHh4dGjBih33//XfPnz4+3fnzrM8aoXr16CggI0KlTp2zz3b59W4UKFVKBAgV069YthYeH235tkyNHjiceAqtYsWJ67733dOrUKbvtWLVqlRo3bqysWbPK29tbuXPnVufOnXXp0iVbnY0bN8pisWjevHkOy509e7YsFou2b98e57oXLlyocuXKKWXKlPL19VXOnDltz2VcLl++LEnKlCmT0+nJkvExAwBcgfULNeuQDqGhofLz89O+fftUq1Yt+fv7q3r16pKku3fv6oMPPlD+/Pnl5eWldOnSqV27doqMjLRbZkxMjPr376+MGTPK19dXlSpV0m+//eaw7riGx9q2bZsaNmyogIAAeXt7K1euXOrdu7ckJej8On/+fJUvX17JkyeXn5+fateurV27djmsf9asWcqXL5+8vLxUoEABzZ49O1H7jmsMrjEAAA9Yrx/++usv1atXT35+fsqWLZv69u2r6Ohou7rR0dEaNmyYChQoIG9vbwUEBKhq1aravHmzrU5UVJQGDhyoHDlyyNPTU1myZFH37t0dhreyDgv3/fffq0SJEvL29tbQoUNt1xhfffWV+vbtqyxZssjLy0t//fWXJOmXX35R9erVlSJFCvn6+qpixYpavXq1w3YdPnxYLVu2VIYMGeTl5aXs2bOrTZs2io6O1qxZs/Taa69JehBUs56PZ82alej991+7HnM2XOmVK1fUrVs32zBxOXPm1Hvvvefw+rAOjfbVV1+pQIEC8vX1VbFixbR06VK7epGRkXrzzTeVLVs2236oWLGifvnll4TtdABIgJEjR8pisWjGjBlOfwzk6empRo0aOZQvX75cJUuWlI+Pj/Lnz6+IiAi76ZGRkerWrZsKFiwoPz8/pU+fXtWqVdPGjRvt6lmHNRw/frwmTpyoHDlyyM/PT+XLl9fWrVsd1vvZZ58pb9688vLyUsGCBTV37lyFhoY69Daa0HPJo0JCQtS6dWtJUrly5WSxWGzf5T/uXsCpU6fUunVrpU+f3vb5e8KECXahKuu2jhs3TmPGjFFQUJB8fHwUEhKiP/74QzExMRowYIAyZ86slClTqkmTJrp48eJj2ytJw4YNU+rUqfXRRx85/WGav7+/atWqZdcGZ+fyR+9nhIeHy2KxaOfOnWratKlSp06tXLlySYr7+kRK2L2OhD7voaGhmjZtmq191kdcw6jfunVL0dHRz+WzdHzXDvfv39fYsWNtr8n06dOrTZs2On36dILX0b9/fwUEBOjdd9+Nt64xRh9//LGKFy8uHx8fpU6dWk2bNtWxY8dsdb755htZLBZNnTrVbt6wsDC5ublp1apVOnHihNKlSydJGjp0qG27Hr3HlVA+Pj764osv5OHhoXHjxtnKE3LMMMYoT548ql27tsNy//nnH6VMmVLdu3dPUrteRM//Zy0AAAB4IZ07d06tW7dW//79NXLkyER/UDp//rzKli2rZMmSaciQIcqVK5e2bNmiDz74QCdOnNDMmTMTtJzmzZtr/PjxGjx4sF599VV5eHgkeX3WL8qLFy+uZs2aaePGjfLw8FC3bt10/Phxbdu2TcmTJ1fHjh115coVTZkyRd9//73tw+OT/KK5UaNG6t+/vzZs2KA2bdpIko4ePary5curY8eOSpkypU6cOKGJEyeqUqVK2rdvnzw8PFS5cmWVKFFC06ZNU8uWLe2WOXXqVJUpU0ZlypSR9OAD58Mf2Lds2aLmzZurefPmCg8Pl7e3t60L88cpXbq0PDw81KtXLw0ZMkTVqlWL8wP049y7d88h0GixWOTm5pboZQEAHFlv9lq/lJMefOHdqFEjde7cWQMGDFBsbKzu37+vxo0ba+PGjerfv78qVKigkydPKiwsTCEhIdqxY4ftV8mdOnXS7Nmz1a9fP9WsWVP79+/XK6+8ops3b8bbnhUrVqhhw4YqUKCAJk6cqOzZs+vEiRNauXKlJMV7fh05cqQGDx6sdu3aafDgwbp7967GjRunypUr67fffrPVmzVrltq1a6fGjRtrwoQJun79usLDwxUdHZ2o6xWuMbjGAAA8EBMTo0aNGqlDhw7q27evNmzYoOHDhytlypQaMmSIJCk2NlZ169bVxo0b1bt3b1WrVk2xsbHaunWrTp06pQoVKsgYo5dfflmrV6/WwIEDVblyZe3du1dhYWHasmWLtmzZYhc02Llzpw4dOqTBgwcrR44cSp48uW7duiXpwZBc5cuX1/Tp05UsWTKlT59eX3/9tdq0aaPGjRvryy+/lIeHhz799FPVrl1bK1assIXJ9uzZo0qVKilt2rQaNmyY8uTJo3Pnzmnx4sW6e/eu6tevr5EjR2rQoEGaNm2aSpYsKUm2G9yJ8V+7HntUVFSUqlatqqNHj2ro0KEqWrSoNm7cqFGjRmn37t1atmyZXf1ly5Zp+/btGjZsmPz8/DR27Fg1adJER44cUc6cOSVJb7zxhnbu3KkRI0Yob968unbtmnbu3GkLwwPAk7p3757WrFmjUqVKKVu2bAmeb8+ePerbt68GDBigDBky6PPPP1eHDh2UO3duValSRdKD4K/0ICyTMWNG/fPPP/rhhx8UEhKi1atXOwSFp02bpvz582vSpEmSpPfff1/16tXT8ePHlTJlSknSjBkz1LlzZ7366qv68MMPdf36dQ0dOtQhTJyYc8mjPv74Y82bN08ffPCBbejth89dzu4FREZGqkKFCrp7966GDx+uoKAgLV26VP369dPRo0cdhrieNm2aihYtqmnTpunatWvq27evGjZsqHLlysnDw0MRERE6efKk+vXrp44dO2rx4sVxPhfnzp3T/v371bx582fWo9wrr7yiFi1aqEuXLrbrD8n59Uli73XE97y///77unXrlr799lu73uji+nycNm1a5c6dWx9//LHSp0+vevXqKV++fInulf7+/ftOOyB4XG+58V07dO3aVTNmzFCPHj3UoEEDnThxQu+//77WrVunnTt32vVIHxd/f38NHjxYvXr10po1a1StWrU463bu3FmzZs1Sz549NWbMGF25csU2pO2ePXuUIUMGtWjRQuvXr1ffvn310ksvqXTp0lqzZo0++OADDRo0SDVr1lR0dLSWL1+uOnXqqEOHDrbhXh9+XyRW5syZVapUKW3evFmxsbFyd3dP0DHDYrHorbfeUu/evfXnn38qT548tmXOnj1bN27c+J8KuckAAAAAj9G2bVuTPHlyu7Lg4GAjyaxevdqhviQTFhbmUB4YGGjatm1r+7tz587Gz8/PnDx50q7e+PHjjSRz4MCBx7YrODjYFCpUyBhjzC+//GIkmSlTphhjjDl+/LiRZMaNG5ek9W3atMm4u7ub3r17m4iICCPJfP7553bzjRs3zkgyx48ff2w7rcLCwowkExkZ6XT6nTt3jCRTt25dp9Pv379vYmJizMmTJ40ks2jRItu0mTNnGklm165dtrLffvvNSDJffvllnG2ybvu1a9cStA0P++KLL4yfn5+RZCSZTJkymTZt2pgNGzbY1Vu7dq2RZBYuXOjQXmcPNze3RLcFAP7XWY+rW7duNTExMebmzZtm6dKlJl26dMbf39+cP3/eGPPgnC7JRERE2M0/b948I8l89913duXbt283kszHH39sjDHm0KFDRpLp06ePXb05c+YYSXbneevxf+3atbayXLlymVy5cpk7d+7EuS1xnV9PnTpl3N3dzVtvvWVXfvPmTZMxY0bTrFkzY4wx9+7dM5kzZzYlS5Y09+/ft9U7ceKE8fDwMIGBgXGu24prDK4xAOB/lfU4un37dluZ9fphwYIFdnXr1atn8uXLZ/t79uzZRpL57LPP4lz+8uXLjSQzduxYu/L58+cbSWbGjBm2ssDAQOPm5maOHDliV9d6/K9SpYpd+a1bt0yaNGlMw4YN7crv3btnihUrZsqWLWsrq1atmkmVKpW5ePFinG1duHChw7XM4/wvXI8Z8+A6KTg42Pb39OnTnb4+xowZYySZlStX2sokmQwZMpgbN27Yys6fP2+SJUtmRo0aZSvz8/MzvXv3jrN9APCkzp8/bySZFi1aJHiewMBA4+3tbfe5786dOyZNmjSmc+fOcc4XGxtrYmJiTPXq1U2TJk1s5dbPlkWKFDGxsbG2cuvnrXnz5hljHpzHMmbMaMqVK2e33JMnTzp8xk3ouSQuzq4DjIn7XsCAAQOMJLNt2za78q5duxqLxWI7h1u3tVixYubevXu2epMmTTKSTKNGjezm7927t5Fkrl+/Hmdbt27daiSZAQMGPHabrKxtmDlzpsO0R+9nWD9jDxkyxKFuXNcnCf1eIKHPuzHGdO/e3SQmzvPbb7+Z7Nmz2z4D+/v7mwYNGpjZs2fbfT9i3ebu3bvb/ra2K67Hxo0bH7vuuK4drNct3bp1syvftm2bkWQGDRr02OU+/JqMjo42OXPmNKVLl7Ztz8Pf3xhjzJYtW4wkM2HCBLvl/P3338bHx8f079/fVhYVFWVKlChhcuTIYQ4ePGgyZMhggoOD7Z6XyMjIOO93OePsO6NHNW/e3EgyFy5ccDo9rmPGjRs3jL+/v+nVq5dd/YIFC5qqVasmqH3/FfTxDwAAgCRJnTr1Y38xE5+lS5eqatWqypw5s2JjY22PunXrSpLWr1+f4GVVr15dtWrV0rBhw+L85XJi1lexYkWNGDFCkyZNUteuXdW6dWt16NAhyduaEMYYh7KLFy+qS5cuypYtm9zd3eXh4aHAwEBJ0qFDh2z1WrZsqfTp09u6MJekKVOmKF26dGrevHmc67T2vtKsWTMtWLBAZ86cSXB727dvr9OnT2vu3Lnq2bOnsmXLpq+//lrBwcF23W0/zuzZs7V9+3a7x7Zt2xLcBgCAvZdeekkeHh7y9/dXgwYNlDFjRv3888/KkCGDXb1XX33V7u+lS5cqVapUatiwod05snjx4sqYMaNtiIm1a9dKkl5//XW7+Zs1a/bYX/VK0h9//KGjR4+qQ4cO8vb2TvS2rVixQrGxsWrTpo1dG729vRUcHGxr45EjR3T27Fm1atXK7hfLgYGBqlChQqLXyzUG1xgAgAe9YTZs2NCurGjRorYhOCXp559/lre392OHprb26PnoME+vvfaakidP7jCsaNGiRZU3b16ny3r0embz5s26cuWK2rZta3c+vn//vurUqaPt27fr1q1bun37ttavX69mzZo9UU8ccfkvX485s2bNGiVPnlxNmza1K7c+x48+p1WrVpW/v7/t7wwZMih9+vR2r6WyZctq1qxZ+uCDD7R161bFxMQ8lbYCwJMqXry4smfPbvvb29tbefPmtTuGSdL06dNVsmRJeXt72z5vrV692u6zllX9+vXtepwuWrSopP8b5vrIkSM6f/68mjVrZjdf9uzZVbFiRbuyhJ5LksLZvYA1a9aoYMGCKlu2rF15aGiojDEOPXnXq1fPrnf1AgUKSHqwDx5mLT916lSS2/s0PHqutnJ2fZLYex3xPe9JUaZMGf31119avny5Bg0apPLly2v16tVq06aNGjVq5PS7gUf16tXL4bP09u3bVbx48SS1yXrd8ui1X9myZVWgQAGnQ8rHxdPTUx988IF27NihBQsWOK2zdOlSWSwWtW7d2u55yJgxo4oVK2b3HvDy8tKCBQt0+fJllSxZUsYYzZs375n3AO/seUjIMcPf31/t2rXTrFmzbD0LrlmzRgcPHlSPHj2eaZtdDSE3AAAAJElSho562IULF7RkyRJ5eHjYPQoVKiRJunTpUqKWN2bMGF26dEnjx49/Kut7/fXX5enpqejoaL3zzjtJ2MLEsX6AzZw5s6QHXYPXqlVL33//vfr376/Vq1frt99+09atWyVJd+7csc3r5eWlzp07a+7cubp27ZoiIyO1YMECdezY0W6ol0dVqVJFP/74oy00kDVrVhUuXFjz5s1LUJtTpkypli1bavLkydq2bZv27t2rDBky6L333tO1a9finb9AgQIqXbq03aNUqVIJWjcAwJE12LNr1y6dPXtWe/fudfjS29fXVylSpLAru3Dhgq5duyZPT0+H8+T58+dt50jrEFEZM2a0m9/d3V0BAQGPbVtkZKQkKWvWrEnatgsXLkh68KXto22cP39+vG2MqywhuMbgGgMA/tf5+vo6hKK8vLwUFRVl+zsyMlKZM2d+7NDgly9flru7u0O4zGKxKGPGjA7DUT7ue4dHp1mvFZo2bepwTh4zZoyMMbpy5YquXr2qe/fuJfmaJD7/5esxZy5fvqyMGTM6DIeWPn16ubu7Ozynztro5eVld/0xf/58tW3bVp9//rnKly+vNGnSqE2bNjp//vxTazeA/21p06aVr6+vjh8/nqj5EnIMmzhxorp27apy5crpu+++09atW7V9+3bVqVPHrl5cy7R+zrLWtR5HHw1LOytL6LkkKZydky9fvuy03PrZ89FzQJo0aez+9vT0fGz5w9cZj7KGDRP7HCZGXNchzsoT+71AfM97Unl4eKh27doaMWKEVqxYob///lshISFaunSpfv7553jnz5o1q8Nn6dKlS8vPzy9J7bG+BuJ6nSR2KPIWLVqoZMmSeu+995yG4C9cuCBjjDJkyODwXGzdutXhecidO7cqV66sqKgovf766098zyshTp48KS8vL9vrPjHHjLfeeks3b97UnDlzJElTp05V1qxZ1bhx42feblfy+J91AAAAAHF49AtMKy8vL0VHRzuUP/qBJW3atCpatKhGjBjhdDnWD8MJVbx4cbVs2VITJ05UvXr1HKYnZn337t3T66+/rtSpU8vLy0sdOnTQr7/+avuA/SwsXrxYkhQSEiJJ2r9/v/bs2aNZs2apbdu2tnp//fWX0/m7du2q0aNHKyIiQlFRUYqNjVWXLl3iXW/jxo3VuHFjRUdHa+vWrRo1apRatWqloKAglS9fPlHbUKhQIbVo0UKTJk3SH3/84fArPgDAs2UN9jyOs/N32rRpFRAQoOXLlzudx9rbhvVL2PPnzytLliy26bGxsfF+MWm9mX369OnH1otL2rRpJUnffvutrccxZx5u46OSemOUawyuMQAA8UuXLp02bdqk+/fvxxl0CwgIUGxsrCIjI+2CbsYYnT9/3tYTqFVc3zs4m2a9VpgyZYpeeuklp/NkyJBB9+7dk5ubW5KvSeLzX74ecyYgIEDbtm2TMcZuuy5evKjY2Fjb85IYadOm1aRJkzRp0iSdOnVKixcv1oABA3Tx4sU49w8AJIabm5uqV6+un3/+WadPn36q4d+vv/5aISEh+uSTT+zK4+oZPD7W4741zP2wRz/jJvRckhTOzl0BAQE6d+6cQ/nZs2dt7XlWMmXKpCJFimjlypW6ffu2fH19H1vfGtZ/9L7F486dcV2HxHUef5r3Op6WgIAA9e7dW+vWrdP+/fudfqfxrNcvSefOnXN4n509ezbRrxGLxaIxY8aoZs2amjFjhsP0tGnTymKxaOPGjU5/mPdo2eeff65ly5apbNmymjp1qpo3b65y5colqk2JcebMGf3+++8KDg629cCbmGNG7ty5VbduXU2bNk1169bV4sWLNXTo0Gfe+5yroSc3AAAAPFVBQUHau3evXdmaNWv0zz//2JU1aNBA+/fvV65cuZz+OigpH/w++OAD3b17V0OHDnWYlpj1hYWFaePGjZozZ47mz5+vPXv2OPS08rR+XSVJe/bs0ciRIxUUFGTret76YfnRD16ffvqp02VkypRJr732mj7++GNNnz5dDRs2tOs+Pz5eXl4KDg7WmDFjJEm7du2Ks+7ly5d19+5dp9MOHz4s6fl9cAcAJF6DBg10+fJl3bt3z+k5Ml++fJL+LyRl/cWo1YIFCxQbG/vYdeTNm1e5cuVSRESE0zC8VVzn19q1a8vd3V1Hjx512kbrzeR8+fIpU6ZMmjdvnt0QECdPntTmzZsTtkOc4BqDawwAwOPVrVtXUVFRmjVrVpx1qlevLunBzbyHfffdd7p165ZtelJUrFhRqVKl0sGDB+O8VvD09JSPj4+Cg4O1cOHCx/Zo8zTPxwnxIlyPOVO9enX9888/+vHHH+3KZ8+ebZv+JLJnz64ePXqoZs2a2rlz5xMtCwAeNnDgQBlj1KlTJ6efQWJiYrRkyZJEL9disTh81tq7d6+2bNmSpHbmy5dPGTNmdBie8dSpUw6fcRN6LnlaqlevroMHDzocn2fPni2LxaKqVas+1fU96v3339fVq1fVs2dPp0NA/vPPP1q5cqWkB0F3b29vh/sWixYteipteRb3OhJzPo6JiYkzsGcd8vJZfpaOq63WIW4fvfbbvn27Dh06lKTrhBo1aqhmzZoaNmyY03tOxhidOXPG6fNQpEgRW919+/apZ8+eatOmjTZu3KiiRYuqefPmunr1arzblRR37txRx44dFRsbq/79+9vKE3vM6NWrl/bu3au2bdvKzc1NnTp1euK2vWjoyQ0AAABP1RtvvKH3339fQ4YMUXBwsA4ePKipU6cqZcqUdvWGDRumVatWqUKFCurZs6fy5cunqKgonThxQj/99JOmT5+e6F/R5ciRQ127dtXkyZMdpiV0fatWrdKoUaP0/vvv2z5kjRo1Sv369VNISIiaNGkiSbYPRJMnT1bbtm3l4eGhfPnyxfuLuN9//10pU6ZUTEyMzp49q9WrV+urr75S+vTptWTJEltPLvnz51euXLk0YMAAGWOUJk0aLVmyRKtWrYpz2b169bL90mjmzJnx7q8hQ4bo9OnTql69urJmzapr165p8uTJ8vDwUHBwcJzzrV27Vr169dLrr7+uChUqKCAgQBcvXtS8efO0fPly27Bk8dm/f7/TL+Fz5crlMHwNAODZadGihebMmaN69eqpV69eKlu2rDw8PHT69GmtXbtWjRs3VpMmTVSgQAG1bt1akyZNkoeHh2rUqKH9+/dr/PjxDkNuOTNt2jQ1bNhQL730kvr06aPs2bPr1KlTWrFihe1GbVzn16CgIA0bNkzvvfeejh07pjp16ih16tS6cOGCfvvtNyVPnlxDhw5VsmTJNHz4cHXs2FFNmjRRp06ddO3aNYWHhyd5uFKJawyuMQAA8WnZsqVmzpypLl266MiRI6pataru37+vbdu2qUCBAmrRooVq1qyp2rVr691339WNGzdUsWJF7d27V2FhYSpRooTeeOONJK/fz89PU6ZMUdu2bXXlyhU1bdpU6dOnV2RkpPbs2aPIyEhbDxkTJ05UpUqVVK5cOQ0YMEC5c+fWhQsXtHjxYn366afy9/dX4cKFJUkzZsyQv7+/vL29lSNHjniHBE2qF+F6zNm1SJs2bTRt2jS1bdtWJ06cUJEiRbRp0yaNHDlS9erVU40aNRK1H65fv66qVauqVatWyp8/v/z9/bV9+3YtX75cr7zySqKWBQCPU758eX3yySfq1q2bSpUqpa5du6pQoUKKiYnRrl27NGPGDBUuXFgNGzZM1HIbNGig4cOHKywsTMHBwTpy5IiGDRumHDlyxBtGdiZZsmQaOnSoOnfurKZNm6p9+/a6du2ahg4dqkyZMtn1nprQc8nT0qdPH82ePVv169fXsGHDFBgYqGXLlunjjz9W165dlTdv3qe2Lmdee+01vf/++xo+fLgOHz6sDh06KFeuXLp9+7a2bdumTz/9VM2bN1etWrVksVjUunVrRUREKFeuXCpWrJh+++03zZ0796m05Vnc67Cej8eMGaO6devKzc1NRYsWddoT/PXr1xUUFKTXXntNNWrUULZs2fTPP/9o3bp1mjx5sgoUKJCg8+ipU6e0detWh/J06dIpV65c8bb10WuHfPny6c0339SUKVOULFky1a1bVydOnND777+vbNmyqU+fPgndHXbGjBmjUqVK6eLFi7YhYaUHP3p488031a5dO+3YsUNVqlRR8uTJde7cOW3atElFihRR165ddevWLTVr1kw5cuTQxx9/LE9PTy1YsEAlS5ZUu3btbOF9f39/BQYGatGiRapevbrSpEmjtGnTKigo6LHts+7H+/fv6/r169q1a5ciIiJ08uRJTZgwQbVq1bLVTewxo2bNmipYsKDWrl2r1q1bK3369Enahy80AwAAADxG27ZtTfLkye3KgoODTaFChZzWj46ONv379zfZsmUzPj4+Jjg42OzevdsEBgaatm3b2tWNjIw0PXv2NDly5DAeHh4mTZo0plSpUua9994z//zzz2PbFVcbIiMjTYoUKYwkM27cuESt7+zZsyZ9+vSmWrVq5t69e7b57t+/bxo2bGhSpUpljh8/bisfOHCgyZw5s0mWLJmRZNauXRtne8PCwowk28PLy8tkypTJ1KpVy0yePNncuHHDYZ6DBw+amjVrGn9/f5M6dWrz2muvmVOnThlJJiwszOl6goKCTIECBR6776yWLl1q6tata7JkyWI8PT1N+vTpTb169czGjRsfO9/ff/9tBg8ebCpWrGgyZsxo3N3djb+/vylXrpyZMmWKiY2NtdVdu3atkWQWLlxoK5s5c6bdvnj08dlnnyWo/QCAB6zH1e3btz+2nrNzulVMTIwZP368KVasmPH29jZ+fn4mf/78pnPnzubPP/+01YuOjjZ9+/Y16dOnN97e3uall14yW7ZscTjPW4//j54bt2zZYurWrWtSpkxpvLy8TK5cuUyfPn3s6jzu/Prjjz+aqlWrmhQpUhgvLy8TGBhomjZtan755Re7ZXz++ecmT548xtPT0+TNm9dERESYtm3bmsDAwMfuI2O4xuAaAwD+dzm7pojr+sF6/nnYnTt3zJAhQ2zn4ICAAFOtWjWzefNmuzrvvvuuCQwMNB4eHiZTpkyma9eu5urVq3bLCgwMNPXr13dYr7Pj/8PWr19v6tevb9KkSWM8PDxMlixZTP369R3qHzx40Lz22msmICDAeHp6muzZs5vQ0FATFRVlqzNp0iSTI0cO4+bmZiSZmTNnJmrfOfOiX48FBweb4OBgu7qXL182Xbp0MZkyZTLu7u4mMDDQDBw40G5fGmOMJNO9e3eH7X643VFRUaZLly6maNGiJkWKFMbHx8fky5fPhIWFmVu3bj1mzwJA0uzevdu0bdvWZM+e3Xh6eprkyZObEiVKmCFDhpiLFy/a6sV1Xnr0uBgdHW369etnsmTJYry9vU3JkiXNjz/+6PB59Pjx404/WxpjnH4umzFjhsmdO7fdZ9zGjRubEiVK2NVL6LnEmbjOZY+7F3Dy5EnTqlUrExAQYDw8PEy+fPnMuHHj7D73xrWtcZ3TE3pOtVq/fr1p2rSpyZQpk/Hw8DApUqQw5cuXN+PGjbP7PHz9+nXTsWNHkyFDBpM8eXLTsGFDc+LECYf9bb3GiYyMdFhXXK8DYxJ2ryMxz3t0dLTp2LGjSZcunbFYLEaS3fcGD4uOjjbjx483devWNdmzZzdeXl7G29vbFChQwPTv399cvnzZYV0Pn5Ot7Yrr8frrrztd78Piuna4d++eGTNmjMmbN6/x8PAwadOmNa1btzZ///13vMt83GuhVatWRpLT12ZERIQpV66cSZ48ufHx8TG5cuUybdq0MTt27DDGGNO6dWvj6+trDhw4YDffwoULjSTz4Ycf2sp++eUXU6JECePl5WUkOdzjetij+9HNzc2kTp3alCpVyvTu3dthfcYk/JjxsPDwcCPJbN26Nc62/JdZjHHSdyMAAACAF87evXtVrFgxTZs2Td26dXvezQEAAP8RXGMAAAAAgOu4du2a8ubNq5dfflkzZsx43s0B8C8qXbq0LBaLtm/f/ryb8lwwXCkAAADwgjt69KhOnjypQYMGKVOmTAoNDX3eTQIAAP8BXGMAAAAAwPN1/vx5jRgxQlWrVlVAQIBOnjypDz/8UDdv3lSvXr2ed/MA/Atu3Lih/fv3a+nSpfr999/1ww8/PO8mPTeE3AAAAIAX3PDhw/XVV1+pQIECWrhwoXx9fZ93kwAAwH8A1xgAAAAA8Hx5eXnpxIkT6tatm65cuSJfX1+99NJLmj59ugoVKvS8mwfgX7Bz505b0DUsLEwvv/zy827Sc8NwpQAAAAAAAAAAAAAAAAAAl5XseTcAAAAAAAAAAAAAAAAAAIC4EHJ7QcyaNUsWi0UWi0Xr1q1zmG6MUe7cuWWxWBQSEvJM2nD27FmFh4dr9+7dCaq/bt06W5stFovc3NyULl06NWzYUDt27HgmbXxWDh48qPDwcJ04cSLB82zbtk1NmjRR9uzZ5eXlpQwZMqh8+fLq27evXb2QkBAVLlzYriwoKMhu3z38iO/5PXHihCwWi8aPH5/gtj5rc+fO1aRJk553MwAAAAAAAAAAAAAAAPACcn/eDUDi+Pv764svvnAIOq1fv15Hjx6Vv7//M1v32bNnNXToUAUFBal48eIJnm/kyJGqWrWqYmJitGvXLg0dOlTBwcHavXu38uTJ88za+zQdPHhQQ4cOVUhIiIKCguKtv2zZMjVq1EghISEaO3asMmXKpHPnzmnHjh365ptvNGHChHiXUbFiRadBtRQpUiRlE56ruXPnav/+/erdu/fzbgoAAAAAAAAAAAAAAABeMITcXjDNmzfXnDlzNG3aNLuw0xdffKHy5cvrxo0bz7F1zuXJk0cvvfSSJKly5cpKlSqV2rZtq6+//lpDhw59zq17NsaOHascOXJoxYoVcnf/v7dZixYtNHbs2AQtI1WqVLb9BgAAAAAAAAAAAAAAAPyvIuT2gmnZsqXmzJmjefPmqXPnzpKk69ev67vvvtNHH33kdEjIK1euaPDgwVq0aJEiIyOVNWtWtWzZUkOGDJGXl5et3sKFCzV+/HgdPnxYMTExypgxo0JCQhQREaF169apatWqkqR27dqpXbt2kqSwsDCFh4cnahtKly4tSbpw4YJd+Z9//qmwsDD98ssvun79unLmzKkePXqoe/fudvWuXbum4cOH64cfftCZM2eUMmVKlS5dWhMnTlT+/PklSXfv3tXYsWP19ddf6/jx40qRIoUaNGigsWPHKl26dLZlBQUFqXDhwurRo4cGDRqkQ4cOKTAwUP3791f79u0lPRgq1rq91n0gSTNnzlRoaKjTbbx8+bLSpk1rF3CzSpbs3x8l2LoNa9as0YIFC7Rw4UIZYxQcHKypU6cqc+bMtrrWfdKhQweFhYXpjz/+UObMmdW7d2/17NnTYZnHjx+3693O+lpZu3atQkJCFBISovXr10uSLBaLrZ4x5tlvOICnyhJ/lWcvKUMx9+uXsGXGVw8A8MxwZfji4HoAAPCscD3wYkhucYmrAQDAf9At7hu9MLgeAAA8K/FdD/z7aRs8kRQpUqhp06aKiIiwlc2bN0/JkiVT8+bNHepHRUWpatWqmj17tt5++20tW7ZMrVu31tixY/XKK6/Y6m3ZskXNmzdXzpw59c0332jZsmUaMmSIYmNjJUklS5bUzJkzJUmDBw/Wli1btGXLFnXs2DHR23D8+HFJUt68eW1lBw8eVJkyZbR//35NmDBBS5cuVf369dWzZ0+73t5u3rypSpUq6dNPP1W7du20ZMkSTZ8+XXnz5tW5c+ckSffv31fjxo01evRotWrVSsuWLdPo0aO1atUqhYSE6M6dO3bt2bNnj/r27as+ffpo0aJFKlq0qDp06KANGzZIkurXr6+RI0dKkqZNm2bb9vr168e5jeXLl9e2bdvUs2dPbdu2TTExMYneT8YYxcbGOjyeJBzWsWNHeXh4aO7cuRo7dqzWrVun1q1bO9TbvXu3evfurT59+uiHH35QhQoV1KtXL6fDp8bn448/VsWKFZUxY0bbvtuyZUuStwHAC2b8+P97AACA/01cDwAAAAAAAAAAnhA9ub2A2rdvr6pVq+rAgQMqVKiQIiIi9Nprr8nf39+h7pdffqm9e/dqwYIFeu211yRJNWvWlJ+fn959912tWrVKNWvW1ObNm2WM0fTp05UyZUrb/NaeylKkSKHChQtLknLlypWoYTTv37+v2NhYxcTEaNeuXerbt68KFixo6ylNkt5++235+/tr06ZNtmFYa9asqejoaI0ePVo9e/ZU6tSpNWnSJB04cECrVq1SjRo1bPM/HNhbsGCBli9fru+++86uvFixYipTpoxmzZqlrl272sovXbqkX3/9VdmzZ5ckValSRatXr9bcuXNVpUoVpUuXTnny5JEkFSxYMEHbPnr0aB0+fFhTpkzRlClT5OHhoTJlyqhhw4bq0aOH/Pz84l3GTz/9JA8PD4fy4cOHa/DgwfHO70ydOnX00Ucf2f6+cuWK+vfvr/Pnzytjxoy28rNnz2rXrl0qVqyYJKlu3bq6ePGihg8frm7dusnX1zfB6yxYsKBSpUolLy8vhl8F/teNH/98ekdJyHrptQUAgH8H1wMAAAAAAAAAgCSgJ7cXUHBwsHLlyqWIiAjt27dP27dvtwuMPWzNmjVKnjy5mjZtalduDa+tXr1aklSmTBlJUrNmzbRgwQKdOXPmqbW3efPm8vDwkK+vrypWrKgbN25o2bJlSpUqlaQHvc2tXr1aTZo0ka+vr12vZfXq1VNUVJS2bt0qSfr555+VN29eu4Dbo5YuXapUqVKpYcOGdssqXry4MmbMqHXr1tnVL168uC3gJkne3t7KmzevTp48meRtDggI0MaNG7V9+3aNHj1ajRs31h9//KGBAweqSJEiunTpUrzLqFSpkrZv3+7w6NChQ5Lb1ahRI7u/ixYtKkkO21qoUCFbwM2qVatWunHjhnbu3Jnk9QP4H/Gi9dbCDW0AAJ4+rgcAAAAAAAAAAE8RPbm9gCwWi9q1a6ePPvpIUVFRyps3rypXruy07uXLl5UxY0ZZHhkbPX369HJ3d9fly5clPei97Mcff9RHH32kNm3aKDo6WoUKFdJ7772nli1bPlF7x4wZo2rVqun27dtauXKlRo0apZdfflnbtm2Tl5eXLl++rNjYWFuvZ85YQ2GRkZF2gTRnLly4oGvXrsnT0/Oxy7IKCAhwqOPl5eUwrGlSlC5dWqVLl5YkxcTE6N1339WHH36osWPHauzYsY+dN2XKlLZ5n5ZHt9XLy0uSHLb14V7dHi2zvmYAQJL9jet+/V6cG9kAAODp4XoAAAAAAAAAAPCMEXJ7QYWGhmrIkCGaPn26RowYEWe9gIAAbdu2TcYYu6DbxYsXFRsbq7Rp09rKGjdurMaNGys6Olpbt27VqFGj1KpVKwUFBal8+fJJbmvOnDltYa0qVarIx8dHgwcP1pQpU9SvXz+lTp1abm5ueuONN9S9e3eny8iRI4ckKV26dDp9+vRj15c2bVoFBARo+fLlTqc7G9b13+Dh4aGwsDB9+OGH2r9//3NpQ0KdP38+zjJrUM7b21uSFB0dbVcvIb3UAfiPePQGdnw3tJ9GDyncOAcAwLVwPQAAAAAAAAAA+BcwXOkLKkuWLHrnnXfUsGFDtW3bNs561atX1z///KMff/zRrnz27Nm26Y/y8vJScHCwxowZI0natWuXrVxy7PUrsfr376/cuXNr9OjRunnzpnx9fVW1alXt2rVLRYsWtfV+9vDDGqyqW7eu/vjjD61ZsybO5Tdo0ECXL1/WvXv3nC4rX758iW5zYrf93LlzTssPHTokScqcOXOi2/BvOnDggPbs2WNXNnfuXPn7+6tkyZKSpKCgIEnS3r177eotXrzYYXlPq2c8AC4ksTeWGQIMAID/Hq4HAAAAAAAAAAD/Enpye4GNHj063jpt2rTRtGnT1LZtW504cUJFihTRpk2bNHLkSNWrV081atSQJA0ZMkSnT59W9erVlTVrVl27dk2TJ0+Wh4eHgoODJUm5cuWSj4+P5syZowIFCsjPz0+ZM2dOdGDLw8NDI0eOVLNmzTR58mQNHjxYkydPVqVKlVS5cmV17dpVQUFBunnzpv766y8tWbLEFmrr3bu35s+fr8aNG2vAgAEqW7as7ty5o/Xr16tBgwaqWrWqWrRooTlz5qhevXrq1auXypYtKw8PD50+fVpr165V48aN1aRJk0S1uXDhwpKkGTNmyN/fX97e3sqRI4fToU4lqXbt2sqaNasaNmyo/Pnz6/79+9q9e7cmTJggPz8/9erVK951Xrt2TVu3bnUo9/LyUokSJRLV/sTKnDmzGjVqpPDwcGXKlElff/21Vq1apTFjxsjX11eSVKZMGeXLl0/9+vVTbGysUqdOrR9++EGbNm1yWF6RIkX0/fff65NPPlGpUqWULFmypz4UK4B/QVJ7TOGGNgAA/x1cDwAAAAAAAAAAngNCbv9x3t7eWrt2rd577z2NGzdOkZGRypIli/r166ewsDBbvXLlymnHjh169913FRkZqVSpUql06dJas2aNChUqJEny9fVVRESEhg4dqlq1aikmJkZhYWEKDw9PdLtee+01lStXThMnTtRbb72lggULaufOnRo+fLgGDx6sixcvKlWqVMqTJ4/q1atnm8/f31+bNm1SeHi4ZsyYoaFDhyp16tQqU6aM3nzzTUmSm5ubFi9erMmTJ+urr77SqFGj5O7urqxZsyo4OFhFihRJdHtz5MihSZMmafLkyQoJCdG9e/c0c+ZMhYaGOq0/ePBgLVq0SB9++KHOnTun6OhoZcqUSTVq1NDAgQNVoECBeNf566+/Oh0mNkuWLPEO2fqkihcvrnbt2iksLEx//vmnMmfOrIkTJ6pPnz62Om5ublqyZIl69OihLl26yMvLSy1atNDUqVNVv359u+X16tVLBw4c0KBBg3T9+nUZY2SMeabbAAA23FQHAABcDwAAAAAAAADAC81iSJoAeEhQUJAKFy6spUuXPu+mAHAxlqT03PIsbigntB3czAaAFwYfSl8cXA8AAJ4VrgdeDMktlufdBADAf9Qtblm/MLgeAAA8K/FdD9CTGwAAeLE8erP60Zvc8U2Pqx4AAHhxcD0AAAAAAAAAAP9TCLkBAPAYoaGhunbtmn788cfn3RTXY70pnJQeXZ5FOxJr/HhubAMA8KS4HgAAAAAAAAAA/AsIuQGwc+LEiefdBCBOBM5cyMM3g13l5nZicUMbAIAnw/UAAAAAAAAAAOBfQsgNAAA8Ha56k7hfvxfvhjsAAC8qrgcAAAAAAAAAAM8AITcAwH/CxIkTNXPmTB07dkxp0qRRw4YNNXbsWPn5+UmSZs2apd69e2v+/Pnq3bu3/v77b1WqVEkzZ85UpkyZJEn37t3TO++8o4iICLm5ualDhw4yxjzPzXJNrnrz+nFexDYDAODKXsRz64vYZgAAAAAAAACAJCnZ824AAODFExUVpRs3bjyVx/Xr1x3KoqOjE92mZMmS6aOPPtL+/fv15Zdfas2aNerfv79dndu3b2v8+PH66quvtGHDBp06dUr9HrrZOWHCBEVEROiLL77Qpk2bdOXKFf3www9PvL/+M7gxDAAAuB4AAAAAAAAAADwH9OQGAEiUqKgo+aRPL928+VSW5+fnp3/++ceuLCwsTOHh4YlaTu/evW3/z5Ejh4YPH66uXbvq448/tpXHxMRo+vTpypUrlySpR48eGjZsmG36pEmTNHDgQL366quSpOnTp2vFihWJ3CIAAAAAAAAAAAAAAPA0JTjkZnmWrXic8eMTV/9xvyp/eFn8+hzA/7ikDsJ59+7dBwG3996TvL2frBFRUfpnxAj9/fffSpEiha3Yy8sr0Ytau3atRo4cqYMHD+rGjRuKjY1VVFSUbt26peTJk0uSfH19bQE3ScqUKZMuXrwoSbp+/brOnTun8uXL26a7u7urdOnSDFn6MM6fAACA6wEAAAAAAAAAwL/s2fXk9mg47d/6Enz8+PjXxRfyAPDkvL2fPOT2/6VIkcIu5JZYJ0+eVL169dSlSxcNHz5cadKk0aZNm9ShQwfFxMTY6nl4eNjNZ7FYCLABAAAAAAAAAAAAAODikj2TpSa297V/EwE3APjP2bFjh2JjYzVhwgS99NJLyps3r86ePZuoZaRMmVKZMmXS1q1bbWWxsbH6/fffn3ZzAQAAAAAAAAAAAABAIiS9J7dHh/505WCbFQE3AHjhXb9+Xbt377YrS5cunWJjYzVlyhQ1bNhQv/76q6ZPn57oZffq1UujR49Wnjx5VKBAAU2cOFHXrl17Og0HAAAAAAAAAAAAAABJkrSe3B4NtMUXcHuScBnBNADAQ9atW6cSJUrYPSIiIjRx4kSNGTNGhQsX1pw5czRq1KhEL7tv375q06aNQkNDVb58efn7+6tJkybPYCuQIC9CgB4AADxbXA8AAAAAAAAAACRZjDEmQRXHj/+/wFliv2R+0qDav70+APgfkKCDvxM3btxQypQppeHDJW/vJ2tEVJT+X3v3HytZWd8P/HPhwi7d9V4EKqABt7JgtYnUDf7YpYKLIq3ECGgrta1BWta0Ggtx0KpFtGBb7RhMaWz7l5hqrUGpMUpDhbLU+iOpVNFYf+DK1h+EQkTvlQ3fYt37/YPOZe7s/Dgzc34855zXK9mwe+/cmWfOOfc8b57nM88TV10VKysrsbS0NN9zUbiFMl9ssO/XtwM02qy5hPLJAwAURR6ohy0LpaYBAFrkQLYpaxIgDwBQlEl5IPt2pbNsSWoAGgDIQ3+xfZbH9pNHAKAZ5AEAAAAAgNbKXuRmixAAYJReTshrAnmW4vpRTHIDQDnkAQAAAAAACpK9yG1aZQ8YG6AGgHKMmyQeXGFlnsnuPCe2AYB8yQMAAAAAAJSouCK3Mg0bPB/3GABgNlkmmYc9ZtbJbRPbAJAeeQAAAAAAgJItrK2trWV64LBPaY8bZK6qqCzFNgEkKNPNf4jV1dVYXl6OuOaaiM2b52vE//t/EVddFSsrK7G0tDTfc1G4Q7LArMroj0flFlkAIEmz5hLKJw8AUBR5oB62LCxU3QQAGupAtilrEiAPAFCUSXngsJmetTcg3OnUZ3C4Lu0EgDYqenUWE9oAkD55AAAAAACAEfLZrjSlQeJhK8yl1D4AYKNevz3LxHPWyXBZAADSJg8AAAAAADDG9EVudRgUrkMbAaCtxk1E976Xd19u5RYASIs8AAAAAADAFLJvV2ogGABIzTT5pOgt0ACAasgDAAAAAACNN91KbgrdAICqzTM5XdTKMABAueQBAAAAAIBWyb6SGwBA1fJafcUqLgBQX/IAAAAAAEDr1KPIrds1+AwAbaLfBwDkAQAAAAAA/s9025VWxRYiANA8/f374CR2GX1/tytjAEDV5AEAAAAAADJIbyU3n9QGgHbo7/NNLgNAO8kDAAAAAABkkFaRW29w2/akANAOw/p8E9wA0C7yAAAAAAAAE9Rju9JJRhXEGRQHgPz1+tc8C9KzbhXW6Yzv96dpk5wAALOTBwAAAAAAKNHsK7lNWnVtloHuvAeXrQYHAGnodIqdRDZBDQDpkwcAAAAAAJjRdEVuvYK2UUVt/YVvg1/Lqn9QWpEaAKRrmonqLH36pMfkuZ25SXAAyIc8AAAAAABACbIXuWVdrS2PVd16A80GnAEgXdNOMmfdgmza15EXAKA68gAAAAAAACVYrLoBI80zQG1wGwCK0elMN5E9+PjexHYeK7Do7wGgGvIAAAAAAAAlm2670jyN2/p02ufJ8jUAYD79k8hZJpSnnXSeZ5I6z63LAIDR5AEAAAAAACpQ3kpu4z6pPc12JQasAaBaeayY0p8JZnm+/p+RDQCgfPIAAAAAAAAlKne70nGDztMUumV9LduWAEC+8phA7vXRWbcpm9Sfz7JlGgAwO3kAAAAAAICSVbdd6TRm3drUViUAkJ9Z+9Rxk8gmmAGgXuQBAAAAAAAqkH6RW16fEAcA0jGpb87Sd89aBA8ApEEeAAAAAAAgo/SL3ACAdsk6oT0LK8UAQD3IAwAAAAAA9KmmyK3TKX9Q2ae6ASA/JocBAHkAAAAAAICSVFPk1ttKxIA4ANRP1v67v8B88GdGPUeRRelyBwDkRx4AAAAAAKBE0xe55TkgPG6wGwBIyywrsRbR18868W1VVwCYnzwAAAAAAEAFFgt/hU5n/CDypAHuST8/TTsAgPlN2y/3r946rj/O8ryDj+k9nwlrACiXPAAAAAAAQIlmK3KbpvBs0uPGbVs6qTguaxsUuAFAseadXJ7m5/r7dZPZAJAOeQAAAAAAgIJMv13pLGYpMsuy+pviNQAoX/82ZVm3LMtr8nmeCW25AQDyIw8AAAAAAFCi6YrcequnzbMtSdbHj2tDlscBAMUa7Ntn3Z5cfw4A9SUPAAAAAABQguzblabwaWcFbgCQrmm2Es/yXIPmee4UcgwAtIE8AAAAAABAAabfrjTLNiSjBqIHv551S5PB55xmUNsgNgCUZ1i/W/WqLbIAAJRLHgAAAAAAIGfZV3KbZtC5V9A260C1wWcAqK/+DKBPB4B2kgcAAAAAAMjR9Cu5zaNX/DbNCm7DGCAHgLRl7eunXelllteyxTkAVEMeAAAAAAAgJ8UUuc0ysNztjh907n1/0uMAgHqZZQK89/esE+CyAwCkTR4AAAAAAGCM7NuV1pEV3wCgPrL228MmtE1aA0AzyAMAAAAAAAxRzEpuvYHlrFuOFDEQrcANAJqnrBwBAKRLHgAAAAAAaJ3sRW69rUOybiEyrNCtzMIzA9wAQD/ZAACQBwAAAAAAamn6ldy63eyDwuNWdBt8TNbXs0IbANAvawG+DAEAzSUPAAAAAAA0WjHblQ6Tx3YinY5PXQMAw42btDahDQDtIA8AAAAAADTS9EVuw7YfnWeltiyv1/szzXMZvAaA5pm0ouxgTpEHAKB55AEAAAAAgNZZnOmnhhW6ZXlsv2ED0gaeAYBxZAUAQB4AAAAAAGid2YrcymJrUgBgFtMU5AMAzSQPAAAAAAA0xvTblY4ybMB4mlXc8mTwGgDap5cv5AAAaC95AAAAAACgkfIrcouYbxDZADQAMA9ZAgCQBwAAAAAAGinfIreIRweUe3/GPWbe1xjHNqcAAAAAAAAAAACNkH+RW1aTCuF6jxn8u09lAwAAAAAAAAAAtMZi1Q3IVLQ2WOBmpTYAAAAAAAAAAIBWqG4lt6xmXblNIRwAAAAAAAAAAEDtpV/k1jNt0ZptTQEAAAAAAAAAAGqvHkVu3e7GorVOZ3wRmwI3AAAAAAAAAACARki/yG3cCm6K2QAAAAAAAAAAABptseoGTJS1kK33uGm3NQUAAAAAAAAAACBZ6a/kNi2ruwFAMRSSAwDyAAAAAAAAFah3kVtvcF1hGwAAAAAAAAAAQCOlv13pOIrbAAAAAAAAAAAAGq3eK7kBAOXrdm1VBgBtJw8AAAAAAFAiRW4AQHb9k9kmtwGgneQBAAAAAABKVu/tSgGA6g1ObHc6j33N1uIA0A7yAAAAAAAABbKSGwCQL6u7AADyAAAAAAAAOVLkBgAUz8Q2ACAPAAAAAAAwI0VuAAAAAAAAAAAAJEuRGwAAAAAAAAAAAMnKXuRmWxEAYB6yBAAgDwAAAAAAMAMruQEAAAAAAAAAAJAsRW4AQHms3gIAyAMAAAAAAExpcapH9w9Edzo5NwUAAAAAAAAAAAA2mn0lt27Xp68BgOnJDwCAPAAAAAAAwBSmW8ltmHED01Z7AwCG6XblBABoO3kAAAAAAICMZl/JLQufzAYARpETAAB5AAAAAACADIotcoswYA0AAAAAAAAAAMDMii9yAwAYRTE8ACAPAAAAAAAwQTlFbgasAYBR5AQAQB4AAAAAAGAMK7kBAAAAAAAAAACQLEVuAED1rN4CAMgDAAAAAACMUF6Rm8FqAAAAAAAAAAAAprRY6qt1uxGdTqkvCQAkTjYAAOQBAAAAAADGKH+7Uiu6AQA9JrQBAHkAAAAAAIAJyl3JDQAgwmQ2ACAPAAAAAACQWTVFbrYtBYD20fcDAPIAAAAAAAAzqG4lN4VuANBc/X28Ph8A2kkeAAAAAAAgJ4dV3QAAoOFMaAMA8gAAAAAAAHOotsit26305QEAAAAAAAAAAEibldwAgHxZqQUAkAcAAAAAAMiRIjcAAAAAAAAAAACStVjJq/pENwA0kz4eAJAHAAAAAADIWfkruRnsBgAAAAAAAAAAIKNyVnJT2AYAAAAAAAAAAMAMilnJrb+oTYEbAAAAAAAAAAAAM8p/JbdeUZviNgBoF30/ACAPAAAAAABQgGJWcgMAAAAAAAAAAIAcKHIDAOZn1RYAQB4AAAAAAKAg+Ra5GdAGgPbR/wMA8gAAAAAAAAXKr8jNgDYAAAAAAAAAAAA5s10pAAAAAAAAAAAAyVLkBgDMzkquAIA8AAAAAABAwRarbgAAUDMmsgEAeQAAAAAAgBLls5KbwW0AaAd9PgAgDwAAAAAAULL5i9wMbgNAe3S7VbcAAKiaPAAAAAAAQMnyWckNAAAAAAAAAAAACjBfkZtV3ACgfazeAgDIAwAAAAAAlGix6gYAADWiwB0AkAcAAAAAACjZ9Cu5GcwGgHaSAQAAeQAAAAAAgArMtpKbQW0AAAAAAAAAAABKMP1KbgAAAAAAAAAAAFASRW4AAAAAAAAAAAAka/oit263gGYAAAAAAAAAAADAoRanenSnU1AzAAAAAAAAAAAA4FDZV3JT4AYAAAAAAAAAAEDJpt+uFAAAAAAAAAAAAEqiyA0AAAAAAAAAAIBkKXIDAAAAAAAAAAAgWYrcAAAAAAAAAAAASJYiNwAAAAAAAAAAAJKlyA0AAAAAAAAAAIBkLaytra1lemDRLQGgVJlu/kOsrq7G8vJyxMpKxNLSfI1YXY1YXo6VlZVYmve5KJwsAEBRZs0llE8eAKAo8kA9bFmQBgAoxoFsU9YkQB4AoCiT8oCV3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAAAAAAAAAAACSpcgNAAAAAAAAAACAZClyAwAAAAAAAAAAIFmK3AAAAAAAAAAAAEiWIjcAauP++++P17zmNXHyySfHpk2b4oQTTojzzjsvPv/5zxf6ujfccEMcffTRhb4GAAAAAAAAADDcYtUNAICsXvayl8VPf/rT+MAHPhBPecpT4r//+7/jtttuiwcffLDqpgEAAAAAAAAABbGSGwC18OMf/zj+7d/+Ld71rnfF7t2748lPfnI8+9nPjje/+c1x/vnnrz9uZWUl9uzZE094whNiaWkpzjnnnLjrrrvWv79v37546UtfGscff3xs3bo1nvWsZ8Wtt946dXs+8YlPxBlnnBGbN2+O4447Li666KL17z3yyCPxxje+MZ70pCfFli1b4jnPeU7s3bt3rvcPAAAAAAAAAG2lyA2AWti6dWts3bo1Pv7xj8f//M//DH3M2tpanH/++XHffffFzTffHHfeeWfs2LEjXvCCF6yv9vbQQw/Fi1/84rj11lvjS1/6Upx33nnxkpe8JL773e9mbsunPvWpuOiii+L888+PL33pS3HbbbfFGWecsf79V7/61fHZz342/uEf/iG+8pWvxK//+q/Hr/7qr8bdd98930EAAAAAAAAAgBZaWFtbW8v0wKJbAkCpMt38h1hdXY3l5eWIlZWIpaX5GrG6GrG8HN/73vdiqe+5Nm3aFJs2bTrk4R/72Mfisssui4cffjh27NgRZ599dlx88cXxjGc8IyIi/uVf/iUuvPDCuP/++zf8/Pbt2+ONb3xj7NmzZ2gzfumXfil+//d/P173utcN/f4NN9wQl19+efz4xz+OiIhdu3bFU57ylPjgBz94yGP37dsXp556anz/+9+PJz7xietff+ELXxjPfvaz40//9E8nH5dEyQIAFGXWXEL55AEAiiIP1MOWBWkAgGIcyDZlTQLkAQCKMikPWMkNgMqddNJJsby8vP7nz/7sz4Y+7mUve1nce++98YlPfCLOO++82Lt3b+zYsSNuuOGGiIi4884746GHHopjjz12feW3rVu3xj333BP79u2LiIgDBw7EG9/4xnj6058eRx99dGzdujW+8Y1vTLWS25e//OV4wQteMPR7//Ef/xFra2tx2mmnbWjDHXfcsd4GAAAAAAAAACC7xaobAADDVnIbZfPmzXHuuefGueeeG29729vi937v9+Lqq6+OSy65JA4ePBgnnnhi7N2795CfO/rooyMi4sorr4xbbrklut1ubN++PY466qh4+ctfHo888kjm9h511FEjv3fw4ME4/PDD484774zDDz98w/e2bt2a+TUAAAAAAAAAgEcpcgOgcktLSxuK3Kbx9Kc/PT7+8Y9HRMSOHTvivvvui8XFxdi2bdvQx3/mM5+JSy65JC688MKIiHjooYdi//79U73mM57xjLjtttvi1a9+9SHfe+Yznxk/+9nP4v7774/nPe95Uz0vAAAAAAAAAHAo25UCUAs//OEP45xzzokPfvCD8ZWvfCXuueeeuPHGG+Pd7353vPSlL42IiBe+8IWxc+fOuOCCC+KWW26J/fv3x+c+97n44z/+4/jiF78YERHbt2+Pm266Kb785S/HXXfdFa985Svj4MGDU7Xl6quvjg9/+MNx9dVXx9e//vX46le/Gu9+97sjIuK0006L3/qt34pXvepVcdNNN8U999wT//7v/x7vete74uabb873oAAAAAAAAABAC1jJDYBa2Lp1azznOc+J6667Lvbt2xc//elP46STTorLLrss3vKWt0RExMLCQtx8883x1re+NS699NJ44IEH4oQTToizzjorjj/++IiIuO666+LSSy+NXbt2xXHHHRdvetObYnV1daq2PP/5z48bb7wxrrnmmvjzP//zWFpairPOOmv9++9///vj2muvjTe84Q3xgx/8II499tjYuXNnvPjFL87vgAAAAAAAAABASyysra2tZXpg0S0BoFSZbv5DrK6uxvLycsTKSsSMW4z2PVnE8nKsrKzMvF0p5ZEFACjKrLmE8skDABRFHqiHLQvSAADFOJBtypoEyAMAFGVSHrBdKQAAAAAAAAAAAMlS5AYAAAAAAAAAAECyFLkBAAAAAAAAAACQLEVuAAAAAAAAAAAAJEuRGwAAAAAAAAAAAMlS5AYAAAAAAAAAAECyFLkBAAAAAAAAAACQLEVuAAAAAAAAAAAAJGu6Irdud/jfAQAAAAAAAAAAoACLUz2609n478FCt8HvAwAAAAAAAAAAwByyr+Q2aRU3BW4A0D5WdgUA5AEAAAAAAAo23XaloyhwA4B263ZNcANA28kDAAAAAAAUZLrtSgEAIoav8KroHQDaRR4AAAAAAKAk86/kZgAbANrByiwAgDwAAAAAAEAFplvJbdhgdv/XFLwBAAAAAAAAAACQo3y3K+12FboBQNvo+wEAeQAAAAAAgALlV+TW6di2BADayKquAIA8AAAAAABAgQ6b6aeGDVj3BrT7B7YVvQFAs0zq2/X9ANB88gAAAAAAACWbrcht1IB1p7OxAM6ntwGgecZNXFvZFQDaQR4AAAAAAKBEsxW5jdLtGsgGgCbrFbArZAeA9pIHAAAAAAAoWb5Fbj0K3QCg+UZtX27CGwDaQx4AAAAAAKAExRS5AQDN1CtkH7d1OQDQbPIAADDB2RFxeNWNAAAAoFEW1tbW1jI9cJbV2QxsAyQr081/iNXV1VheXo5YWYlYWpqvEaurEcvLsbKyEkvzPheFmzoLyAEAZDRrLqF88gAARZEH6mHLwkKmxz01Ir4dET8rtDUANMmBbFPWJCBrHgCAaU3KA9Ov5JZ1gNpANgAAAAAAtM43Q4EbAAAA+Spmu1IFbgDQbp3O6C3MAIB2kAcAAAAAAMhJftuVKmwDqBXblTKthYhHJ6qzTljLBgBkZEOS+pAHACiKPFAPticDoCi2K60PeQCAouS/Xemg3oB1t+sT2gDQdNNMVMsFANBM8gAAAAAAACVbnPknhw1q+4Q2ADRXb5I6a38vFwBA88gDAAAAAABUYLaV3AYHqTud8gaurRgHANXqbVE2qP9rJrQBoNnkAQAAAAAASjTbSm7jiswMYgNA8/VngU5n46ouitEBoB3kAQAAAAAASjLbSm7jlLHSmkI6AKje4EotvQygnwaA9pAHAAAAAAAowWwruWUxz6D2sJ/tL5wbVkRnAB0Ayqf/BQDkAQAAAAAACrawtra2lumBddlqxOA6QCaZbv5DrK6uxvLycsTKSsTS0nyNWF2NWF6OlZWVWJr3uSjcwuCWZCno3xYNgNqaNZdQPnkAgKLIA/WwZWGh6iYA0FAHsk1ZkwB5AICiTMoDxa3kVpW8BretFgcAww3rD6ua8O5tiwYAlEseAAAAAACgRM1byW2YaQbXs7xPxW5AA1jJjWnNlAWK7jPn2R4dgGT4rHZ9yAMAFEUeqAcrtwBQFCu51Yc8AEBRJuWBdhS5FcEAOlBzityY1sxZYJ4+c9SktaJ0gEYxjF0f8gAARZEH6sGkNgBFUeRWH/IAAEVp33alZfFJcQDIZtatywYnrqedVM9rC3MAYH7yAAAAAAAAc7CS27wMlAM1ZSU3plXrLKC/Bkiaz2rXhzwAQFHkgXqwcgsARbGSW33IAwAUxUpuRfOpcABI3+CE/Cz99qRJfVkAANImDwAAA7ZExDER8b2qGwIAAMBEitwAgPYZVqQ+78o0tjIHgHqRBwCg9XZGxBURcX7VDQEAAGAi25XmadxAthXfgMTYrpRpyQJT0N8DTMWGJPUhD0xBHgCYijxQD03bnmxLRPx8ROyvuB0A2K60TpqWBwBIh+1Ky5RlsD+P7VGayrFhVll+91xPQJms4gIAyAMAkLwD//cHAACA9Clyq1odVnjLo43DBvcnFSaZEGCSaVaR6H+s6wooWqejHwOAtpMHAAAAAAByY7vS1Awb/B5WnFPWylWTXifrawwrlLP6FvOY957k2rJdKVOTBWbkfgMwkQ1J6kMemJE8ADCRPFAPticDoCi2K60PeQCAokzKA4eV1A6y6nYP/TPs+1mfq2hZVmOb5eemfRztksd1Mc3vEsA83GsAAHkAAAAAAGAuityarjeQPmtBT5ZPm0967v7v1WF7VtKW9+SQySagDFnuNe5HANBs8gAAAAAAwMwWq24AJRgsMiuqwGza5+50DOCThiJ/LwB6BrcfH9YHDn7NvWl6Cuopw+B1NirTug6BQfJAOeQByiAPAAAAAJRqYW0t2wbnC4qRmmXUANtgsc+05z1L4dqw1x73MwYD6Vfkvahl11qmm/8Qq6ursby8HLGyErG0NF8jVlcjlpdjZWUlluZ9LgonC1SsTveoMiaW+zPLsGuzTseL+pj1Puh6nGjWXEL55IGK1el+Ig/QVPJAYeSBetiysFB1EwBoqAPZpqxJgDwAQFEm5QErubXVuJWr5ilSm3agL+t2LQYCKWMyzbUGpGzSRPHgyjDDvj5M3ve9SSvSTPt64963QgvKNM/1JmMAeZEHsr0WFEUeAAAAAKiMldx4zKRV2CYNxBWx6lvW16b5yroHtehas5Ib05IFGi6v+1+W6yRLpuhflWXw8b2vDdvObZ4JdBglr/ufa3Ikn9WuD3mg4eQBGE0eKJw8UA9WbgGgKFZyqw95AICiTMoDh5XUDupg0mBd3pMZ0zyfiRQAqIdRk3adzqErysyTPXoT273nnKZ4HqaR53WV5boHaAJ5gKaRBwAAAAAqp8iN6YwahCtrK0naq4xPOvs0NdBm/VuPFzHxNs1zDk5QD/5s/6ous26ZDlVyvQKpkgegPK5XAAAAgKkocmN6gwPLZQ7KGQBst8FP/U/7swCMN9jPztLvDtsybNJjBr+WxxZnWZ8HxinyGnJ9AqmSB2AjeQAAAAAgCQtra9k2OF8w6EKKFC4xy8RH72f6P/k/6rFNMvC+M938h1hdXY3l5eWIlZWIpaX52rS6GrG8HCsrK7E073NROFmgxUbdGwfvp/1fy/I8/T8/60T2pJ9r8n2d4pVx33ONRsTsuYTyyQMtJg/QVvJAaeSBetiysFB1EwBoqAPZpqxJgDwAQFEm5QFFbjTHYMGSAcJ2cx1sNGQlBUVuTEsWaLmsE8r9E9ST7sHTXlNZJ8/H/RxkUfb9zjVqUrtG5IGWkwdoE3mgdPJAPZjUBqAoitzqQx4AoCiT8sBiSe2A4nW7BgR5jGvhMSYigTxMmsAe97hRsq7YMvjc7vEAUA15AAAAAACoyGFVNwBy1T8wrrCHtut2/R4AxZp0jynqHjTt87oXAkBx5AEAAAAAoASK3ACaqIyJpuWIWJjzz/L8zQASN67gdp570bDVW8at6GJim2l0OlYIAsiTPEAdyQMAAAAASVHkRrMZwKZtrN4GpGrUNmazThwOu9dVtZIMzTXv5HaWnzV5DrSJPEAdyQMAAAAASVhYW1tby/TAIgcBO52NA5sGHMmLQULaYob75tqMvx+rq6uxvLwcESsRsTTTc/Q9W0Qsx8rKSiwtzftcFK3QLABlkxGYVdZ74eA1NljI0f88bb0eu931957pf0pJgjxAo7T1/sv85IH8yAO1s2VhoeomANBQB7JNWZMAeQCAokzKA4sltSM7A+bkrW+wDBrHPRNgNvOuHEN7jbpmJl1Tg18fNrndFm18z0Ca5AFmJQ/Mr43vGQAAAJhLeduVjhswNKhBEfqvOdcYANCvf0Jx2FbPsgPTmnUrs7YVVvhdA1IiD5A3eSAbv2sAAADADMorcjNYQdkGB6hdg9SdQWCaqm0TOqRh2D2097XB/wLzG1Y8Av3kAaogD0C55AEAAABgDuUVuUEKDKRRR1YUoOlcz5Spv4hiWEGF6xHy5/eKLFwnlEkegPL5vQIAAADmpMiN5pp1iwhIybiVBaAN3MvJ27SrvLrnwnz8npEHeYC8yQNQLr9nAAAAQA4Wq24AFGbU4Fi3a4KENBn0hUO55pul03FOoQ38npM311SzyAPQDn7PAQAAgJxZyY12MtBGSoZtRwpt1b9Si4Lk5und6+p0buvUVkiBTEMe5IFmkweg+eQBAAAAoACK3GgvA26Urb+Yrfd31yEcqv/3hGbK49wOm2zOYwJ62JZ47tcwmd8T8iYPNJ88AM3j9wQAAAAo0MLa2tpapgcaoKDJ+rdL8Qlt8pbo/XNtxmt9dXU1lpeXI2IlIpbmbMVqRCzHyspKLC3N+1wULZcs0LvuEv29oCbK2Oas/x7Z/1pyAhwqh9/HWXMJ5ZMHSIY8AGmRB1pjy8JC1U0AoKEOZJuyJgHyAABFmZQHrOQGEY8OxJlooQiuJ9jIJ/vJQxnXUP+1arINRnNPZxbyAHmQByAd7ukAAABACRS5QY8BOQBgmP6MIC/QVsO2Wvf7ALSJ+x/IAwAAAEClFLnBMAboyIPrCKDerNgCj5JpgDaTB+BR8gAAAABQscWqGwDQSAZ/Aeqj05nuvt2/zTk0UZbfB1kHaBp5ADaSBwAAAIDELKytra1leqBBC9rIgDWzqsE9c23G63t1dTWWl5cjYiUiluZsxWpELMfKykosLc37XBRNFoABcgJNU+F9ftZcQvnkARjg/kXTyANMsGVhoeomANBQB7JNWZMAeQCAokzKA1ZyA8iTST+A9rCCC3XTf83KLAD5kAeoG3kAAIiIp0bEN6tuBADAlBS5wTgGq8nCoDBAe/X6gP68ID+QksGcIrcA5E8eIHXyAAAw4EkR8e2I+FnVDQEAmMJhVTcAoNYMDAMQ8Wh/0N8n6B8AoH3kAQAAauJfQoEbAFA/itwAZmXCAgAAAAAAAACgcIrcYBKFTAzjugBgmP7tyvQVANBO8gAAAAAA5E6RG8C0TFIAMIm+ghS4DgGq5T5MClyHAAAAQEMsVt0AgOQZEAZgVt3uo6u4QNnkF4B0yANURR4AAAAAGsRKbpCFQcH2cu4BmJe+hLK55gDS495M2VxzAAAAQMMocgMYxYAwAAAAAAAAAEDlbFcKMEr/djIK3gCAFMkoAIA8AAAAALSAIjeAUQwSAwApk1UAAHkAAAAAaAnblQIM6nYNEgOQL/0KeXNNAdSPezd5c00BAAAALaLIDQAAoE5MaAMA8gAAAADQMorcAPoZJAagKPoY8uA6Aqg393Hy4DoCAAAAWmix6gYAAAAwholsAEAeAAAAAFpOkRtk1e1GdDpVtwIAgLYwmQ0AyAMAAAAAEWG7UpiOgcVmc34ByMuownh9DVm5VgDqTx5gXq4VAAAAgHWK3AAAIG8mJJmH6wegGdzPmYfrBwAAAGAD25XCtGxb2kwGjwGAVAxmTTkFANpHHgAAAADYQJEb0E69YkWDxACUSaE8WcgnAM0mD5CFPAAAAACwge1KYRYGGpvBeQSgbPoexul2XSMAbeBezzjyAAAAAMBQitxgVgYcAQAAAAAAAACgcIrcYB4K3erHJ6IBgBTJJwCAPAAAAAAw0mLVDYDa6x+A7HSqaweTGSwGAFIkowAA8gAAAADAWFZygzz1VgmzWlhanA9oFwXHpE6fBFA8eYDUyQMAAAAAMBUruUGRJg1am3gByJ8JQwBAHgAAAAAAaBQruQHNp5gQgNQovgAA5AEAAAAAyEyRG1RF4RUAAAocAAB5AAAAAGAiRW5As3W7BosBZqEYu3j6J1wDQOrkgeLpC3ANAAAAAGSyWHUDAAphkBgASJWcAgDIAwAAAABTsZIbVMVgZnEcWwDqQp/VLlaYBWAYfUO7yAMAAAAAM1HkBgDAoUy8lcexBiBV+qjyONYAAAAAMJYiNwAAqJqJ7WazYgsAWegrmk0eAAAAAJjLYtUNAMiFgWIA6q6/L+t0qmsH+ZJRAJiGPNBM8gAAAADA3BS5Ac3Q6Rg0BqA5THDXn1wCwLzkgfqTBwAAAAByo8gNqtQb7DRYPR+DxgA0mQluAEAeAAAAAKDlFLkB9abADYA2mdTvmfQGgOaTBwAAAABoocOqbgC0Vqfz2B9mo8ANAB4jU6TF+QCgCvqftDgfAAAAALmxkhtUwSDnfBS3AQCpk1cAAHkAAAAAIDdWcoOyKXCbjwFiACB18goAIA8AAAAA5MpKblAWxW0AAM1lIhsAkAcAAAAACqPIDYqksC0/BooBYLxuV/YAgLaTBwAAAABoKNuVQlEMKudHgRsAAAAAAAAAQGspcoMiKHDLjwI3AAAAAAAAAIBWU+QGeVPglh8FbgBAqvpziswCAO0kDwAAAACUZrHqBkBjKG7Ll8FhACBlnY68AgBtJw8AAAAAlMZKbpAHBW4AQApMspbDcQYgZfqpcjjOAAAAAKVS5AbzUuCWr27XQDEAzKPXj+pPi+H4AlAH+qtiOb4AAAAApbNdKcxKcVv+DA4DwOwGs4msAgDtIw8AAAAA0FCK3GAWBokBgJTIJsVTjA9A6uSB4skDAAAAAJWxXSlMy6BxMQwUA8BsZJPiySkApE4eKJ48AAAAAFApK7kBAAAMYzIbAJAHAAAAAJKgyA2m4ZPRAO3U6ZjcIi0yCUD55AFSIw8AABTqwoj4t4h4oOqGAAAQEYrcIDuDx8Xodh89tiaMgJS5P5ECWaQ8fueBYdwbSIE8UB6/8wDQep+MiJ9W3QgAANYdVnUDAAwcA8AEJrTLI5cAkCp5oDzyAAAQCtwAAFKjyA2oTv+gsQFkABjOhHZ55BEAUiUPlEceAAAAAEiSIjegOgbpgaZo2/2sbe+3Kr3tvCmHCW1gXm27Z7ft/VZFHiiXPAAAAACQLEVukIUBZQDGadNkWKfTrvdbFdmjXK5pIA9tupfIA+WQB8rlmgYAAABImiI32mOWwWGfmAaAjUz+FU/2KJdrGmB67p3FkwfK5ZoGAAAASJ4iN5qtV6TWGxwe/G//46iGYw8Aj9EvlsuENgApkgfKJQ8AAAAA1MJi1Q2AwnW7GweIFboVr3+AuHdchw0aO+YANM2s27fpE8tnQhuAosgD9SEPAAAAANSGldxotlGDlQYxizN4bLtdxxuA9jChDQDIAwAAAACQOyu50T69webBFd6YzqyFa445ADxGvwgAyAMAAAAAMJGV3GgHK4mlYdjAvXMDQFuZ0K6O/AFAKuSB6sgDAAAAALViJTfaxQDm9MYds/7BeMcWAKgDmQUAkAcAAAAAakeRG/XU6Tw2INkrtMpajEV+DAoDwGxkEwBAHgAAAACAzBS5kb5hBW2Dfx983ODXR/0M4+VZxObYA8Cj9InVGpUZAaBM8kC15AEAAACA2jms6gbAWL1B31kHfw0az66swV7nCAAoQ7c7/IMTAEB7yAMAAAAAtWUlN9I2TaGVQcr8FFHg1u2OPjc+QQ0AFKk/Z8gcANBO8gAAAABArVnJjWYYNVDZ/wndeZ/fAGhxHFsAoEg+BAEAyAMAAAAAtWYlN+qlvxiqNzg5rEAqz6KpUQV0BkenN2y1PQVuALSJ/FAdK8cCkAp5oDryAAAAAEBtKXKjvsYNSo4rgMv79Zs2OF3WYG+qg8oK8AAoStMyAwAwPXkAAAAAAGaiyI3mGDZQXMbgcbfbnEHqoou6UjtO4z7BrcANgJ5h/Vev/8/aX6TWBwIA05EHAAAAAKBSitxohqoHiute6FZGcVu3u/E4pbJFyGA76n4uAcjXqD6h/+v6jfpIJX8AUC/yQLPIAwAAAAC1pMiN5hgcoCx7gLn/9es0uF32wO60n3Qvoy397anTuQOgHOMKoPUb1RmWvUblC+cJgHnJA2mSBwAAAABaQ5EbzZBCwVS/uhRMVXXcUjtfEYeeq8F/p9hmAIrXX5w9rG9Iva9vqsF+eVw/7RwBMC95IE3yAAAAAECrHFZ1A6DRFEY9KuXj0NtGdRID4gDtNWpCu/+/FKu/v87ab/f+DD4PAMxCHqiePAAAAADQaorcaK68Bi2HDYg2gUFdANogrz5cv1m+/onswe3Fx2libgNgPvJAfckDAAAAAPwfRW60wzwD0VlX+qqLJr2XPDXtPAPwqMF7ex4TnvqL/PWvytL7M+6DBrOcR+cNoL3kgXqQBwAAAAAYY7HqBkCh+gcvxw1kjhoYnWfwM+VPDRvUBaCt+rca618VpP/fk36+X8r9fV2M2nZMXgGgKPJAeuQBAAAAACZQ5AYRxQ6aDnvuqgbADQ4D0DaDW1r198GD/fG47a96jzWJna9Zs4lMA8A05IG0yQMAAAAAZGC7UijKuO0vqxiINfg7mYkKgObo39pqngnpcVtkMZ8issm4c+U8ArSPPJA+eQAAAACAjKzkBlXp3x6F9CkSBGgHfXM5iupXJ+WrcavzAECPPFAOeQAAAACAKShygyopdEuH8wDQLLNMXPZ+Rp8AAM0gDwAAAABAY9iuFNrC9iqjTTouPuEN0Hz99/rBv+fZD+T9fEzHsQdgHHmgHRx7AAAAgFqykhtUrezV3GzLMR3HCqA5+u/p/X3vsHt9Eff/URPnbStCL7pvneY8p0I+AyiPPJAGeeBQ8gAAAADAWIrcIAVlD2wbOAWg7Yb1g6P64LImYdswuV1W/kh5MntUDuu1ObX2AjSZPFANeUAeAAAAAJiB7UohNQYyyzfumLdhggGYnntD84w7p2Wd76ZngDLeX122Zx9sY9PPPTRVHe43TEceKJ488Bh5AAAAAGAqitwgIr3BTwObaUnt+gAgXynd55uYAbrdct/X4OuldH77pdougLZK6b4sD+T/eimd336ptgsAAAAgQYrcICLNAeQU21QHvU9sT/PJ7SyPM/AM9HOPbq7ehGiR53hSn9Lfhrpfa1W2P9VjN2zCfdQ1IX9A2lK9zzA/eSBf8sCh5AEAAACAqS1W3QBgjN6gZxEDmp3OoYO9vdfJOgg87eOLksKWMmVJ5ZgDNEnZ99RpX6/bbV5/xmMGz61zDVANeYAqyQMAAAAAE1nJDeqgjE9u94remlTg1hT973HYOWrDMQBou6r721mk0OZUV7+ZJXMBQIp92iQptFkeAAAAAGgERW5QF0Wt5tb7M8unyKscJJ5mO9I6m3RuZjl3QPracH9ro3n7rt79vg73/araaJUboEncz5pJHijndf3+AAAAADSOIjdIXRXFXKNer78orkpVvX4VrztpUqAOExvA9PxupyHV81B1P0x+Ur3GgDS4R6Qh1fMgDzRHqtcYAAAAQGIWq24AMEIVhW39n3Y2YD5cWdu0jlqhbfC8GAwHKF7/vXbUfThLvznvPbsOfXMK/VIKbZiG1WYA6kEeyC6FvjiFNkxDHgAAAACYSJEbpCjllcqqHihOZdC3im1CU3nvAG0yeK8fde8vo/gZAKiGPAAAAAAAlVPkBlWo88B0FcVdqeo/j0Uck8HV9QBop/7+YFR/U3VfIRsAQLHkAQAAAABa7rCqGwCtYRvQZsv7vHa7j00QVLVKAABpmNQfTPoeaZMNAchCHmg2eQAAAABgIkVuUBaDzc1X5KB0f9Fb798A0E/f0EzOKwDT0G80k/MKAAAAYLtSyFVbtvIsepvOOpv3Gph0bB1vAMaxzXV9OE8AFEUeqA/nCQAAACAzRW7UU6rFZIMDySm2MQ9NfV95GRyknnS8DGoDUHe9vkxGyJfjCUCdyAPFcDwBAAAAIsJ2pdRZ6oVBowYhe1+v6yBl1e2u+vVn0X+tDv593HWc+jUOQJqq7Cv1XfmpY+YBIB3yQDPIAwAAAADrFLlRL5OKglLQ7U4ehLR1SPvU4doFoDlMbAMA8gAAAAAADaLIrQoG+vIzaoUsilP1ca769fOg4G0u3/ve9+J3f/d344lPfGIceeSR8eQnPzn+8A//MH74wx9W3bR46KGH4ogjjoiPfOQjG77+ile8IhYWFmLfvn0bvn7KKafEW97ylkzPvW3btnjve9+bV1OBNugV3lsBJE3OCwBlkAfS5rwAAAAAZKbIrQoGsGYzqiiojgVDdWvvoCraX8fzTO6+853vxBlnnBHf+ta34sMf/nB8+9vfjr/5m7+J2267LXbu3BkPPvhgpe3bunVrnHHGGXH77bdv+Podd9wRJ5100oavf//734/vfOc7sXv37lLb+Mgjj5T6etAqKfdT8mf9yD4A9ZTyvVseqB95AAAAAGCdIjeapQ4Df3VoYxYGWotnAuIQr33ta+PII4+Mf/7nf46zzz47Tj755Pi1X/u1uPXWW+MHP/hBvPWtb11/7LZt2+Kaa66JV77ylbF169Z44hOfGNdff/2G51tZWYk9e/bEE57whFhaWopzzjkn7rrrrvXvv/3tb49f/uVfjr/7u7+Lbdu2xfLyclx88cXxk5/8ZGQbd+/eHXv37l3/99e//vV4+OGH4w/+4A82fP3222+PI444Is4888zYt29fvPSlL43jjz8+tm7dGs961rPi1ltvXX/s85///Piv//qvuOKKK2JhYSEWFhbWv/e5z30uzjrrrDjqqKPipJNOite//vVx4MCBDcfh2muvjUsuuSSWl5fjsssum+qYAxn1+kR9Y/v0MlFR2cg1BVAf8kB7yQMAAAAAhVtYW1tbq7oRANTH6upqLC8vR8RKRCzN+2wRsRwrKyuxtDT+uR588ME47rjj4p3vfGe8+c1vPuT7e/bsiY9+9KPxwx/+MBYWFmLbtm3x4IMPxlve8pa46KKL4pZbbokrrrgi/umf/inOPffcWFtbi+c973lxzDHHxNve9rZYXl6Ov/3bv40bbrghvvWtb8UxxxwTb3/72+M973lPvOhFL4p3vOMd8aMf/Sh+4zd+Iy699NJ45zvfObSdn/70p+NFL3pR3HvvvXHiiSfG+973vrj55pvjbW97W1x00UXx/e9/PyIiLr300rj77rvjM5/5TNx1113xhS98IXbt2hWbN2+OD3zgA/Ge97wnvvnNb8bJJ58cDz74YJx++umxZ8+e9SK1E044Ib761a/Grl274pprronzzz8/HnjggXjd614Xp59+erz//e+PiEeL3H70ox/FVVddFRdccEFERGzfvn2ms7Wg8BKaoaxJ0ibfMyYdw1HvfdZjn+Kx7L2XnNq2ZvK+NuQBaAh5YH7ygDzQUlv6PngHAHk6YMq6NuQBAIoyKQ8sltQOABpnNbfnWF3d+FybNm2KTZs2bfja3XffHWtra/G0pz1t6DM97WlPix/96EfxwAMPxBOe8ISIiDjzzDPjj/7ojyIi4rTTTovPfvazcd1118W5554bt99+e3z1q1+N+++/f/21ut1ufPzjH4+PfvSjsWfPnoiIOHjwYNxwww3xuMc9LiIifud3fiduu+22kUVuZ555ZhxxxBGxd+/e+M3f/M3Yu3dvnH322bFjx45YWVmJu+++O0499dTYu3dv/PZv/3ZERJx++ulx+umnrz/HtddeG//4j/8Yn/jEJ+J1r3tdHHPMMXH44YfH4x73uDjhhBPWH/cXf/EX8cpXvjIuv/zyiIg49dRT4y//8i/j7LPPjr/+67+OzZs3R0TEOeecEx2TBUBPbxKy7PvCLJOgOU+c5iLLcWviPXfYe0rpvAAwHXlgPvLAY1I6LwAAAECjKXIDYCpHHnlknHDCCXHffSfl8nxbt26Nk07a+FxXX311vP3tb5/qeXoLk/Zv5blz584Nj9m5c2e8973vjYiIO++8Mx566KE49thjNzzm4Ycfjn379q3/e9u2besFbhERJ554Ytx///0j2/FzP/dz8exnP3u9yO2OO+6IK6+8MhYXF+PMM8+MvXv3xqZNm+Kee+6Jc845JyIiDhw4EO94xzvik5/8ZNx7773xv//7v/Hwww/Hd7/73bHv+c4774xvf/vb8aEPfWjDcTh48GDcc8896wWBZ5xxxtjnAebQ6dR3Yq/bLXbydfC5pz1OqU1oN3GiOqss13lq5wugTPLAaPJAc8gDAAAAQMUUuQEwlc2bN8c999wTjzzySC7Pt7a2tqEwLSIOWcUt4tEtNhcWFuI///M/17fd7PeNb3wjHv/4x8dxxx039vV6r3Xw4ME48cQTY+/evYc85uijj17/+xFHHHHIzx88eHDsa+zevTs+8pGPxNe+9rV4+OGHY8eOHRERcfbZZ8ftt98eRx55ZGzevDme+9znRkTElVdeGbfcckt0u93Yvn17HHXUUfHyl7984jE+ePBgvOY1r4nXv/71h3zv5JNPXv/7li1bxj4PMIc6T+ClOEk7OHmawvFN4ThVWTyRdUI7S/vmLXQASFWd72cp9HOD5IHRbZAHAAAAgBZT5AbA1DZv3ry+FWZZjj322Dj33HPjfe97X1xxxRVx1FFHrX/vvvvuiw996EPxqle9akPB3Be+8IUNz/GFL3whfvEXfzEiInbs2BH33XdfLC4uxrZt23Jt6+7du+Paa6+Nv//7v49f+ZVficMPPzwiHi1yu/7662PTpk2xc+fO9WP4mc98Ji655JK48MILIyLioYceiv379294ziOPPDJ+9rOfbfjajh074mtf+1ps37491/YDDVf1JO2oFT56X6+6falKbZWgaVZz6Rn2+NTeF0BbVN3fygOzSa3flAcAAACAEilyA6A2/uqv/ip27doV5513Xlx77bXxC7/wC/G1r30trrzyynjSk54U73znOzc8/rOf/Wy8+93vjgsuuCA+/elPx4033hif+tSnIiLihS98YezcuTMuuOCCeNe73hVPfepT4957742bb745Lrjggrm2+Ny1a1ds2rQprr/++njrW9+6/vVnPetZsbKyEh/72MfiyiuvXP/69u3b46abboqXvOQlsbCwEFddddUhq8Vt27Yt/vVf/zUuvvji2LRpUxx33HHxpje9KZ773OfGa1/72rjssstiy5Yt8fWvfz0+/elPx/XXXz9z+4EGq3rCuP/1q25LHZW9BdiwCedpzmGWdprQBihf1X2wPDAfeQAAAABoqcOqbgAAZHXqqafGF7/4xTjllFPiFa94RZxyyimxZ8+e2L17d3z+85+PY445ZsPj3/CGN8Sdd94Zz3zmM+Oaa66J97znPXHeeedFxKPbjt58881x1llnxaWXXhqnnXZaXHzxxbF///44/vjj52pnbyvSn/zkJ/H85z9//etHHHFE7Ny5M37yk5/E7t27179+3XXXxeMf//jYtWtXvOQlL4nzzjtvfYvTnj/5kz+J/fv3xymnnBI///M/HxERz3jGM+KOO+6Iu+++O573vOfFM5/5zLjqqqvixBNPnKv9QENVPYk8zet3u2lMdlZ9zEZJrV0pnCsAsqm6D5EH8pNau1I4VwAAAECjLaytra1V3QgAyNu2bdvi8ssvj8svv7zqpjTGgkkLSFdqk5zzSuV+U4fjWuSxyvr+c2jDWh2ONREhD0DSmnYvTeV+U4fjKg9Qoi0LC1U3AYCGOmDKujbkAQCKMikPWMkNAADqptN5bMLRZGC7FXn+eyvoZJm0dh0ClE8eoEceAAAAAFpAkRsAANRVUycSm/q+ilLE8Rp8zlET3L1/p7LaDkBb9N+nm9pvNvV9FUUeAGiNIyLi96puBAAAVECRGwCNtH//fluVAvU2bqIy62oaddZbncYEdzZlTGz3mMgGKM+4e7E8wCB5AKAVfhoRH6i6EQAAUIHFqhsAAACl6nTqMxnXP6lYlzbnpW3vNw/DJqGzHsdpJ8WdH6Du5IF6aNv7zYM8ANAKP626AQAAUAEruQEA0C7dbj1WAxmcNJxnwrJuUnlfqbRjHoOr3wz+fdLqOHX4XQGYhTyQvlTeVyrtmIc8AAAAADSAldwAAGiPuk7QjZtc7U3S12WynmqMmtgGaKO63gflAeYlDwAAAAA1ZiU3AICUTFpFgdkNHtc6HOduN9vqISa0mUYTVqSBppMHiiMPwKPkAQAAAKBmFLkBAKSk6ZNNVU28jnrd3tcVE6QlhXPR5Guiqe8LmkQeKPd15YE0pXAumnxNNPV9AQAAAI2lyA0AIDVNnthO8b3Vfdumpk6+Vvmemng8gfpJsc/MS4rvTR5IkzwAAAAAwP9R5AYAAHXV9MnXpk7Y10mKhSgAbNT0vlIeqJ48AAAAACRgYW1tba3qRgAAAAAAAAAAAMAwVnIDAAAAAAAAAAAgWYrcAAAAAAAAAAAASJYiNwAAAAAAAAAAAJKlyA0AAAAAAAAAAIBkKXIDAAAAAAAAAAAgWYrcAAAAAAAAAAAASJYiNwAAAAAAAAAAAJKlyA0AAAAAAAAAAIBkKXIDAAAAAAAAAAAgWf8fr4qhrotFzLYAAAAASUVORK5CYII=",
-      "text/plain": [
-       "<Figure size 2500x1000 with 6 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 2500x1000 with 6 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "\n"
-     ]
-    }
-   ],
-   "source": [
-    "import matplotlib.pyplot as plt\n",
-    "import numpy as np\n",
-    "from matplotlib.colors import ListedColormap\n",
-    "\n",
-    "\n",
-    "def plot_model_predictions_with_errors(\n",
-    "    X: np.array, y_true: np.array, y_pred: np.array, year: int, day: int\n",
-    "):\n",
-    "    \"\"\"\n",
-    "    Parameters:\n",
-    "     - X: np.array of shape (window_size, window_size, n_forecasts)\n",
-    "     - y_true: np.array of 0,1,2 (window_size, window_size, 1)\n",
-    "     - y_pred: model output (window_size, window_size, 1)\n",
-    "     - year: Integer year like 2023, for titling plot\n",
-    "     - day: Integer day like 18, for titling plot\n",
-    "    \"\"\"\n",
-    "    X_cmap = ListedColormap([\"#0000FF\", \"#00FFFF\", \"#008B8B\"])\n",
-    "    binary_cmap = cmap = ListedColormap([\"#008B8B\", \"#00FFFF\"])\n",
-    "    # Calculate the incorrect predictions (difference between the predicted and true labels)\n",
-    "    incorrect_predictions = np.not_equal(np.round(y_pred), y_true).astype(int)\n",
-    "    current_SIE = X[:, :, -1].copy()\n",
-    "    current_SIE[current_SIE == 2] = 0\n",
-    "    change_from_curr_to_next = np.not_equal(current_SIE, y_true[:, :, 0]).astype(int)\n",
-    "\n",
-    "    # Plotting the first example of the batch\n",
-    "    fig, axes = plt.subplots(1, 5, figsize=(25, 10))\n",
-    "\n",
-    "    # Plot the last channel of input which is the most recent SIE\n",
-    "    im1 = axes[0].imshow(X[:, :, -1], cmap=X_cmap)\n",
-    "    axes[0].set_title(\"Most Recent SIE Input\")\n",
-    "    axes[0].axis(\"off\")\n",
-    "\n",
-    "    # Plot the true label for next day's SIE\n",
-    "    im2 = axes[1].imshow(y_true[:, :, 0], cmap=binary_cmap)\n",
-    "    axes[1].set_title(\"True Next Day's SIE\")\n",
-    "    axes[1].axis(\"off\")\n",
-    "\n",
-    "    # Plot the predicted next day's SIE\n",
-    "    im3 = axes[2].imshow(y_pred[:, :, 0], cmap=binary_cmap)\n",
-    "    axes[2].set_title(\"Predicted Next Day's SIE\")\n",
-    "    axes[2].axis(\"off\")\n",
-    "\n",
-    "    # Plot the incorrect predictions\n",
-    "    im4 = axes[3].imshow(incorrect_predictions[:, :, 0], cmap=\"hot\")\n",
-    "    axes[3].set_title(\"Incorrect Predictions\")\n",
-    "    axes[3].axis(\"off\")\n",
-    "\n",
-    "    # Plot the SIE change\n",
-    "    axes[4].imshow(change_from_curr_to_next, cmap=\"hot\")\n",
-    "    axes[4].set_title(\"Change from Current SIE to Next Day\")\n",
-    "    axes[4].axis(\"off\")\n",
-    "\n",
-    "    # Add a color bar for the SIE plots\n",
-    "    cbar = fig.colorbar(im1, ax=axes[0], fraction=0.046, pad=0.04)\n",
-    "    cbar.set_ticks([0, 1, 2])\n",
-    "    cbar.set_ticklabels([\"Open Water\", \"Sea Ice\", \"Land\"])\n",
-    "\n",
-    "    fig.suptitle(f\"Model Predictions for {year} {day}'s Next Day Forecast\", fontsize=14)\n",
-    "    plt.tight_layout()\n",
-    "    plt.show()\n",
-    "    print()\n",
-    "    print()\n",
-    "\n",
-    "\n",
-    "# Get the batch data for test set\n",
-    "batch_index = 200\n",
-    "X_batch, y_true_batch = test_generator[batch_index]\n",
-    "\n",
-    "# Predict using the model\n",
-    "y_pred_batch = model.predict(X_batch)\n",
-    "dates = test_generator.get_years_days_of_batch(batch_index)\n",
-    "\n",
-    "# iterate over the batched predictions\n",
-    "for i in range(X_batch.shape[0]):\n",
-    "    # Assuming 'model' is your trained model and 'test_generator' is an instance of AllDataGenerator\n",
-    "    plot_model_predictions_with_errors(\n",
-    "        X_batch[i], y_true_batch[i], y_pred_batch[i], dates[i][0], dates[i][1]\n",
-    "    )"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 80,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/",
-     "height": 211
-    },
-    "id": "_1dQwATjBex_",
-    "outputId": "7c05ae92-4efe-4a51-e554-c96d046278e5"
-   },
-   "outputs": [],
-   "source": [
-    "import numpy as np\n",
-    "from scipy import ndimage\n",
-    "from scipy.ndimage import sobel, binary_erosion, label\n",
-    "\n",
-    "\n",
-    "def find_sea_ice_edges(image):\n",
-    "    \"\"\"\n",
-    "    Finds edges in the sea ice class (assumed to be labeled as 1) of a binary image.\n",
-    "\n",
-    "    Args:\n",
-    "    - image: A numpy array of shape (height, width), where pixels are 0 (no ice) or 1 (ice).\n",
-    "\n",
-    "    Returns:\n",
-    "    - A list of coordinates for the contiguous edge pixels.\n",
-    "    \"\"\"\n",
-    "    # Apply the Sobel filter to detect edges\n",
-    "    sx = sobel(image, axis=0, mode=\"constant\")\n",
-    "    sy = sobel(image, axis=1, mode=\"constant\")\n",
-    "    sobel_mag = np.hypot(sx, sy)\n",
-    "\n",
-    "    # Threshold the Sobel magnitude to get a binary edge map\n",
-    "    edge_map = sobel_mag > np.mean(sobel_mag) / 10\n",
-    "\n",
-    "    # Optionally, perform erosion to thin out the edges\n",
-    "    # edge_map_thin = binary_erosion(edge_map)\n",
-    "\n",
-    "    # Find connected components in the thinned edge map\n",
-    "    labeled_array, num_features = label(edge_map)\n",
-    "\n",
-    "    # Extract the coordinates of the edge pixels\n",
-    "    edge_indices = np.argwhere(labeled_array > 0)\n",
-    "\n",
-    "    # Optionally, return the labeled array for visualization or further analysis\n",
-    "    return edge_indices, labeled_array\n",
-    "\n",
-    "\n",
-    "# Assuming `predictions` is a numpy array from your model with shape (height, width) and binary values\n",
-    "predictions = y_pred_batch[0, :, :, 0]  # Dummy data for demonstration\n",
-    "edge_indices, labeled_edges = find_sea_ice_edges(predictions)\n",
-    "\n",
-    "# edge_indices contains the coordinates of all edge pixels\n",
-    "# labeled_edges is the labeled edge map, useful for visualization or further analysis"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 81,
-   "metadata": {
-    "id": "iCg6CNxpd8Cv"
-   },
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "array([[   0,    0],\n",
-       "       [   0,    1],\n",
-       "       [   0,    2],\n",
-       "       ...,\n",
-       "       [1999, 1873],\n",
-       "       [1999, 1874],\n",
-       "       [1999, 1875]])"
-      ]
-     },
-     "execution_count": 81,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "edge_indices"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 82,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "<matplotlib.image.AxesImage at 0x781e144b9ee0>"
-      ]
-     },
-     "execution_count": 82,
-     "metadata": {},
-     "output_type": "execute_result"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "plt.imshow(labeled_edges, cmap=\"binary_r\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 84,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import skimage"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 99,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "for i in [0.1, 0.5]:\n",
-    "    contours = skimage.measure.find_contours(labeled_edges, i)\n",
-    "\n",
-    "    # Display the image and plot all contours found\n",
-    "    fig, ax = plt.subplots()\n",
-    "    ax.imshow(y_pred_batch[0, :, :, 0], cmap=plt.cm.gray)\n",
-    "\n",
-    "    for contour in contours:\n",
-    "        ax.plot(contour[:, 1], contour[:, 0], linewidth=2)\n",
-    "\n",
-    "    ax.axis(\"image\")\n",
-    "    ax.set_xticks([])\n",
-    "    ax.set_yticks([])\n",
-    "    plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  }
- ],
- "metadata": {
-  "accelerator": "GPU",
-  "colab": {
-   "gpuType": "T4",
-   "provenance": []
-  },
-  "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.9.15"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 4
-}
diff --git a/unet_model/jbacon_unet_modeling.ipynb b/unet_model/jbacon_unet_modeling.ipynb
index 3328016..aab19d9 100644
--- a/unet_model/jbacon_unet_modeling.ipynb
+++ b/unet_model/jbacon_unet_modeling.ipynb
@@ -11,13 +11,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 2,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
     },
     "id": "1EWdW30icnzz",
-    "outputId": "2e03f665-5e0d-4e17-8de1-51e3686a0672"
+    "outputId": "4803df24-0a0e-47a7-eb00-479f89deedf3"
    },
    "outputs": [],
    "source": [
@@ -46,13 +46,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 3,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
     },
     "id": "Wyo0Xe4vbX6n",
-    "outputId": "c96ca0ee-7e29-4316-e97b-79b12d977606"
+    "outputId": "31bb6f31-40a6-459d-dfd5-ca5d91cd44ee"
    },
    "outputs": [],
    "source": [
@@ -93,14 +93,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/",
      "height": 35
     },
     "id": "9gpzZZtUh30m",
-    "outputId": "9728f5a5-934c-4332-b5e4-e19e15de1e00"
+    "outputId": "ead2e019-cc19-4131-c41e-fbed771b0102"
    },
    "outputs": [
     {
@@ -109,7 +109,7 @@
        "'/home/bacon/code/personal/icedyno/unet_model'"
       ]
      },
-     "execution_count": 3,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -130,7 +130,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {
     "id": "f2Hsz8qQlGvo"
    },
@@ -138,6 +138,7 @@
    "source": [
     "import glob, json, os\n",
     "import datetime as dt\n",
+    "import datetime\n",
     "from IPython.display import HTML\n",
     "\n",
     "import pandas as pd\n",
@@ -147,6 +148,9 @@
     "from sklearn.model_selection import train_test_split\n",
     "from sklearn.ensemble import RandomForestClassifier\n",
     "from sklearn.metrics import accuracy_score\n",
+    "from scipy import ndimage\n",
+    "from scipy.ndimage import sobel, binary_erosion, label\n",
+    "import skimage\n",
     "\n",
     "import tensorflow as tf\n",
     "from tensorflow.keras.models import Model\n",
@@ -177,9 +181,18 @@
     "from tensorflow.keras.models import Model"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "3AgGBD1OkNDt"
+   },
+   "source": [
+    "# Set configuration constants"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {
     "id": "dzcSg_26piy6"
    },
@@ -188,9 +201,12 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "2024-03-23 12:35:41.844238: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2024-03-23 12:35:41.887384: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2024-03-23 12:35:41.887706: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n"
+      "2024-03-26 19:17:47.398220: E tensorflow/stream_executor/cuda/cuda_driver.cc:271] failed call to cuInit: CUDA_ERROR_UNKNOWN: unknown error\n",
+      "2024-03-26 19:17:47.398291: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: sven\n",
+      "2024-03-26 19:17:47.398301: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: sven\n",
+      "2024-03-26 19:17:47.398673: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:200] libcuda reported version is: 535.104.5\n",
+      "2024-03-26 19:17:47.398707: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:204] kernel reported version is: 535.104.5\n",
+      "2024-03-26 19:17:47.398714: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:310] kernel version seems to match DSO: 535.104.5\n"
      ]
     }
    ],
@@ -202,6 +218,19 @@
     "WINDOW_SIZE = 2000  # km"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 69,
+   "metadata": {
+    "id": "UxHWAk-CkNDv"
+   },
+   "outputs": [],
+   "source": [
+    "batch_size = 1\n",
+    "test_batch_size = 1\n",
+    "dim = (WINDOW_SIZE, WINDOW_SIZE, 5)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {
@@ -213,7 +242,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 8,
    "metadata": {
     "id": "J_uYc9VMujWV"
    },
@@ -236,7 +265,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 9,
    "metadata": {
     "id": "ui-tRfxBjY6B"
    },
@@ -264,7 +293,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 10,
    "metadata": {
     "id": "TbKOBHXaytLN"
    },
@@ -277,7 +306,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 11,
    "metadata": {
     "id": "l0SnpTY11oiw"
    },
@@ -297,7 +326,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 12,
    "metadata": {
     "id": "SR_qBhtuzRYP"
    },
@@ -324,7 +353,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 13,
    "metadata": {
     "id": "XK74-F6I6uv5"
    },
@@ -361,14 +390,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 13,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/",
      "height": 435
     },
     "id": "gbF6ZpdDukLT",
-    "outputId": "210099b5-9e63-4eab-d66b-54dcf33c131f"
+    "outputId": "4d33b3ae-b116-46d9-debd-253de823d62f"
    },
    "outputs": [
     {
@@ -415,7 +444,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 14,
    "metadata": {
     "id": "RVeGt-1c_zww"
    },
@@ -528,27 +557,27 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 15,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
     },
     "id": "kJL6GG2AWPmr",
-    "outputId": "06dea8b2-2efd-4b49-ec88-cf5300ff20c7"
+    "outputId": "835b8516-0c9d-441b-bd1c-c4f79a86d560"
    },
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Number of Train files is:  2344\n",
-      "Number of Test files is:  669\n",
+      "Number of Train files is:  334\n",
+      "Number of Test files is:  2679\n",
       "Number of Validation files is:  336\n"
      ]
     }
    ],
    "source": [
-    "test_frac = 0.2\n",
+    "test_frac = 0.8\n",
     "validation_frac = 0.1\n",
     "\n",
     "train_frac = 1 - validation_frac - test_frac\n",
@@ -578,31 +607,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 16,
    "metadata": {
     "id": "xWORYNWN_xyC"
    },
    "outputs": [],
    "source": [
-    "def simple_unet_with_sigmoid(input_shape=(4000 * 2, 4000 * 2, 5)):\n",
-    "    inputs = Input(input_shape)\n",
-    "    # Downsample\n",
-    "    c1 = Conv2D(16, (3, 3), activation=\"relu\", padding=\"same\")(inputs)\n",
-    "    p1 = MaxPooling2D((2, 2))(c1)\n",
-    "    c2 = Conv2D(32, (3, 3), activation=\"relu\", padding=\"same\")(p1)\n",
-    "    p2 = MaxPooling2D((2, 2))(c2)\n",
-    "    # Bottleneck\n",
-    "    b = Conv2D(64, (3, 3), activation=\"relu\", padding=\"same\")(p2)\n",
-    "    # Upsample\n",
-    "    u1 = Conv2DTranspose(32, (2, 2), strides=(2, 2), padding=\"same\")(b)\n",
-    "    u1 = concatenate([u1, c2])\n",
-    "    u2 = Conv2DTranspose(16, (2, 2), strides=(2, 2), padding=\"same\")(u1)\n",
-    "    u2 = concatenate([u2, c1])\n",
-    "    outputs = Conv2D(1, (1, 1), activation=\"sigmoid\")(u2)  # 0, 1, 2\n",
-    "    model = Model(inputs=[inputs], outputs=[outputs])\n",
-    "    return model\n",
-    "\n",
-    "\n",
     "def simple_unet_with_softmax(input_shape=(4000 * 2, 4000 * 2, 5)):\n",
     "    inputs = Input(input_shape)\n",
     "    # Downsample\n",
@@ -669,23 +679,19 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 17,
    "metadata": {
     "id": "TscI1oFMQo3y"
    },
    "outputs": [],
    "source": [
-    "# Example usage\n",
-    "batch_size = 1\n",
-    "dim = (WINDOW_SIZE, WINDOW_SIZE, 3)\n",
-    "\n",
     "train_generator = AllDataGenerator(train_files, batch_size=batch_size, dim=dim)\n",
-    "test_generator = AllDataGenerator(test_files, batch_size=batch_size, dim=dim)"
+    "test_generator = AllDataGenerator(test_files, batch_size=test_batch_size, dim=dim)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 18,
    "metadata": {
     "id": "HsQeOTpEbQPT"
    },
@@ -694,64 +700,71 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "2024-03-23 12:35:42.780161: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2 AVX AVX2 FMA\n",
+      "2024-03-26 19:17:49.982882: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2 AVX AVX2 FMA\n",
       "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
      ]
     }
    ],
    "source": [
     "# Setup model checkpointing\n",
-    "import datetime\n",
-    "\n",
-    "datetime_string = datetime.datetime.now().strftime(\"%I:%M%p_%B_%d_%Y\")\n",
-    "\n",
-    "# Model checkpoint foldernames now generated by datetime (won't overwrite previous runs)\n",
-    "checkpoint_dir = f\"./model_checkpoints/jbacon/unet_{datetime_string}_{WINDOW_SIZE}km/\"\n",
-    "if not os.path.exists(checkpoint_dir):\n",
-    "    os.makedirs(checkpoint_dir)\n",
-    "checkpoint_path = os.path.join(checkpoint_dir, \"cp-{epoch:04d}.ckpt\")\n",
+    "train = False\n",
+    "if train:\n",
+    "    datetime_string = datetime.datetime.now().strftime(\"%I:%M%p_%B_%d_%Y\")\n",
     "\n",
-    "checkpoint_callback = ModelCheckpoint(\n",
-    "    filepath=checkpoint_path,\n",
-    "    save_weights_only=False,\n",
-    "    monitor=\"loss\",\n",
-    "    mode=\"min\",\n",
-    "    save_best_only=True,\n",
-    "    verbose=1,\n",
-    ")\n",
+    "    # Model checkpoint foldernames now generated by datetime (won't overwrite previous runs)\n",
+    "    checkpoint_dir = (\n",
+    "        f\"./model_checkpoints/jbacon/unet_{datetime_string}_{WINDOW_SIZE}km/\"\n",
+    "    )\n",
+    "    if not os.path.exists(checkpoint_dir):\n",
+    "        os.makedirs(checkpoint_dir)\n",
+    "    checkpoint_path = os.path.join(checkpoint_dir, \"cp-{epoch:04d}.ckpt\")\n",
+    "\n",
+    "    checkpoint_callback = ModelCheckpoint(\n",
+    "        filepath=checkpoint_path,\n",
+    "        save_weights_only=False,\n",
+    "        monitor=\"loss\",\n",
+    "        mode=\"min\",\n",
+    "        save_best_only=True,\n",
+    "        verbose=1,\n",
+    "    )\n",
     "\n",
-    "early_stopping_callback = EarlyStopping(\n",
-    "    monitor=\"loss\", patience=10, verbose=1, mode=\"min\"\n",
-    ")\n",
+    "    early_stopping_callback = EarlyStopping(\n",
+    "        monitor=\"loss\", patience=10, verbose=1, mode=\"min\"\n",
+    "    )\n",
     "\n",
-    "model = simple_unet_with_softmax(input_shape=dim)  # skip(input_shape=dim)\n",
-    "model.compile(\n",
-    "    optimizer=\"adam\", loss=\"binary_crossentropy\", metrics=[\"accuracy\"]\n",
-    ")  #'binary_crossentropy', metrics=['accuracy']) 'sparse_categorical_crossentropy'"
+    "    model = simple_unet_with_softmax(input_shape=dim)  # skip(input_shape=dim)\n",
+    "    model.compile(\n",
+    "        optimizer=\"adam\", loss=\"binary_crossentropy\", metrics=[\"accuracy\"]\n",
+    "    )  #'binary_crossentropy', metrics=['accuracy']) 'sparse_categorical_crossentropy'\n",
+    "else:\n",
+    "    model = tf.keras.models.load_model(\n",
+    "        \"./model_checkpoints/jbacon/unet_12:35PM_March_23_2024_2000km/cp-0003.ckpt\"\n",
+    "    )"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 19,
    "metadata": {
     "id": "MsLENCJXfvkZ"
    },
    "outputs": [],
    "source": [
-    "# Log model parameters\n",
-    "with open(os.path.join(checkpoint_dir, \"model_params.json\"), \"w\") as f:\n",
-    "    f.write(model.to_json())"
+    "if train:\n",
+    "    # Log model parameters\n",
+    "    with open(os.path.join(checkpoint_dir, \"model_params.json\"), \"w\") as f:\n",
+    "        f.write(model.to_json())"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 20,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
     },
     "id": "X9LCjLY83Jkk",
-    "outputId": "b003fdfc-7934-456f-f55a-8dda5a51d7ab"
+    "outputId": "60c3b7c5-c4db-4c2f-c41d-83b5c6dfb632"
    },
    "outputs": [
     {
@@ -799,85 +812,43 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 21,
    "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/"
-    },
-    "id": "BIK7kxOlZTxw",
-    "outputId": "bcf637b9-a78b-4e35-9a71-d0ed8c55379f"
+    "id": "BIK7kxOlZTxw"
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 1/20\n",
-      "2344/2344 [==============================] - 7278s 3s/step - loss: 1.2299e-04 - accuracy: 1.0000\n",
-      "\n",
-      "Epoch 00001: loss improved from inf to 0.00012, saving model to ./model_checkpoints/jbacon/unet_12:35PM_March_23_2024_2000km/cp-0001.ckpt\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "2024-03-23 14:43:53.983464: W tensorflow/python/util/util.cc:348] Sets are not currently considered sequences, but this may change in the future, so consider avoiding using them.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "INFO:tensorflow:Assets written to: ./model_checkpoints/jbacon/unet_12:35PM_March_23_2024_2000km/cp-0001.ckpt/assets\n",
-      "Epoch 2/20\n",
-      "2344/2344 [==============================] - 7258s 3s/step - loss: 1.6113e-07 - accuracy: 1.0000\n",
-      "\n",
-      "Epoch 00002: loss improved from 0.00012 to 0.00000, saving model to ./model_checkpoints/jbacon/unet_12:35PM_March_23_2024_2000km/cp-0002.ckpt\n",
-      "INFO:tensorflow:Assets written to: ./model_checkpoints/jbacon/unet_12:35PM_March_23_2024_2000km/cp-0002.ckpt/assets\n",
-      "Epoch 3/20\n",
-      "2344/2344 [==============================] - 7407s 3s/step - loss: 3.6367e-08 - accuracy: 1.0000\n",
-      "\n",
-      "Epoch 00003: loss improved from 0.00000 to 0.00000, saving model to ./model_checkpoints/jbacon/unet_12:35PM_March_23_2024_2000km/cp-0003.ckpt\n",
-      "INFO:tensorflow:Assets written to: ./model_checkpoints/jbacon/unet_12:35PM_March_23_2024_2000km/cp-0003.ckpt/assets\n",
-      "Epoch 4/20\n",
-      "1885/2344 [=======================>......] - ETA: 24:22 - loss: 1.3922e-08 - accuracy: 1.0000"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
-    "# Train the model\n",
-    "history = model.fit(\n",
-    "    train_generator,\n",
-    "    epochs=20,\n",
-    "    use_multiprocessing=True,\n",
-    "    callbacks=[checkpoint_callback, early_stopping_callback],\n",
-    ")\n",
+    "if train:\n",
+    "    # Train the model\n",
+    "    history = model.fit(\n",
+    "        train_generator,\n",
+    "        epochs=20,\n",
+    "        # Only for training on CPU\n",
+    "        use_multiprocessing=True,\n",
+    "        workers=4,\n",
+    "        callbacks=[checkpoint_callback, early_stopping_callback],\n",
+    "    )\n",
     "\n",
-    "# Save the final model\n",
-    "model.save(os.path.join(checkpoint_path, \"unet_with_2d_output_model.h5\"))"
+    "    # Save the final model\n",
+    "    model.save(os.path.join(checkpoint_path, \"unet_with_2d_output_model.h5\"))"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 22,
    "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/",
-     "height": 449
-    },
-    "id": "85HoU8we3HsR",
-    "outputId": "c631f942-da43-426d-e686-9ab76257bd23"
+    "id": "85HoU8we3HsR"
    },
    "outputs": [],
    "source": [
-    "# Plot training history\n",
-    "plt.plot(history.history[\"loss\"], label=\"Loss\")\n",
-    "plt.plot(history.history[\"accuracy\"], label=\"Accuracy\")\n",
-    "plt.xlabel(\"Epochs\")\n",
-    "plt.ylabel(\"Metric\")\n",
-    "plt.legend()\n",
-    "plt.show()"
+    "if train:\n",
+    "    # Plot training history\n",
+    "    plt.plot(history.history[\"loss\"], label=\"Loss\")\n",
+    "    plt.plot(history.history[\"accuracy\"], label=\"Accuracy\")\n",
+    "    plt.xlabel(\"Epochs\")\n",
+    "    plt.ylabel(\"Metric\")\n",
+    "    plt.legend()\n",
+    "    plt.show()"
    ]
   },
   {
@@ -891,111 +862,43 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/",
-     "height": 626
-    },
-    "id": "b5LAJ53_LrhC",
-    "outputId": "566904e4-059f-47e2-83b4-054ba324073d"
-   },
-   "outputs": [],
-   "source": [
-    "def plot_model_predictions_with_errors(\n",
-    "    X: np.array, y_true: np.array, y_pred: np.array, year: int, day: int\n",
-    "):\n",
-    "    \"\"\"\n",
-    "    Parameters:\n",
-    "     - X: np.array of shape (window_size, window_size, n_forecasts)\n",
-    "     - y_true: np.array of 0,1,2 (window_size, window_size, 1)\n",
-    "     - y_pred: model output (window_size, window_size, 1)\n",
-    "     - year: Integer year like 2023, for titling plot\n",
-    "     - day: Integer day like 18, for titling plot\n",
-    "    \"\"\"\n",
-    "    # Calculate the incorrect predictions (difference between the predicted and true labels)\n",
-    "    incorrect_predictions = np.not_equal(np.round(y_pred), y_true).astype(int)\n",
-    "\n",
-    "    # Plotting the first example of the batch\n",
-    "    fig, axes = plt.subplots(1, 4, figsize=(25, 10))\n",
-    "\n",
-    "    # Plot the last channel of input which is the most recent SIE\n",
-    "    axes[0].imshow(X[:, :, -1], cmap=\"viridis\")\n",
-    "    axes[0].set_title(\"Most Recent SIE Input\")\n",
-    "    axes[0].axis(\"off\")\n",
-    "\n",
-    "    # Plot the true label for next day's SIE\n",
-    "    axes[1].imshow(y_true[:, :, 0], cmap=\"viridis\")\n",
-    "    axes[1].set_title(\"True Next Day's SIE\")\n",
-    "    axes[1].axis(\"off\")\n",
-    "\n",
-    "    # Plot the predicted next day's SIE\n",
-    "    axes[2].imshow(y_pred[:, :, 0], cmap=\"viridis\")\n",
-    "    axes[2].set_title(\"Predicted Next Day's SIE\")\n",
-    "    axes[2].axis(\"off\")\n",
-    "\n",
-    "    # Plot the incorrect predictions\n",
-    "    axes[3].imshow(incorrect_predictions[:, :, 0], cmap=\"hot\")\n",
-    "    axes[3].set_title(\"Incorrect Predictions\")\n",
-    "    axes[3].axis(\"off\")\n",
-    "\n",
-    "    # Add a legend for the values\n",
-    "    from matplotlib.colors import ListedColormap\n",
-    "\n",
-    "    cmap = ListedColormap([\"blue\", \"white\", \"yellow\"])\n",
-    "    labels = [\"Open Water\", \"Sea Ice\", \"Land\"]\n",
-    "    patches = [\n",
-    "        plt.plot(\n",
-    "            [],\n",
-    "            [],\n",
-    "            marker=\"o\",\n",
-    "            ms=10,\n",
-    "            ls=\"\",\n",
-    "            mec=None,\n",
-    "            color=cmap(i),\n",
-    "            label=\"{:s}\".format(labels[i]),\n",
-    "        )[0]\n",
-    "        for i in range(len(labels))\n",
-    "    ]\n",
-    "    plt.legend(handles=patches, bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.0)\n",
-    "\n",
-    "    fig.suptitle(f\"Model Predictions for {year} {day}'s Next Day Forecast\", fontsize=14)\n",
-    "    plt.tight_layout()\n",
-    "    plt.show()\n",
-    "    print()\n",
-    "    print()\n",
-    "\n",
-    "\n",
-    "# Get the batch data for test set\n",
-    "batch_index = 10\n",
-    "X_batch, y_true_batch = test_generator[batch_index]\n",
-    "# Predict using the model\n",
-    "y_pred_batch = model.predict(X_batch)\n",
-    "dates = test_generator.get_years_days_of_batch(batch_index)\n",
-    "\n",
-    "# iterate over the batched predictions\n",
-    "for i in range(X_batch.shape[0]):\n",
-    "    # Assuming 'model' is your trained model and 'test_generator' is an instance of AllDataGenerator\n",
-    "    plot_model_predictions_with_errors(\n",
-    "        X_batch[i], y_true_batch[i], y_pred_batch[i], dates[i][0], dates[i][1]\n",
-    "    )"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 23,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/",
-     "height": 617
+     "height": 1000
     },
     "id": "MXq734JLZAha",
-    "outputId": "08290386-35c7-4c6a-fc03-48060903152f"
+    "outputId": "3042bba0-6b66-444f-f93b-f84726582987"
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2024-03-26 19:17:53.199807: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 2500x1000 with 6 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
    "source": [
-    "import matplotlib.pyplot as plt\n",
-    "import numpy as np\n",
     "from matplotlib.colors import ListedColormap\n",
     "\n",
     "\n",
@@ -1010,25 +913,33 @@
     "     - year: Integer year like 2023, for titling plot\n",
     "     - day: Integer day like 18, for titling plot\n",
     "    \"\"\"\n",
-    "    cmap = ListedColormap([\"#0000FF\", \"#00FFFF\", \"#008B8B\"])\n",
+    "    print()\n",
+    "    print()\n",
+    "\n",
+    "    X_cmap = ListedColormap([\"#0000FF\", \"#00FFFF\", \"#008B8B\"])\n",
+    "    binary_cmap = cmap = ListedColormap([\"#008B8B\", \"#00FFFF\"])\n",
+    "\n",
     "    # Calculate the incorrect predictions (difference between the predicted and true labels)\n",
     "    incorrect_predictions = np.not_equal(np.round(y_pred), y_true).astype(int)\n",
+    "    current_SIE = X[:, :, -1].copy()\n",
+    "    current_SIE[current_SIE == 2] = 0\n",
+    "    change_from_curr_to_next = np.not_equal(current_SIE, y_true[:, :, 0]).astype(int)\n",
     "\n",
     "    # Plotting the first example of the batch\n",
-    "    fig, axes = plt.subplots(1, 4, figsize=(25, 10))\n",
+    "    fig, axes = plt.subplots(1, 5, figsize=(25, 10))\n",
     "\n",
     "    # Plot the last channel of input which is the most recent SIE\n",
-    "    im1 = axes[0].imshow(X[:, :, -1], cmap=cmap)\n",
+    "    im1 = axes[0].imshow(X[:, :, -1], cmap=X_cmap)\n",
     "    axes[0].set_title(\"Most Recent SIE Input\")\n",
     "    axes[0].axis(\"off\")\n",
     "\n",
     "    # Plot the true label for next day's SIE\n",
-    "    im2 = axes[1].imshow(y_true[:, :, 0], cmap=cmap)\n",
+    "    im2 = axes[1].imshow(y_true[:, :, 0], cmap=binary_cmap)\n",
     "    axes[1].set_title(\"True Next Day's SIE\")\n",
     "    axes[1].axis(\"off\")\n",
     "\n",
     "    # Plot the predicted next day's SIE\n",
-    "    im3 = axes[2].imshow(y_pred[:, :, 0], cmap=cmap)\n",
+    "    im3 = axes[2].imshow(y_pred[:, :, 0], cmap=binary_cmap)\n",
     "    axes[2].set_title(\"Predicted Next Day's SIE\")\n",
     "    axes[2].axis(\"off\")\n",
     "\n",
@@ -1037,6 +948,11 @@
     "    axes[3].set_title(\"Incorrect Predictions\")\n",
     "    axes[3].axis(\"off\")\n",
     "\n",
+    "    # Plot the SIE change\n",
+    "    axes[4].imshow(change_from_curr_to_next, cmap=\"hot\")\n",
+    "    axes[4].set_title(\"Change from Current SIE to Next Day\")\n",
+    "    axes[4].axis(\"off\")\n",
+    "\n",
     "    # Add a color bar for the SIE plots\n",
     "    cbar = fig.colorbar(im1, ax=axes[0], fraction=0.046, pad=0.04)\n",
     "    cbar.set_ticks([0, 1, 2])\n",
@@ -1045,173 +961,892 @@
     "    fig.suptitle(f\"Model Predictions for {year} {day}'s Next Day Forecast\", fontsize=14)\n",
     "    plt.tight_layout()\n",
     "    plt.show()\n",
-    "    print()\n",
-    "    print()\n",
     "\n",
     "\n",
     "# Get the batch data for test set\n",
-    "batch_index = 0\n",
+    "batch_index = 200\n",
     "X_batch, y_true_batch = test_generator[batch_index]\n",
-    "# Predict using the model\n",
+    "\n",
+    "# Generate predictions from the test data\n",
     "y_pred_batch = model.predict(X_batch)\n",
     "dates = test_generator.get_years_days_of_batch(batch_index)\n",
     "\n",
     "# iterate over the batched predictions\n",
     "for i in range(X_batch.shape[0]):\n",
-    "    # Assuming 'model' is your trained model and 'test_generator' is an instance of AllDataGenerator\n",
     "    plot_model_predictions_with_errors(\n",
     "        X_batch[i], y_true_batch[i], y_pred_batch[i], dates[i][0], dates[i][1]\n",
     "    )"
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": null,
+   "cell_type": "markdown",
    "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/"
-    },
-    "id": "36a9t8wpLxTH",
-    "outputId": "05d90f1b-fd6c-4c14-a32a-1055d3205bfd"
+    "id": "WB3g46XikND-"
    },
-   "outputs": [],
    "source": [
-    "(np.sum(y_true_batch == 0) + np.sum(y_true_batch == 1)) == y_true_batch.size"
+    "# Edge Detection"
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": null,
+   "cell_type": "markdown",
    "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/"
-    },
-    "id": "xC0edj7LA4Yn",
-    "outputId": "3401a434-68ab-4633-af3c-3a7eb92fd5d2"
+    "id": "-czr-A7KkNEA"
    },
-   "outputs": [],
    "source": [
-    "(np.sum(y_pred_batch == 0) + np.sum(y_pred_batch == 1)) == y_pred_batch.size"
+    "## Contour-based edge detection\n",
+    "This is maybe ok. It's closer to what IceNet does, but they also utilize masking out all boundaries/edges that aren't sea-ice to open ocean.\n",
+    "\n",
+    "It'll be very hard to line up which boundary should get compared to which. We get a collection of contiguous lines from this contour based approach, but when we get a different collection of lines/edges from the true target, how do we compare them? How do you know that you're comparing the contour around one island to that same island in the second image?\n",
+    "\n",
+    "Additionally, I don't think this is differentiable, which is fine for evaluation, but it would be great to embed this into our model if possible to force it to learn how to forecast edges better."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 24,
    "metadata": {
     "colab": {
-     "base_uri": "https://localhost:8080/"
+     "base_uri": "https://localhost:8080/",
+     "height": 1000
     },
-    "id": "_G_L9lhcBHCJ",
-    "outputId": "6d7e7131-fc2a-45d6-a443-9cc857fdc7e2"
+    "id": "uEhpfXM4kNEB",
+    "outputId": "8d84e714-e3f9-4c8d-9ce8-1b788b5215d5"
    },
-   "outputs": [],
-   "source": [
-    "(np.sum(np.isclose(y_pred_batch, 0)) + np.sum(np.isclose(y_pred_batch, 1)))"
-   ]
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAGKCAYAAAASfgYQAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAABue0lEQVR4nO3ddXhcVfrA8e/4TNytkVrq7kahtBSHomWBxRa3hSK7yw+HXVhcFmhxK+5QKKUtdVfqTS2NNS6TmWT0/v4YOul0kknaTLTv53l4mHuunWmSeeceeY9KURQFIYQQogHqtq6AEEKI9k0ChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIC0x3ui2+0mPz+f8PBwVCpVMOskhBCihSmKgtlsJiUlBbU68DPDcQeK/Px80tLSjvd0IYQQ7UBOTg6pqakBjznuQBEeHu69SURERL3HKIqCLSuL7L9cXu/+2BtvJO6mG5t0v215lUx/cxUAjopDHHr/Tu8+FXDtUC29Y9V8sMnB9hIFlSGU1Fve8x5zzqBknr5okHd7wqcTcCpOn3v8OO1H4kPim1QfIYToyKqqqkhLS/N+lgdy3IHicHNTREQEERER1O7ahWXpUhS3Am4XrvJyyj74EIAwjabeaxgKCxsMMkcLq1JQG0I85yV2R6UzoDhsACjAuxt9P/QVmwXzpl+IHH0RAL/sruT1I+6lGBU01NUr3hRPj6QeTaqLEEJ0Fk3pOjjuQHGk8s+/4NAjjzR63Dqrla8qK3g6OeWY7xEbpvfZjhgxjcqVnwc8p2r1195AcXTqw+6R3dlXuc+7PfeiucdcJyGEOBE0O1Ds+OR3Il4KHCT22mzMNZv5trKCLzO6estVen3DJx3FXOv7xBB50hVUrvoSFHeD57hrqrAXH0Af39Vv338n/pfZO2YTb4rnmgHXoNc0vS5CCHEiaXag2LQwn4lHbD9eeIi/x8UTrlaT43Dw9/w8dttshKnV/NC1G9HaultGTju/yffplRjGmO4xrNpXBoBKpfY0P9lrGjxHF5dRb5AA6BPThyfGP9Hk+wshxImq2YFCb6v0vj5z316yHQ4+q6hACxz5DHBnXBxJOp132zhwIKGjRjX5PiqVik9vGMPEZ38np6zh4HCk0H4nN/n6Qggh6tfsQNFz7zcA/FxVRbbD4S0/MkhMDgvjyugY77YqJIQuLzx/zPdSqVR0jQ1tNFAYuw8n7ow70ITHEeGoJMZRDoClopyKwkOYwsOJSQk8HEwIIYRHswOFo6aUO4qKWFhd3eAx1xwRJAB6LpiPNjq6ubcmYsT5uGurcVnKse5ZjVofQuSYS4gYdQEA3Sz7OaeorpN65k2/eF/HpXflzNtmkNC1e7PrIYQQnVmzA8XkffsYZjLxYpdUjBoNVW4XH5aUsLW21nuM6YhZfz3m/tKsIKE+YihX1ElXBjy2u3V/g/tKDh7gi8f+xU0zP0BnMB53fYQQorNrdqAYaDDycXqGT9k5oWEcdDh47FABuQ4H/YyeD2KVwYC+a9dm3W9K3wQW7y5u0rGTesVRuGFXg/ttVguVRYXEpWU0eIwQQpzomh0oHk1Kqrc8XafjnbR038IGJt4di7+O7cqQtGgOlFpwKwov/Lab7FKrzzFnD0rmP9MGsuLdnRT+WfZHbgE6jYYe8THojxh5FRIZ1ew6CSFEZ9bsQJF2xEimmGuuoeLLL3FbLPUea+rXr7m3A2BgaiQDUyMBOH9IF6x2J3anZz6FSa/BoPUEJJW6LjANSk2uv07hTZsZLoQQJ6qgphlP/Oc/6L1+HWlvv03M1Vf57Is4+2xSnj/2kU5NEaLXEhWiJypE7w0SAD1HjYUA09O7Dh4mmW+FEKIRKkU5OrlF01RVVREZGcmanpmEaTQkP/0UUdOm+Ryj2O1Y16/H0Ls32piY+i/UwioKD1GWn+PdttfUYK2sILlnb6JTumAMDWuTegkhRFs6/BleWVnZaM69oOR6AlAbTX5lKr2e0LFjg3ULL7fVgctsRxNpQG0M/BaiEpOISqy/H0UIIUTjghIoNLGxhI4fF4xLNapmWwmln+wEl+dBKPaa/pj6tM3TihBCnAia3UeR9s479FwwH00TcpoHg2VdoTdIAJS+vw233dUq9xZCiBNRs58oQoYMRm1smQlrGzZsYMGCBahUKiZNmsTQoUPRp4ZTu6PM5ziVVpb+FkKIltLszuymdIQcj9WrV/PLL7/4ld9/730ou6qxrCpAmxBC6OgkDOl196+u3kVp6SIOv62IyMFER42R0U1CCHGENunMDqaSkpJ6gwTA+x9+wK233kro8ES/fbW1BaxZey6K4t8UFRk5CrXaQEryhSQlnRf0OgshRGfV7EBRaneysczM7TuyqXA4cShgVKlwKApJBh3fDO1JulHPBRv3sK7SQpROw7LRfYnSNXzrgoKCBvcVFRU1uK+yalO9QQKgsnINAOXlS4mOHoPBkNDEdyiEECe2Zjfuj161nemb91Js9wQJgFpFwQXk2RyMXrWD5EWbWVVpwQmUOFz0WbaVrw6V4XDX3+qVmZkZ8J7btm1rtF7bttVSWVl/0Phjy82Nni+EEMKjzXqBb99xkIHLt+Kup4vEaDRyxx13NHhuUwLF0qUWLrowm6v+epCffjL77LMGyCorhBDCV4sFCsXlbPSYCqeL5eWedSycboXl5WYWlFZhcbqIjY1tMFhkZDSe7fWiiyKJidHw4UfpnHOO79BdnTaq8TcghBACCEIfhfWrj7F+/iGqiEjiP6vrgLZ+/SnVb76Etlc/VAYDitWCM2sn6sRk4j/92Xtcjs1OpcNJ72Vbfa67YGRv+sfGcuedd/LKK6/47EtISEBR3BQV/UJl1SZveUnxfO/r+HgtV14ZVW+dhwx5txnvWAghTizNHh57JE1qOqF/vQEcTqrffx13Sf3rRpjOvoCIex5u9B6XJUXzZGYqtsoKXn31VU+FVSoeeeQRiot/a7SvYeVKC2PHhvqU9exxPxkZNzV6byGE6MzabHisK/cgVU891OhxyhGr3wXy2aFyhkSEck2XOB599FFvudvtJCf3w4DnPvLwIdLSdH6BQq+PQ1FcqFTNXxtDCCFOBG0yj8K2aqlfmbvaTM0PX+LYtZ2ox57zlh+yObyvFcXNocIf2L79Hr/zKypcPPTQIQBycxyYzW4SErRMOCmUPn3qZo5v33E/23fcz8kTN6PVSuZYIYRoTFCbno6FOjqW0MuvA70ey6fv4T6U792nHzqK6OdnebcPTRoCwPoNl1NRsbre633+eQVvvVlW777TTw/jvvv9501MOmU7arXhuN+DEEJ0VB1iZra7vBTza8+CwYiu7wA0icm4S4tx5R7EXeO7tOkuSy0ZmjK/IJGVZWP+b9WUlztZtsz3nCP9+ms1O3faePSxRNLS9N7y/IKvSe1yeXDfmBBCdDJtm8JDoyX27c/RdqlbW7tm/s9UPe3bz1Fsd5CsKfUpu/wvBykqanwI7mHZ2Q6uvSaX995P9QaLXbseIiX5UtTqdpnJRAgh2oWgzKNQ6QygOvZLaRKTfYIEgGnKWUQ/N9OnTHtUQj+bzU1xcdODxJEefLDQZ9taI5PvhBAikGYHiqSrXyJ9xtdk/P0TjN2GNXq8pmt3dP0GAeDKz/FrZgLQDxnpsz0sIgSttm7SnMGgZuhQ3xX1EhO1jB0bQo8eegLJy3WQnW2vKzi+LhohhDhhNDtQ6KJTPC8MYYQPPiPgsYaJU4h792ti/vcBMa9/BFotJZeejm31sgbP+W+vVPRqNSEhXX3KH3+iLnvsmWeFM/uTdJ54MolZb6ZywYUNd8zExmrIyAgcTIQQQtQJagqPkN7jSL1xFsbuI+rdrx82yvta12cApqnnoliqqfjXHVQ+9WC950xP8ixzmpPzgU+50ajGaFQREqLinnviffYNHNjwQkp3/j3Ot85HBSAhhBC+gp7rSRPdhZhJ1x1VqEU/fDSao/ojIu59GFWoZy5D7YJfUBwOn/17Jw7EqFGzd+9z7M563GdfTY0bh0PhuedT/OpQbXbXWzedTsXYsSHe7e7d7katlqcLIYQIpEWSArpqfLO1ht98N9HPzsQwfLTfsXEffOt54XZTfPGUoyqnwmzezoHsN3zKs7JsLFtq4ZNP0+nVy38eRHKKrt566fUq1Oq6jvH09OvqPU4IIUSdoI8LtexaTsWi97zb6rgEQi5qeK6COiYOXf/BOA/sRZOc6re/onK9z/ZfLsumqsrNnJ+7NXjNozu6D9MckbUjNvYUNJqQeo8TQghRJ+iBouL3d3FW1g1BNZ3hu+yobdVSDGNO8imLefX9eq+lO2pYbFaWjVGjQrj6muiAdZj7i7ne8gsurJtJLrmehBCiaYLe9HT0EFlVuG+aD8OYk6j+8M1Gr3NnegJatQoVdcEiM9PA3TPiiYlpOL6tWWPl088q6t03cmTdk0ZoSM9G6yCEECLITxSKy0ltzhafMk1svP9xNZaA13lvQFfODAN2zyNWMYGiAlXT5ju88HwxJSX+S6AmJGh9kgNmZMhyqEII0RTNDhTO6jLKv/03AI6Sg7gs5b4HqPzPceXlUPPbHEynne2375leqZxZtRFemwaACTi5zxkUTbzEe4y1Zj/Z2TP9zgXQauu5IXDjjTHe1xpNKDpd4CRYQgghPJrd9FQy/xX0F56D4cJzUEL8h5q6igvrOQuq33rFr+w/mV24KjECPprmU67ZPZ+UlIu9/+Xlza73mvPnmzl0yD+1R1ychlMm1aUUT0/7W6C3JIQQ4gjNfqKIenUW6j/nQhAdQ+XDR60VofHvNA676W5K/3oetnWrMIwYA8CVybFclxoPi572v4nbtynJ6az2vl661MKeLBv7D9hZuaL+DLL33uvb/NW16y2NvS0hhBB/CmofhXHCqVSnd8N1sC7Rnru0xP+mXdJIXLjRpyzJoAOnDRY95Xf84cYkRXFz6ND3gKe/orjYybJlFnRaFevX1dSbtkmjgaHD6jqxQ0K6yyQ7IYQ4BkEf9aQf4pu+o+anr5t03tUp0fB0egN7PRGgtHQx23fc6y2Nj9fyr38lcO998cz+JB2Tyb9/wuUCh6Muglit+3A4KptUJyGEEC0QKCLueoDEhRtJ/HU1ETMeRKk2U3jGmIDnLBnVh/jsxeBseC1tRVHY/Mf1De6PitJw8in1L2363HPFPtt2e3G9xwkhhPDXIik8ANDpMZ1zEerYeLDb6g0WOpWKm9Li6RVigE/qRjXZnArTvzqyv0GFxbqn0Vse+eRwpFtuifXZDg2VORRCCNFULb+0m/bPW9htOPfvQdvN8yEdq9OyaVx/dGoV7Jzjc8pJ71k4M/OIqqm1GA3+yf+O9N57ZSxeVO1XrlZDbGzdtYYO+fA434gQQpyYmv1E8Xq/hvoVPDRxCd7Xzvxc7+uBYSZPkFAU+KwuF9TaPBfrC9w8dsoRqcKnvc7Bg2/Xe/2CfAcXTDvA7I8rcNaz6N2Vf/VN9xEdPS5gfYUQQvhq9hPF1LgoCrqlsclcg83tZmt1DQ9m5Xn3x7z6PtY532BfuRTj+FO85Y/0/PMJIes3n+vd8GMNw5OPil+pI9m/yTez7GEff1yOuYG04oMGGbnqqrpAkZJ8KSpV/RPyhBBC1C8oTU8qlYqhEZ5MrKMjQ3knt5j9NXXLjYacfSEhZ1/oc06PkD/Tg+9f7C0rsbrZUuTm5TOOeJoIS4IY/0yxc38xc/CgHYdD4Y2ZqURGanG5FN54o5gVyz39G4MH+y5g1K3bnc16n0IIcSIKeh+FSqVi2ei+/FZSxTVb99d7zE/DMtGr/Vu94kLUVD8QgenIWo273e+4Rx85hN2uMHpMCDfe5NtRffvtcaxYfhDwT+dRUbmOJOO5x/iOhBDixNYindkalYoz4iPZO3EgZ63PIq/W83RhVKv5cFA3hkWE1h0c6zsCyXR0jQb/xVNRbThOpyd9+KOPJTWpHnn5vivmHSr4hqRECRRCCHEsWnTUU6hGw+JRfQIfNPgysBTDrp8hf6P/frsFQuMIDe1F5VGLGNVn1sxS7+uFC6oJMam5/Q7POtk2u/8scSGEEIG13DyKptKZ4OT74cZFcOFb/vujMwDo0eM+b5HZBb+btay3aHAdMXWistLF4sV1KcydTpg7t24Ro+rq7Vit9TeHCSGEqF/Lz6M4FoMuhT5nQ8VBCIkDY92iR2GhmajVekrsDh4vqMvdZHHbmRjuGRe7fl0tmZl6srLqOtKjo32TEhaXzCcj/YYWfiNCCNF5tP0TxdH0oZDQF8LiQVuXvE+niyKj90s+QQIg31H3Fk6dHMobM1MZPMRIdLSGjAwdDz6Y4HN8dFTgdCJCCCF8ta8nikasrazwKzuv+5lQ8QOHEwcCPP98/bO4e3S/l4iIgS1UOyGE6Jza3xNFADXOGr+ynhk3cNKE1fTr93yj52dk3NQS1RJCiE6tQwWK8V3G+5VlRGSg18cSYgqcSgRApepQb1cIIdqFDtX0lBmdyW8X/8Y3Wd8QpgtjdPJoQnQhlJevYsPGKwKem5h4XivVUgghOpcOFSgAkkKTuHXIrT5lpWVL6z02MfFcrNZ9xMVNobuk7xBCiOPS4QJFfeJiJ5GdPdOnbNjQT4iOHt1GNRJCiM6jUwSKqKgRjB+3lLKyFWg0RsLC+hIa2qOtqyWEEJ1CpwgUAEZjCikpF7d1NYQQotORYUBCCCECkkAhhBAiIAkUQgghAmp2oHBbrcGohxBCiHaq2YEi66SJuMzmxg8UQgjRIQWl6angoYdRFKXxA4UQQnQ4QQkU5rlz2dm3H1Xz5qG43cG4pBBCiHYiqJ3ZeXf+nZ39+lM1dy5uqxXF6VlQyG23e18LIYToWJo94a4isjth1dk+ZXl33V3vsaEnTyTlqafQxsQ097ZCCCFaSbOfKHK6TOKj8jL22GyNHmtZvISsceMxL1jQ3NsKIYRoJc0OFAcr83mhuJjzDuznzH17+b6ykmWW6oDn5N52O5YVK5p7ayGEEK1ApRzncKWqqioiIyPRaQw4XP5PEyEqFWeERxCn1bLHbqOXwcCdcfHe/bouXei5YP7x11wIIcRxO/wZXllZSURERMBjm91HUV+QALAqCt9UVXq3F1ZXs8pi5ZOMDM95eXm4a2pQm0zNrYIQQogW1KopPDbV1pBjt3u3yz76uDVvL4QQ4ji0eq6nDTU13tfFL7yAqzpwf4YQQoi21eqB4smiQp/t4hdfau0qCCGEOAatHigsbjc/HtF3UT57NlU//9za1RBCCNFEbZJm/PniYp/tvBn3UPPHH21RFSGEEI1ok0BR5HRy4YH9PmXZV13dFlURQgjRiDZbuGinzcZtubnebaW2ltrt29uqOkIIIRrQpivc/X7UDO79F16Eu7a2jWojhBCiPm2+FOotuTk+24cefayNaiKEEKI+bR4ollosLKyuWyGv8rvvcOTnt2GNhBBCHKnNA4UbuD0vz6fs0ONPtE1lhBBC+GnzQAEQq9H4bFcvWoSjqKiNaiOEEOJIbR4oehkMLO2Z6Vdu37u3DWojhBDiaM3OHtucG7+Xls7wkJB694eMGNG6FRJCCFGvVg8UfQwGrouJ5ZwG8p+HjB5NlxdfQKXTtXLNhBBC1KdFA4UaeDmlC+PCwnABYSoVbqA0zESFy01UTd1aFqmvv074qZNasjpCCCGOQ4sGiqEmE5PDw33KNmYkUhgVBkDfvBJ6R8bR/ccfUKnbvLtECCFEPVr00/nMcP/mpcNBAmBHlzjSvvpCgoQQQrRjLfZEoQIuj472bkdecAHa2Bhit6+n1FI3wU5vqr8zWwghRPvQIoFCA7yflo4mtif6zNNxW0qovnQaq7b8QcqgaxmmVQiLjiGha4+WuL0QQoggapFAcVp4OMPDozGNuROVzkgtDj7/4Qdcipt9+/Zx9tlnM2joyJa4tRBCiCBrkc6BHnoDKp0Jlc4IgE3lwKW4vfvnzJnTErcVQgjRAlokUKgBpbYCV8VBACKVEML1od79U6ZMaYnbCiGEaAEqRVGU4zmxqqqKyMjIevct79GTaK2nVSv9vXcJHTv2+GsohBAi6A5/hldWVhLRwATow4L+RDHIaPQGCQDNESOfhBBCdDxBDRSxGg2fZXT1KTP07h3MWwghhGhlQQ0UM+LjfbbT33sXlUoVzFsIIYRoZUELFOk6HRdERnm3I6dNk74JIYToBIIWKE4JC/PZTv7Pv4N1aSGEEG2o2RPuPkpLp6vRSOwR+Zribr9d8jcJIUQn0exA0dtgIOyooBB28sTmXlYIIUQ7EfSv/ZqYGIwDBgT7skIIIdpI0HI9GXr3JvovlxF10UUy0kkIITqRoAUK265dRF92WbAuJ4QQop0IatOTy2xu/CAhhBAdStACRcioUWiOWvZUCCFExxeUpqfEhx4kevr0YFxKCCFEO9PsQJEx+2NiRo0KRl2EEEK0Q81uejL26ROMegghhGinZPq0EEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQISAKFEEKIgCRQCCGECEgChRBCiIAkUAghhAhIAoUQQoiAJFAIIYQIqNmBoqj4N+z2UsodTuaVVFLpcAajXkIIIdoJbXMvsH37vWzNjuRe7QdUuxQAxkaF0jvURIRGzRUpsWSYDM2uqBBCiLbR7ECRTVf+rXoR/gwSACsrLKyssADwe5mZ30b2bu5thBBCtJFmNz09oXoi4P6dltrm3kIIIUQbavHO7EuSolv6FkIIIVpQ8AKFWwFF8SnSqVS80Cc9aLcQQgjR+prdRwGg2WtGt6cKAPuIWNyxRgB6hxqxuFyEajTBuI0QQog20OwnivOLFvLPA2+TocpDG7mO0PzPUTuLANhaXUOPJVsotjtQXC7KPvyIrEmnUjVvXrMrLoQQonWoFOWo9qImqqqqIjIykqL7oogzafgtxMQ/EuJwqp3onXGMKnyCCKubJf1N7E/S8YejmNI77/Se33PR7+iSkoL2RoQQQjTd4c/wyspKIiIiAh7b7CeK73Ons7/mE/ZZ7uH87PO5aP9F9C8cxICDdtJLnFy52AyKwoYDudi7uqk6z4krWqHomWebe2shhBCtoNl9FD0ihlCmsXFIXeEtC3eE+Rxz3fwqvp5q4rL7PbO2q89w06Prjc29tRBCiFbQ7CeKWpygC/cps1bafLa7lLmwu/f4lO2rebO5txZCCNEKmh0obp/Sm3MmRTHz5GmYFD1p1dEsO+QbKAojNaxjtE9ZSpfLmntrIYQQraDZTU9mo8kbba6wncS8/cvIrdrHk3+LJiqvBl2OlYpQcJaG81z617yRqSUtMhO1WtfcWwshhGgFzQ4UW5afzYGYLlwx5C0A3lj9CbWHdlLhVtANvx4dWrrmFVO938TGM3vzTmkcT8RIkkAhhOgomh0oJmV0QWPSkL7qVjKe2YpbcQNQu2geoUP+xsyFT5FQUwFAycoorntuFk9kpjb3tkIIIVpJUGZmAxSnu71BAgC3i7SfniZBVe0tiq62km6UJichhOhIghYoAEJ6hlBbEUbEPx5H17MXO7duho9noQDb+15DYeJILny3ANuAnhhCJGAIIURH0OyZ2X3f6IvG5MnlVH0wjJoJb/gcZ6yt5eRN2xmV3dVbFpUYwhWPjTn+WgshhGiWVp2ZfSTVgFP9ymqNRhYNG+JTVl0ua1QIIURH0fwV7l7MJuGZB3BrorGFjKj3GJtejfb8VJQ5+ST3jGT8xT2be1sAFEUhb8c2qkqKiEvvSkLX7kG5rhBCiDrNDhSG21+jJmJIwGP6hRq5cHw34s/s1dzb+Vjz3Zcs++xD7/Z59/4fmSPHBvUeQghxomt205O2W+NPBzq1imqnu9HjjtXejet8tpfOfj/o9xBCiBNdUEc9NWSzuYaxq3cwMiKU74f1RK1SNfncUruTeaWVDAkPoW+YCQCry82snCL2WJ0c2dhUXpDHK3tyCTMaiNBqOCMukjCtLJokhBDNEbRA4Sos8LxQqVBqrNjXrcJ0/iWotHXDYNdWWUhZtJmz4iLpHWrklvQEIgJ8kO+01HDKml3e7YFhJjJMen4qrgQgfch4uufs9u7fn9qTr3JKfK7xYPdkbkyLR69u8eXBhRCiU2r28NjGGE89g8gHn0LjVrhnp42ph5z8nKzlhT4G1Iobk0qFW6NBhQqVCg4/a6gAs6sJzVVHVj/Ak8qW8f2J1x//3I28WjvlDqdPWWaoEYMEICFEB3Qsw2NbPFAAoNZw7vNf8fqhWG/RJcNrmLRzDXqXkz3xXZjfb+Qx18G2ehnm155Fm94dDAZQFFRqNYaTJmM8eYrf8QPCTExLiKJniJHT4iLQNNIEVmx38FNxJU/vK6DS6ar3mH91S+b61DhCpYlLCNGBtL9AAZyeOYG3L/wPANbonSzr8h2W8kRKS9MB+F/6gD+fCP58rFCpPM1YDgfq0DDclmpwObF++TH2TetQaqwoluqGb2gwEvfuV2iSuzR4yKz+Geyx2FhfZfHb51Jgcbm5ye9vx4QBROtapctHCCGarV0GCoC/j7uaGadcQdbkm71lu3aOo6ioB4899tjxVCMwnQ7T6ecRMePBoF3S+sOXaOITMYyd6LfvvIQo3uzfNWj3EkKIltJuAwVAfKyBT7+o+5a/a9dQnnt2A/v37z+eajSNTo9+8DBUeiNhN9yBNuPYJuZZv/0M28rF2DeuA5enn0IdE0fkg0+hH+I7yTBJr2X9uP6NNmsJIURbateBAuCWW2K56GLPuTfflMuePfbjus7x0nRJJ+zGO1HHxqNSa3Bs/4Pq99+o91iltgacznr3AZimTSfizn/6lX80sBunxR3fv48QQrS0dh8oABIStJSXO3E4jvsS7UbUf1/DMHKcX3n2yYNkVJQQol1qs6SAx6KoqHMECQBd34H1lg9fsR1XPXG42O7g1exCZueXUtuUIcBCCNGG2uyJojNRJyQR/9kvgGfioSYx2Wd/n1AjLkXBpPHE5T/MNT77+4cZ0aBiUHgIj2d2IURzYj+F1Lrc2P/8tXzjYBGLy82Mjgzl4R4pqKTvR9TD4XLz/LzdrNxbglqtYkrfRLbmVZJb7vlb25LnmaQ7uU8Cj5zbn/TYkLasbrvQIZqeOiN1QhLuinISflyKSnf8k/vOiIvghT7pxBzncFuz08XKimpGRoYGdciuze1meXk1tW434RoN46LDjrnT3u52s6y8msxQI2lGPQBuRWF1pYVyh5N3c0tYVlH/sOch4SHMHRHcxJKiYzPXOnjypx18vi7nmM57YtoARnWNoXdSeAvVrP2TQNEORD8/C23v/qhDQv32KU6HN7WJUlODymSq9xo3pMZxV0YSsXrPh32hzcHH+aWU/DlDPEqr4fzEKBaXmTlQ4xkQYHa6+Kqw3HuNs+IiSTLoOCMukokxx/9HUWx3MHD5Nr/ya7vEeV+PiwrjnPjIer/1O9wKXxwq455ddX/QY6NC+U9mKueu3MD5xQup1IbzS9wEXKqGg9uuCQOIlPkqArDanUx85ndKqo9/MMzUfonM+uvwE/JJVQJFO6IbMoKQ8y5BpTdQu3AutQvn+h+k1hD6l2sI+9vt9V4jw6hHrYL9Nc0bHdbVpOedAd3oH1Z/YApk+qa9TZqAODE6jFf7ZpCg13r/+BRFofuSLdS46++PmbX9UcIsq/gwMpz+BTpitw7Gro9ka8IkjKowFo9LIDfOE1g/GNiN02U02QnP7Va4dfYG5m475FNu2bkUxVGL2hgBKhXWHUuw7l1L5LjpRI66sN5rje4Ww4d/G4XhBMuuIIGig1KFhhEy/WrCrrw+8IFONyqzAyVSD2r/b0KmGgsX/vIRKUW5FMR34eOLbvHu62rS81RmKhVOFyrglJhwoprwDX3C6h3ssdp8yjKqXTy7qZZcx252anLIj4zl54FjcKs9f3AbxvYjxahnXaWFczZkAdDDehCXSs0BU6r3OgeWnMLYrqncGG+jl9FNjd3IU5ZnORiRzLgdNUzcXssT02MAGB4Rwpzh0vx0IiuorOHSWSvJKavr63OU5VH4yT9xWcoDnAm6hO4Y0/oTM+Umv31vXDGMMwcm13NW53QsgUKe4dsRxVKN5d3XsHzyDmHX3OrJV6XTe76Zq1WgUqPU2gnZYEX1Z/OMdYQJ1GDfuAbL555FnE5OjCQlxZNXK7k4D+eBvWi79gDgQI2dv/yxz3vPeJ2WdeP6NTqMN0zj+fDX2najLVtDzKFo7tlURYomjYUZJagVSK0oZlDuXjalez7Ih63cTv4pgym0e4a3/S33K/6991UA1kQM4LyhrwFwd69/0Fv9Ib2MnieOTfrh7DWkALArVc9J22sZv72GrRl6zCH159wSJ46fNhf4BAmgSUECwFG0D0fRPqr/+I30GV/57Ltl9gZmXjmcMwYkBbW+nYE8UXQwof0nEXfOPd7tvLduxlmW63PMgC5JXDN+uHf73i/mEPfVfDQxsTTk95G9vet9gKe56GCtHb1aRbJBzyN78nhr/26Ss+9ixI4oBu430LugnNTyahZOOpWuMWOIUyJYFGPnxWE9683ke2jxyQAsZSR76MqHmWezI6UbAJnV3/Jo6McArGMEBmx8pVxOhbkbUzbW0L3QgQYVT0yPaXYmYNGxdfvXHG/SaEfFIYo++z+clYXHfB21MZzIcdOJGDnNp/ytq0ZwWr/EINS0fesQ8yjE8bHuXumz7bZW+B2zNe8Q7y3zrP63aOdeAEqvvyTgdSet3cWOas+3NIdb4fT1uxm9agdDV2wn6fdNzDpYxODK1Uw+qKVXbjihoRnUDjiHsHNf4/zIyxjs6koXdwxXlCSxdEE1g8r9v/k/1fV6iohlARPIJpWTszYTXuNJyJgVdgGfK5fiViDGtoEFjmHsUfeixKjmG10NL0fWssLgeTJ5al/Bsf2jiU7B7Vb4eFW2z8oCJT8+e1xBAsBda6Z84dsUfDjDp/yGD9fx8xb5HTuSPFF0RCo1MVNvxbLtd2x5O0Bp2qQ9dVwCkY88g8powr52Bbp+g9APGsb4TWvpe2APC0eMo/vggWyoslDm8P2gvzXnUx7eNxOAKkc8Va73Gr3fqZPCqNL7PlmM/uENhkbWtQP/NHAcuTEJ3u3E0ldwW9YCUBM6kRrr5YRuq2CQTYsKWHtmPLV6NdvGD/COBhMnhveW7+exH7d7txWXk4PPTQvKtfVJmSRf/aJP2dr/m0J8uCEo12+PpDNbNNmgiZP5rLQUnLUAnPHy+9j0/n8ch5uNDsut/ane65lVNegULUY8TUOXjw1h6/5NOLZuwvrNJ5jsNh646i/oKopxhUXy/rnXURHm+T0yKVbCcm4EVCioUNShVFQ/xdSddkbY6oLCE9Nj6B9mZMHIPsH4JxAdxN/eX8uCnUXe7aJvnqQma1XQrq9P6kny1S95tzt757Z0Zosmezh6POFjPYs8WRY8gtblwnbUMdUfvsl9rm48292T4det+M5q1SWFEj4pjSX561i1ejWKAt3sg0hSdDz488dc8fFblFs930ciIsMxlHoe69WVpYRWV1ARFskNymucwkJyk8J4qdhIUcKjuHSeP1JTWTXk+A4N3lZdy43bDjCrX8YJOQb+RFe+6P2gBgkA+6E9VP8xj7BBU4N63c5A+ihOYJGGMEb3qVsJUNdtErlXnkfV80/g2LsbxeHA8sm7qNZ/xGyliN6/Kny2LZVDtrd9ruM4ZEEToWfV6tUA7HbF829FwxjDTUxN/ozi+8IZmuT5VTvy8TXCauPCx/7J2St/4xRlIQCpumqSirp7gwTAN+PCvK/zout+ZX8oquDW7dlB+/cQHUf0KdeQdNUL6BK6NfkcfXIvoiffSPSp16NPyvQ/QKOVINEAeaI4gVXafFNlPLBrNfbyUpjzDTVzvgFAG6mlz8t/NvFMieWmp7I5O2sm49OHcX6/uiBTPOsPIvUhVKqt5LkjSVcVEqWqWzlwTKqGjYfcHKo088XaP7h0xEAmZP05WuvDd6nO1VB1sadfpEv5DjYePlFRCLWauePN00iOzqDAWUbC1EUA6F0KC3PLSCqq4NEeKVyfGo+2nnklonMY0z3Wp+nJkNyLlGtfxVmWh+K0Ub1jKZZtC0FRUBQ3uN24a6oAMHYbRuKlj3vPjRg5DVveTmqzN2HKHINKpUZx+j5LD02Pbp031opK86r56X+bsVmdpAzwzxrRkHbXR5GcnExUVBRZWVk4j1gHon9/A5GRGtassQZaHkIcI5POyFeXv8r+shweX/gaRZZSn/2G7pGkPX0RWschtI6DHHj+ANVbPAFmYteRzJ7+vPdYKzZW6HbxlCMdGyoOGK/w7hv9toU1eXUd5Aa1mo2ZdRPnygY7qb3JjcOt5aHXUtil60PITTcw/af3SM/3NHk9+dMCKqy1YDAy7otlzFprJdphwaFfx5WjB7InJIM70xMI02o4Jz6KnFo76yotxOu1XJAYTfgJNvO2s3G5FebvKOSmj9Y3+RzFace88We/IbBN8cn1oxnXM67xAzuAouwqvn52PW5n3cd9jd3Cfe+d1zE6s8OHhBMxMgLzZjMxRTFcffXV3n2Hl0c9ZVIoDz5YN655yuR9ftc50aTpdFwVHUOly8UH5WWYG0iP0VzRL72DftAwAExVP5N3w6M4iuvyww/vMoC7x19DcngCveK6ApCHm+l4gsn51V+x8MsP2VHsxvXnb9rwHpPolz6KmMIN3OjKQQXcnJuDe6hCftc7cXWZAIArdzN3OVZ47/Xthq0s3+Npanr9oic5t+dJJOj/jl7t+X2YMPJD9oRkNPheZF3zzsHtVvh2Yx7P/LoTm9NNhbVl1ivoHh/KwntOaZFrtxZFUdi8IIflX+3x23csgaJN/2o0YRoy7sogtkLPOKUfsVV6lB3rsXbri9tY12F6+eVRbVfJduq55BQG/plM8Na4OPrt2tki9zkcJADMFX19ggTA+rytLDuwnnsmXEeZtZIp715NsaUMgNPOPp/oEUOYNP02st58E5fNRmx4MtdO+XMN815TufSbmzhQug+L2w3LIGP8BO+1NamDUecuJePUfCK7VmNb7mLNZ4OImPEgj3XPxLmpiosrKjjcm/FU1ktcMth3iOORXsku5JGeXRrcLzoGtVrFRcNTuWi4Jw1MUVUtby/bT155DW5Fwa0o/Lqt/rkVX98yjuEZ0fy+s4jvN+XhcCtEmnSogNmrD/ocOyKjYzc9leZX89nja4JyrTYNFGqjp2Oyz4Fwqk1Oqk1O0opMGAsOYO3WD71KxalhYWx7xsaAK0zUDnTjjmrLGrcfAxvIOBtsNXN/wHTGeQDULlngtz9EZ+L/JnlySRl1Bm4dcwWPLXiViIgIxo0YAkBMTAz9+vVj48aNRIT4/vFVKSpPkPiTdeUXjBj+Fywqhe37lrKEldze1TOjfPR4PZGZT6NJTEZVZee5kmpedz3HPP39pKlLOKliA1sWTGXg5Hn1vpc3coq5IyPxuNO3i/YpIcLIA2f19SlTFIUNByvYdciTyFKrVnFKn3gSwo0ATOqTwKQ+CT7nPHBWXz5alc3OgipOyoznlN7xrfMGWkBpXjWfPeEfJGbNfYjiyjwenP7uMV2vTf9iHCUOyn4phZRYVArszDDz+/BiTv08gddmP8k/4xO4IvrPD5ZPPf+N37+7LavcblyWfYDPMroCMLO05JjPv2roNE7tMZbvts/nu+2/NXhc1TOPYP3qY1SmEBw7tvjtt7vs5KhL2K0pINEdRX6V55uc3e47nLW62tMUtb9wO+uyFjAiczIAFRbfuk8z9mCw1bNOxW5XHLvsvhP/knXFFJGMdq8ZlUthhHo3aeq6a8RrbZT9/TpiXq7/D+G+XTm8M6DpI2VEx6RSqRieEc3wY3gqCDVoufnkHi1Yq5ZXU21n+Vd72LXKN6vu7rxNzF78HKVmz9D07KKdJESlNfm6bdZHoc3sS+QDT6LN6M6ENfMZu2ERAG+frKFmy3m4NDr+s/cVLhu4F12Im+p8AznLY5heWc2OwkKfju4TlRaI1mgpdjX93yJsQBhdr8+AKBX35F/FlMoxjHr9YgrMRQ2e83hiEhdHRVHsdPFhxiUMzZxMqfkQT35xHW7FyaOP3opKZcJiVnAX5RFvqeTtJWswa/Rcf/31mM1m3nrrLczmujTl0WEJlFfX3VMblYTbXsMrf/2GfkY1mUYNbkVh+P+m8fjzJnr1MnqPvWTzvzB2GYB+QynhWNli9M22q3qsCk1aBoYJp+LYshHjaWcTcu7F3v37Jw7yrjYoRGfx69tb2bPO/+/4w4VPsybL98vg+aNvZEK/c9p/H0XEXf9Cm9EdgGWjpngDha3wVIhMQgP0HKFDp/E0S4Sl2Fg0dRIXRno6td955x1yc3Pru/QJwwnHFCQAut7b1fv6+ZQPmVI5hkADSnvq9VwcFQWAPqYnQ/98EogNTyI1tgcTzyhm7Lg5bGIor6o8fQ8h1mpu42nu/WKOd0DC0Y4MEhFjL2XgyWdyunotaw/9zPlR5+PAyYqITUy6ZBxp3bf7njvjBlCp0EZ3If68+4lZnMbJhl08PNHA7b94ZpinHMpn8tzvcSoKK60WjhwIPLuglKtT4tDJUFrRwZnLatm+LJ91Px+od/97859k/d7f/cp/3TibCf3OafJ92iRQjOh5KpN2RWMrMbOup4G9yXrvvppsMCTB4Mo/2F+jQxveheExedSojBRG1o18Gjhw4AkfKI6Hs9qJNsz3x1511HyKI1W46pp+9PYqn32W2ir+epVn1byPudZbbg0Jo8RsoalGTryKj1U70CoZRHV5hyqXkQ8j1byT+C2kwr8KQnkpzQrAwgV/1lVRcJblUvD+nQB8B3y30xM0dcDs9AzitH++z0VzmXTpVd6Mtg9m5fFKdiGbx/WXWd2iQ/tl5haKD/ovKPbFsldYsu37Bs+z2a2UVh1qcP/RWv3526QP5YpT7iPDFkKvfAeXL6mm6vnHefT737j3izkYknoSbS9jYtlyCmoiWFLUnVJbCKHU+lzn4MGDDdxBBLJrxi7MWzy/WOVvlJL234lU260NHl/icnHJgQMAZJXnMuuHuzyv8zdTZS1j8WLPB/c0fHP7f7Huj3qvl5qqIyWlLlDFxsZyum4LM/XV/KIJp8D+NrOKfuOd5CU+502ZvI+LL87mP/9puInssDCNpi5I/El91PDhIruTlRVND2ZCtDV7rZM964vIWltIrcXBjhX5fkGiylrOPe+eEzBIACgofLrk+YDHHKnV+yiM+lCeu/YHn7I73jzNM5MSMHYfwaDzbuHS/G+8+2Mcq3hkQRk5VQqnn3465eXlbN68GZvt6KxEoi2MPq0nsUNOYv36Q9hOOxvzrBdwHfCf6zJtWgS33+GZwLR1ay13/T2fCy+8kIEDB3qPqcw4i0+7ep4wTeZ5hJV/hNvpZvv12/2uF8jdcfHcEFu3/sakNz6t9zhZ20K0d067iy2L8ljxjf9ciMPMNRW88uM9FJQfOObrt9sJdxeMuZnJg+vWR7h91mTfSmkNnDNqBCenRgHw8HfzsNpbZlKNaCaVmvQZX6PSej5sS+e9TvXGn+s9dP6C7j7bUybv492re3Nu11q20YtFjOGNk33XNa6+dQKWnZ5v/rFTY4k+ORrLdguHPjuE4vL/1dUBh39TVBFRTH3kv2wePAKlgRX8zoqL5N2BMgpKtC+WShsOmwvFrfDJo6sbPf6Jz6+lsOL4WlnabWf2t6tmsjNvPbed9TSLt37rt19x2vhxxXLm67TUOGR0U3um0hm8QQJAF53S4LHvvVvGtdd51r622xVGpqi5tqtnuN7JrOb1pZXUEuFZAhawrVvpDRK6OB3Jl3um1hm7GDFvNlO91bdv5f8SErksKoqtNhvXHMzGVlXBurhEtEcFCcuXHxN6yZUA/FxSydIyMyfFhDfnn0GIoCg+aOarp9fhdjf9+/uTn1933EGiqdokUKTH96ZXyhD+N+d+duY2nLdFgkT7p9hrqP7jN8IGnQZAzb6Gf56zZ1eweo2VM88M55PZFQwI8+1IXrFuLZUL11AzegK6vgOxfvtZ3c6j+pwVp+8fUpha7Z1zM9ho5LqYGN6prGBURT7akmy29BlOjcmTBK36jee9gQLgks17+XpID8ZHS7AQrc9aZWfT/INsX5aPzXpsn3k/rnmHQxUtn0G51ZuedFoDL/6trmmioeFbov0alNSbhybdRkpEAnfP+Q9rcv9AG5OK21qBu7bhEVT1efQUA4+c7FkoKfypKqrtDR8bMymGlKs9Tyw779qJs6LujypCrWbVEUkGv6ioYEVcOFP715U9e/OT2NatpOL+W9EPHUX087N8rv9CnzQuS4rBoSjssdqI0mroYtQjRHNVl9s4tK+SFV/vwW5zYrN4fnfVGhXueppQAdyKm3VZnmwIOSVZLNryDZdMuIOJ/c8HPP0Sj336V2odDQ9GaYp22UcRbormqavqRsh8veJ1ft/y9fFUQbSRlTd/QWpkknc77b8Tm3W9UB1YmtgFpTaocdvqT4C4PrMXpj+bmWbk53HSlHEYjkjX8bAqlpofvkKp8fxh6YePJvrZmQHvNzDMxK8jeqGWYbTiOG2cd5AV3+7xXYylETPnPsj2g6tx17PMcXpcLyJCYtiVtwGHK8A3qyZql30U5ppyVu+ax+jengVCrJoizjzzTLKystizp+FefdF+HA4SbrWNirTfuf76GH78sYrCwuNrKmxqkAAaDBIAw7N2c1FkJMVOJ6utVtbPX8Y/zjwFgFKLFeucOT7H29evpuyu64l56e16ruaxpbqGD/JLubZL50g3LVrftmV5AYNESVUBDqeNjfsWs2Lnz35pbY52sKT10xi1WQqPhMhUalxmZtxzl7ds9uzZEiw6gITQWFbc/DnWbosp6jvbW94e07+rgKTIcAoq/SclHabtnknUky+hSfI0a6msTrTbK9CU2lC0KuzD49h33ghJ+yGOy8zbF+Fyer7gLN76HRv2LiLcFEW3xP5s2LeI7KKWyfzcVO3yieKwospcov5MDXGYwWBom8qIY1JkKaXn81N45ZUu9KN9/8wUCBgkAJz7snAVHfIGCk2uBU2pZ46OyqlgWF3MO8OKuT0jMdBlhPCzYV62N0hU1pTy5fJXvfs27V/aVtU6Zm36FamiooKNG72LXnLoUNOnlIu298gjdT8vi6VlFk5qLZWP3ed97db69kcoejVP7ivAcQxDFoXYvCCHld/s9W5XVB97luf2os1XuAOIioqiqqoKdwut0iZaVmqqjtzczjEhUpOWgbuggNABk4k9/TYA7ENicCea+Fe3ZP7eVZ4qROOcdhez7lzsU/bIJ1d603y3J01pemoXja4VFRUSJDqwzhAk+sapuWGYjsSKgyhOO9WbfiH7v+dQrtqEO9GzSNSnBaWNXEUIOLSv0i9IPDT7L+0ySDSVLPUlTngp4Sq23xbm3Y57xkxpjedBu/qdVzFNPRuAA7V2iu0OyQ0l/Gz4NZt9m4qpLK6httr3i9PP6z7wSavfEbWLJwoh2tK4NI3Pdnyop49Cn6BHVVOC61C+d99z+6UfTfj69PHVrPx2L4X7q/yCRKn5EPM3f9FGNQseeaIQJ7xvd/jO/yi2KCRcmEDCeZ41lSv3vY496UkAPsgv5d5uSfJUIVDcCl8/u56yfN909S63ixpbNR/+/jQ7ctaiHMtMu3ZKAoU44bkUz/KpZ/TU8vt+JzYXDPgzSABEpmZTfMTxM3bm8NGg7v4XEieUL55aS0lOXcoaq83Mgx9fht1ZG+CsjkkChRB/mrun7smiems1YQM8/RaWPXZIrzvut9IqrC43ITIB74TidrnJ2VlObbWD3z/a6Z0fcVhnDRIggUKIeh147gCRYyIJ7R1K0fdFGEs+JPTSq7z7q50uCRQnCEVR2LOuiHnvbGvwmCMXX+uM5DddiHoMNhp5KiecW+a6CTMrVM98kdplC73711TKMqonih9e3tRgkLDUVnHPu+d06iAB8kQhRL0eSEhkoMkzfyJMo+YfRWX0qNRRYndj06t57WAR5yREtW0lRYtb8+M+cneW+5Tll+5j2Y6fsNRWsSV7ZadtbjqSBAoh6nE4SACcFhGF5dQXiaqOg28reHNqBNUhrjasnWgtu9cW+pXtLdzCkm3ft0Ft2s4JESiGZXThtH6ZmGttvL98nay/LRpld7vR/7m2xS5NFFGhdWnG04udVHRpq5qJ1pKzvYzKohq/8n5pI9ugNm2r0/dRqFUqLh89hPjwULrHx3D2oD5tXSXRAYzM2s2cqiqqXC4e3OO7vGtBjIYsq62NaiZaw9w3t/LDK5vq3WeuqWzdyrQDnf6J4ujJLjXyNCGawAHcV1A3I/v2WZM588ZXKD11DAXRnf7P5oRWuL+KvRv8U24UVeZSY6vmpzXvtUGt2lan/41XFHh1wQrumDwOgI0H8xs5Q4j6zZ39AAmXedYQyAxp3+twiOO38KMdPtuv/HQvu/M2NnD0iaFTBQqTzsjkHmMxag0s3r+GYksZANml5fzjq59xuzvDZHrRVhRLNW5LNerQMLKsNtyKImtpdzKH9lf6pOT435z7T/ggAZ0sULx94b+Z2LWuo2nim5ezvzwXAJcsOiOC7Oz1WfwyoldbV0ME0Z51dU1O5dVF7MxdH+DoE0en6swelTrIZzsztmvbVER0Ws4DdeuCbzRb+cNsbcPaiGCzWev6ML9e8UYb1qR96VSB4sVl7/ts7y072DYVEZ1W+d+v9dk+a/3uNqqJCLbSvGp2rqxLI19cmduGtWlf2sVSqMGkUWk4vdcE1uVupdhS1ilS/LaEESNGkJGRwR9//EFWVlZbV6dD0Y+dSPS/X/Zuzx3eiyERIW1YI3G83C43igIludV89fQ6n333vXceNfbOn6qlKUuhdrpAIRqXmJjIzTffDIDT6eSZZ57B4ZBhw02mUhH35W9oYmIBzwiopaP7tnGlxLHK2VHGvHe2+S02BPDBwqdYmzW/DWrV+jrMmtmidR35S6HVagkPD2/D2nRAikLVvx/wbmZZbbyS7Z/qQbRvfyzMqTdIvDf/yRMmSDRVpxr11N6NHm3i7HMi2L3bxuyPKzi+Z7nmMx2RxwhAo9E0cKRoiH3jGtzlZaijYwD4z74CdCoVt6QnNHKmaA8slTYObCn1KauylvPBwv+wK29DG9Wq/eqQgcI49Vwi7n4AlcFIxUN3Y1u+qK2r1Ci1Gh74v0Q0BiMJ/VIoKlrPvF/bJhXArl27fLarq6sbOFIEUnrz5cR/Pte7/djefNZVWZieFMOkmAh0aplj0d4oisIvM7ewf3OJT/nMX/6PbQdXS59mAzpkH0XiwroJMIrbTdGU4W1Sj2NhNKr44rt+3LvkMWwuIwAHX7gIxdE2OYO0Wi0XX3wxGzduZPfu3Rznr8EJT5PchbjZP9W77+LEaOL0Wq7tEkeGSWZyt7X8PRV8+5z/08KSbd/zxbJX2qBG7UNT+ig65BPFkZwH9mLUwtNTDJzeQ8srq+28sa79dczW1ipsKh7gDRIAmvA4nGV5AKg0Ks+3mVbKXu10Ovnss89a52admKsgj+LLzyb+kzl++74q9KxjMDOnmC4GHafEhHN9ajxqlQoVeP5Tef6fatRjUEuXYUvJWltY7+JDny19iWXbf2yDGnUsHfKJQmU0EfXUq+gHD6fstqu4TLeTDy+oa3dXPVbVJvVqTFRcCJF/+8K7nfPSdNw2C12u60L0xGgAnJVOdv59Z1tVURwnVXgEhjEnYTr3YvQDhhzz+XE6Dd8Ny6RniLHxg8UxURSF12/53aeswlLC459dfUIsOtSYTvtEodTWUH739d7t1An6NqxN01WUWKl45jxiTr8Ny7bfcdtr0EZqvUECQBup9XzFlJagDkUxV1H72xxqf5uDKiIKdVgYhrETMZ19Edqu3Rs9v8Th4tvCcu7rltwKtT2x7FhR4LO9bPtPfLb0xTaqTcfUIQPF0Z5ZbmdIkoZL++s4UNHO165V3JTNfdW76TS7cdW60BiPGHkkQaJDU6oqcFVVYP36E6xff4IqLJyQaZehTkxChQrUKk+bEypMZ5znPU/SkbWM/N0VvgUyxuCYdYpA4VJg+lc1XPd9DZb21z0RmBt237ebyNGRpFyZQtGP/nnwRcemVJuxfPxWvfuMp5yGyuhpNj07XiawBpPT7mLfpmJ2rT7kU+52O9uoRh1XpwgUh3W4IPEnl9lF2fwyyuaXtXVV2o3zIyLoazDyfVUlO2ydeDU5dd2TpPRPBI+9xslbdy+pd19xpaxJc6xkmIVod4abTDyVnMJVMTF83bVb524pUOqaSr/5c5SUaL4ti+tP6Pf96rdYvsN/hJoIrFM9UYjOob/R95t1Z+7bt2/ZhGHEGAD21XTiJ6dWVlHon/79odl/obxamnaPxwkdKFTA2J4ZdImKZM3+HLJLy+mfkMmVQ89nf1kO763/Goe0Z7a6T8vL+VtMLPFaz69nOx+e0CyWj9/2BgrRPPYaJ2t+2k/5IQtF2WaffWuy5kuQaIYTOlD0TIjlwmEDABjdPY17v5jD11e8SqjekzL6lO6jufzzGW1ZxROSAzh57x666vQccNjbujqig9i+PJ/NC3L8yl/+8R6y8je1foU6kRO6jyI5yneSiQq8QQIgRCedi23pRAsS/ztYRH7tifWeg6WqtIblX+3xK3e5XeSX7qvnDHEsTuhAsSzrAIWVdY+oCnDzdw97t9/f8G0b1EqcWOp6X1SKws2ffMWjjz7KvHnz2rBOHUtZvoWP/m9lvfu+WPYyFlv7zNTQkXTIFB7BFh1iotxa490O04egVWuoqDUHOEuI5gu75R5CL7kSgLTSQs7eWveBd80119C1a9c2qlnHYK918tZd9Q+D/ej3Z1i9+9dWrlHHIwsXNdGRQQKg2m6VICFaXPhd/+cNEgBqxbfbftGiRa1co47DXuNk7Zz99QaJGruF+Zs/lyARRCd0Z7Y4cQ0damT6ZVEUFTmZ+UYpVmvrDsDVDR1JyHkX+5QZUtNx7dqAxumZOXrWWWe1ap06ksWf7WL3at9VBQ8U7eC5b29voxp1bhIoAtCkZaCOjsO5PwvFLO2cncmzz6V4X0dFaXj4odZZylQVEYW2Ww9inn/Tp3zHhAFE67Qwtn+r1KOjUhSFOa//QfZRq9Mt3zGHn9d90Ea16vwkUDTAdMFfiLjjflQVNtSlNqp+eJvqHz8ixqTiykE6Sq0KX2xz4OjMg/xPEO6W/hnq9Bgnn4l+8HBMp5/rt/unYZmeICECstU4+f2jnX5B4h/vXyAd1i1MfjsbYBg9HmpdGFZ7lkyM7Ted2iXzeO/MKs7rrQPgP5MNZLwky4h2RPffX8Azz3hSev/0Ywt+yKjVxL79Odq0rvXu7m4yMDwipN59oo7T7uLtenI3vfzjPRIkWoEEigbU/v4rpu6+S6yqTeGc19vT8V2mVpOfaEClsaC4OleCCRUw1GTikMNBvrNzzkzfsL6Gs8/aj9sNDkcL/fxUKuK/XoA6Mspvl2dd7XDOTYhCperU2ayazeVwM+vOxb5lbhcvfHcH2cW7GjhLBJMMjw1AFRlNl+tnodF6vvHlvnYVF6RX8b/Lwjk7NRmbWo11j5V9T3aeCT1Dup3E1d1H0kdlJ6VgJR8UHuSFkuK2rlaHox87keh/v+xX/vngHgyPCCFMq6nnLHE0W43T70kiK38zs+Y+SK3DP5+TOHZNGR4rgaIJ9Mm9sBfuBbdnQevE8ZHE35Dm3b/j9h24qltpsesWdO7QMZzSpx9qXTpa4wRUKhUjFtzMmD1ZbV21jkOtIfz2+wiZNt2nOFKrYceEAajl6aHJXA43M+9Y5FO2Ye8i3p3/RNtUqJOSeRRBYi/Y7Q0SAHat7z+b4vTE2ghDGKdnTiDa1PECqEmv4+TMWBRXIa7atbgdewFYUH1i9cF07dqVxMTE4zpXP2w0ifPX+QWJNKOebeMlSByro4PE5v3LJUi0EemjOA7lS8tJujQJTain+UBxKOjUWpbf9BlRJk9knvrutewo3tuW1TwmITqdz7bbVYC9Opz/FLXOsNH24IwzzmDylB4kJOwnK8vNk0/MbfK5+mGjiX5upl/5z8MzGRYRGsxqdnq2GieLP/Hte9i0fylvz3u0bSokJFAcFzfsuG0HxnQjtQdrAegel+YNEgDDUvp3qEBRZvVt7/1u9bcs3rU5aNdXRUShWKrB1X47x8eMGcmA/l+h09vo0gV+/NHI5k21Ac9RRUQS9ehz6IeM8Cm/LCmG/+uRTLxe18CZoj4ul9uvT6KsukiCRBuTpqdmOBwkAHaV7PfZt7Vwd2tXp1kUBR789leKzdWsO5DL+gM7gnbtsBvuJOG730n8bS2Gk6cE7brBtmrV7+j0dYsHTZ0aHvB4VXgECd8t8gsSXwzuwUt90yVIHIec7f7LAX+w4D9tUBNxJOnMDrLpA8/ij0O72F1yAJfS8Tu4gyFx4Ubva2fuQUqvOr8Na9MwrRbm/trdu/32W6V89lllvceqo2OJ/3q+X/l3Q3syJiqsxerY2c19cyt7N3gWGCqqzOXfX/wNlywe1qKa0pktTU9B9vmWnwntfyrJlz6KyhDCoQ9n4CjJbrH7pep0HHI4aM6fUs8YNe+eZ+SkDC3/nF/Lf5e33JoIji0bWuzazeV0wh3vWzBNiUOvgoItDS9NGna9b06hq1JieaZ3WgNHi6bIWlfoDRIAv26YLUGinZCmpxYQd84MNGHRqHUG4s5puRXy3uiSyrzuPfijdx8GGI9/kaXHTzFwUobnO8PTUxq/TmbyYEZmTsGgMzXp+oVThmP98StsKxdT/eGbjZ/QRlR6FZrJ8dgVFdVuFbbBUfUepx8+GtOZ07zb05NiJEgEwcZ5B322t2avaqOaiKPJE0ULc1aVtMh1I9VqTg6ra+K4MDKSrbWBO14boj2GrwsDM8Zx0xl1QxRvnzW58ZPcbswv/vs4ata6jp5h7yz3/zarGzKC6Gd9Rze90EeCRHM5HS5qLQ7v9vPf3SGpOdoReaJoAbmvX+t9Xbnqi6BfX6NWYwoP5ciPtY01NQ0e35grv63hvY2e5qZHFzXc3AIwsf95x32fds8FWf9XN7mwaqP/B1XE3Q/6bH81pAcamR/RLKt/2Mdbdy3BXFr3Reeeaa9i0ktfT3shndktSgUEN4+QQavlX2edQpjRAIB7+SZyam18WF5GZTPToDaltrHhyTx2+cfe7SY9UXQ0Af4h4r9ZiDoqGvAEiQnRgUdGicAslTbe/8fyevd99Pt/Wb1bloRtadKZ3eaCn2yuf0qiN0gAPFtVQbHZEpRrN6W2peYC7pg1hYFdx/HHgfr/wOtz2aCzuWTgmfy+dxVvrP60fY8IC/QPofY8hKcZdRIkgmD5V3sa3KdRy8dTeyE/iXbswgsvZODAgbhcLp599lkcIWHsNEWg4PnSC2CxtdwIpYYoKMcUJFLCE3j2zH8AMCp1EJsKdrAse31LVe+YjB5t4u4Z8dhsCk8+UUhWVsP/nvoxJ6GO8DxF61XSahsMlor6mzptjhp25K5r5dqIhkigaKcSExMZOHAgABqNhtRTp1J93+MAPAec/fANrNmfQ62j/Q8fNGj1Ptvu42vtDAptr34YRo1HqbFQu+g3rro1jrg4z4fVy6+kcNaZB/zOUUfHEnrl9YRccJm3LEon2V+DYfCpaeRnVXi35238hA37llBcmYvNcfz9biK4JFC0U2az2We7eqpvJ/K7yzrOt6395bn8d/Gb/OPkG73brUqrJeymuzAMH4O2aw9vcfht99Fbuci7rderifr3y7grK6h+73Xc5Z6V1CL+9SSGEWN8LvnvzNTWqXsn131oPL1GJbJ7jSenWL/00fyw5p02rpU4mgSKdspqtfLaa69x2223sWfPHqp3/ELY4BGNn9hO/W/Vx3y86Xsqas2NHxxkhtETCL3oinr3zeE8zuYHAMqJxjB2IgCmMxoe3fXTsEyGyKp0jaoospK3qxyAyHgTSd0jObClFJvV4XOc5ojx2amxPYgNT6LUfKhV6yoCk0DRjpWUlPDYY495t2sumkz4bfe160lrgbRFkABQx8SB040m1woouLqEgk5NiMtKYi5YavuwNyWKWWF/a/Rae08aSKgsOtSobUvzWDT7+FafCzVESKBoZyRQdCDu8jIqn/xXW1ejTamNaty1xzgMWFHQ7jWjPeBZW0O3q4pHbh3GJZ+eRFjNn6v3HYK+seXc0ecBnPpwBof7PjHE63Xc1TVRgsRRrFV2rFV28naVs3FeNpbK5g+ucCvNG+Ytgk8ChWjXwtVqRoaEsKGmhqi70wgbEEbNH7UYvoyhsOJgk7552lYvQ9vvGhJthaiAQ4ZEav/4si5I/OmM0uVkRe2CkY0/WQjYsSKfRR/vQqMoxGlVRAAROhUVLjeVmipUihqtMwy3y4VG0/SPGgkU7Y8ECtFuaYFfu/cgSuP5Fn9/nIYDwGXqf5ByThqR3Zeyr3wBb7y3joKChkd/uYsL6Tr/Mc7t6Vm5zqHWcsu63+s/2Oqf5lpA8UEzmxfkUFZgofigmYwBseRvK6WXQU1vo+/HyDrNXg7qDgCwaP4yFi9fQIghHL3WwJNXfu5zrEanxuXwDQyVlpZJeyOOnwQK0W6l6/VEajSURUQSXVVJr1yFfXHhJJm7EZ6xkqRhn5IEjJuczoXzr8VdXoZh5DgUpxPbst+xfPouOJ1EG+G8XnHUaF0Y7Rp0gTKSdh3fau+voyg+aOaL/6z1KcveWko/o5pMo39T3D5N3aqI8cmeWexWmxn1UXNPLntoFNHJobxxa13Q3rB3keR4aockUIh2I02nY4DRyDKLBbPbzT67nXv+/n9s7N0fdakN/foSyFLxcqSVJ5O3+pxrOu0cn21dZh9Cp1/NzW9fwq3Darg85QBFWs+v+zU/Z9RfAVMMZIxrkffWkf088496y3UN5LhyquqeEDIzM+uO19ZlFAiPNRLbJQx7rW/Q/mal/3Kyou3J9FLRLqRodfzavQfPp3RhdWYvdCoVmozubOwzAM2BavTrSzk8H92hUvPQgasavaYqJITHBlfzeUSYN0gAnDHkYP0n9J8WhHfSuSz5fDfVZXWzp6usZewv3M6Xy//HLZ/dwuL9nieN7Ip8Dpk9fT5WVd3xen3dZMuLxt3qfZ2QXn/6k9TYHvWWi7YlTxSiXZgYFuqz3S2mG+cOm0HXz0pZFlXBKgxwxDdYe63CXws+IjG5jAJV3eS3ixKjmVNcQa3bM/u7ShPKKdYa3oqqS2C5Mm44/fOOWEyq15kQlwnj7mihd9cxbZiXzZbf6yZHmmvKeeCjS3yOufKLe3y2o02RnHvheXTr6Vkp0H1EosqEyLqfkzHMs0ysWqNCpVah/PnzuvnMf/PNijdYuOWr4L4Z0SwSKES78F1lJQ8nJoFGjzZ5CNcPGospOhO7+VNGVhxiJPBB6hVU6eqyXOq3lDN+6EDu7prI6koLU2IieHhPnjdIAPQe/xOP7HuDRyp38liUk6roy7nwtOvh90RQa6D/BZDa+ETGTVVWvi0sp9DuIM2op3+YiXMTojp1ivFN83N8th//7OpGzymvqeTD2R8xYsQIRo0axddff+3d53TVDZ11/7n2h1anod+EFLYtyfPumzrsLxIo2hlJMy7alZ8ueZrB3T39BEV2G7/nveTdtzhmAn9EDvQ5PvzcDAaGmbi3WxIvHShkXmnjHaHTk2J4uW96k+tUbHcwbMV2HPX8qQwKN5EZYuTxnl2I1Xee712WChvv/7Mu8eO/PrwYc015s645uNsEbpjqmUBqCNGS1D2SkWd3Izo5hLfuWuI9zmqr4v73L2jWvY6XSqVi/PjxxMTEsHbtWgoKCtqkHq1J0oyLDudwkABI0Bt89u3fvYkKxzbCBp+GNjQWR/8oiu1OFpaZWVjW9FnfG6uOLS37Hqut3iAB8Ie5hj/MNXxdWM458ZG8PaDbMV27PaqptvsEiZySrGYHCYA9BXWd4jark+ytpWRvLfU7Lrtod7Pv1VT9+vVj8uTJmM1mvvnmG1JSUpg82bPGytChQ6murmbLli3Mm3dir4shndlNFBenYdoFEfTrZ2j8YHHcHl31Ht+m6lgSr+G7nQv451dz2VlQxE+bd5C15Csql35M3mvXUGHYjjOhdb7n1PdH4q6x+pX9VFzJWet3t2l23OayVNp4995lPmVLt/0QlGu73Y2vQeJ2u3j9538G5X5NceaZZxITE0NGRgannHKKXytJWFgYY8eObfQbd2cnTxRN9OxzyaSleUZw3HpLLrt3t/46ECeCr08fyk/9jQDYDd1xfu/i7aW+Y/hRFCofvx8A/chxmM66AJXBgGHMSU26h/oY+xUyTL5fDsr+fh2OLRsBMJ13CRF3PeDdt6HKyuhVO/h+aE+MGjUxuo7xJ6a4FRZ/uottS/N9yn9a+x4rdv7c7OuHGML5918DLwu8YPOX/LjmHZQWWPCrIWFHrDs/dOhQcnPrz2ys0ZzYqVs6xm9xGzMaVd4gAXDlldE8/HBhgDM6PmO6EUeFA1dV665Ed+SHvb7/oEaPt69dgX3tCtBoSZizHJVe3+g5Z8QdW99akkHHkPAQNpmtaLKrSTrtAZzDcin68lE+NPzExYt/pVpjYtDYb7FqTOTU2hm2crv3/LnDe7XrbLOKW+H1W/1nqv+y/iPmbvi4njOO3bg+Z6PTNPyzqXVY+XZV68+hsFgshIbWjbhLTa0/ffzUqVPp1q0bGzdu5Ndff22t6rUb7abpKVGr5broGCaFBm9B9W6J/Th96OUkR3dt1nXsdt9vOEXF7X+xoOZIuSaFno/3pO8rfYk/L75V71310n+8r2t++a7J56kMhoBBIjPEwC/DezFnWCb/6JZ0zPWaEO35vdTuqkStM6FPyuT0Cy7i4n6eYZ5hrhoGmetvWz9j/W4u27SX4xw30uJ+n73Tr+yblTOZs+79oN0jMjQ24H6jrm0CqcvVtC9Cffr0wWAwMGbMGOLi4lq4Vu1PuwkUL6d04d6EBF5LTeW0sOavRRwZEsc9017l3FF/4/8ufYfY8GP/cDjM7YZHH6lLPvf7wupm1+8wFSp07Wxt4JhTYryv486IQ6/VMG1of+49fSIju7bsgj01P3xJ0QWnUnrTX6h69rHGTziskeakUoeToREhDI8MRdWMIa2qIz7riyL7++zbdeftVH/4JrULf0Wp9V2dbVG5mUs27qG2Nh+lHSW9s1Ta2LG8bmRPTkkWd7x5Ggv/+DKo9zkyfYe7HY0orqo69nQht912m0+T1Ymg3QSKQSaT9/WM+OZ/i02O9k3TEGJoXvBZtszKuefs54zT97FtW/3r/B6rgYm9WH3rl+y7byE5/1jCtL5TgnLd5rLsqhsVZN5kZnhGFyZkdiUpMpzpowa3+P2VynKcWf7fcgOeY6mmdlld84liq6X8/ltwlXlG1ZQ5XGw4xtFO9XGH1QX1vbt3M2Sm50vDG6VjmXbd/Vye2hPju69SdNY4yu+92XusWnFxWsUdLF9xEgt/z8Tl8u8Mb20uh5v3/+G79vlz397eIoHM7qwFYM6IEP5zcTSvj3XjOuLTZ0fO2gbObFnffPNNveUmk4kRIxqeXzNq1KiWqlK71GZfZUN1cO1QHQaNinc3+nYM5zkcDZzVdDvz1lNhKSEq1POYWGn1H4Z3rGpqgtt0cMnAM0kOT/Buv3rew3y3Y35Q73E89j+1n4iRERiSDZQtLGNw98zGT2oHKh+egTkxGVQq3Ic8nbKKxQwxnmaPB3bn0S/MSI3LzelxkUxL9CSsW15u5vNDZQwKD+GalDi0av+vvEkGTxOTfXwi6jwLKpub6h/XsLnQjeaJau5/cDyf24bgRo3xr7NQvXgp9g2rKTxzLIm/rCSTXfQky3u99RsuY9TI4IwmOlYup5s/fs9l/S8HfMq/WvEarkAJE5th+Y45jBv9Fzb08AxUKE2P46NNizmnMI680r28Ne+RFrlvY8rLy1m8eDEnn3yyT/mUKVMYPnw4J510EsXFntQkixcvJifHMwkxJKT99jm1hKAHCp1KBYpCYx/1/5xgYNLpkbwcHcWUa2D6k3l8HtIFgPfLgpPq+cGPp5OR0Ifckj0t9gfQHNsKs3y2txftaaOa+KtaW/dIPnfrbpIiw+mbnMChyrZZpa6p3IW+E6Sq3/4fUY89B8Ams5VNZs83+W+LKrh5e7bPsV8cKufBrDxuSotnWkI0apUnu5QKeDCrbuawu4un81PbuzeO7Rtxu93kuSNxH/GAro2Ix1GSjem0swHIw7fJzoERi9PVJgsh7V5TyIqv/X/X1uz+rcXuWVKVz7Pf34lm2kfesp2Wfaz79K4Wu2dTLVq0CJvNxtSpU71lP/74I2lpaSQkJHiHzIaGhjJr1iyAdtvf1FKCGiimhIXxbHIKBrWat0pLebGkuMFjHzhJz+Ckum/T9juSGXz/TlQqFfYg/hCyi46tCaM1fb7lZ7Ir8rhr/LUMTOrFvT8/3dZVqpfL7eadpWsxaDXYnK07Cqq5HDvqz3wayKycYmblNPy7C+CuKMe2YrF3+4sXHiTpjk+92y5LOWpTKIMunYGlwklxVAQvKvdxEZ+zlUH8aj6bkqVbuDI5lgd6JLfaMFpbjZMVX+/B7apEcZei1qajUnnubdKHYbW13BeBkuytJB6xbf9jQ4vd61itXLmStLQ0+vbt6y0rKSkhIaHuM+pECw5HCupv5ytd6r413RAbGzBQDHzDgvrZum2X1eV5CjnBfhircjZz2Wd3tXU1mqSjBQkAd0kx5rdexXjq6ah0OlQ6PZrkLsd9PVdBHmUzbsBdWgzOuqdUm9VM9n/PJWriVVh3LcNdW80Vpz3A2OW1QC3b0vR8M24M6xjjc72PC0r5uKCUtWP7kWZsfGhvc815bTPWqnzsVXXf7HWRf2fexk8oNbd8uorCU4diuuAyXHk52De2Tb9EQ7744guuvvpqunbt2uixJ1rQCGqgWGapZkITh7duL3Zj+FcWmU952r+Lvi0KZlWE8LJ++i7WT9/1bqsiogg5/1JQq8HpQBUSSs2cb1HstZjOugB1eKRnnwpQqT0jqlQqnNn7qPn+S3A11IypULHkA+/W2G6TvK/759ipv9vUY+TK7awf248uLRgsFLdCwZ5KXLbtPuV/f2tqA2e0jJpvP2vV+zWV0Wj0CRJHT7JbvXq197UEima4MTeXG2Ji+FtMLDc3MMPxSLYCG1uv3YpKo0Jxnlj/8KLtKFUVWD56s959lg/rLz8ev2/5mkkDLwJgd4qnM7x24VzP6Cy7Hf3Ykwg5+0Lv8cNXbufgyYPQq1t2MKLWOByXbX2L3qOjiYuL48Ybb/Qp69atLm/X7t272bx5s3fbam37UWutKegNo2+VlfHWsXRGK0iQEJ3S1yteZ9XOuXSJ7cHmwvXYnqwGR90IP9uKRbgOHiD8lhnesjWVFiZEN38eUSAqdRiGqLsoLf2FT1bMadF7dQTx8fHceuutPmWnnXaad9Elt9vNJ5984rN/w4b207/SGtrNPArReaRHJvutj3yiyivbx5qs37BVlfkEicOsX36Ebe0K7/bFm/byQ1FFi9RFpVZhCPV8N1Sp1MTFnc2d571Oz+TGU6V0Zl26+PdZjRw50vv6zTd9nzJnzpyJ2dy+R/8Fm/w1i6D67LKXWH7z52Tfv4jM2AbWphY+bCuX+GzfuO0Al2/ey5N781lYWhXU9vC/PjHWr+yu814kMSotaPfoaNRHNPXFx8fz0EMPeZ8mLBYLhw7VZWXIzc2lsLBz53mrjwQKETTJ4QmMzxjm3T6/X/uYad7e1f42B+v3vplVF5aZ+d/BIi7/Yx9rKps/o/wwQ4iOW9+YxJRr+/mUD8jwDyAniiNTuhQXF/ts2+2+T4Hvvfdeq9WrPZFAIYKmwOw7cm1T/o42qsnxGTRoEHfffTf3338/06dP9/mm2ZIUSzXml5+i9Ibp9e4/f2PTJmLut9pYUFqFq5EnEJVKRe/RSYw8p66ztr6mwpHj4/nf+0OZv6A7EyeG+u1vz2JjY0lMTCQhIYH4+Hi02oa7Y/fv3++zvWrVKu/ro8976KGHGDZsGCea9pWNTnR43Z6dxBvnP8bWwt2syT32yW5tJT09nQsuqFt+s0+fPowePZqVK1e2Wh2ce3dTeOZYdD16oRs8gvAb7vDueyQrj39F2zFufA8ctTD1CdDWrZMxr6SSq7bUfeAdmjSk0fvFpdYNZY+P8G2nT+03ldKTbuPZHRo0O528+cgMpkze1+C1encZRlxEChv2LqLGHrykmcciLi6OzMxMTj75ZAwG/wXG3nnnHZ/1JtLS0khL829y27hxI+PGeVZaDA8PJzQ0FIul7qnu3HPPpbKykr1797bAu2ic2qAmYmQE2nAtlt0WavbWNH5SM0mgEEHldLu44dsH27oax6xXr15+ZfFBSE55zGy1OLb/gWPHFkIvuxp1uGdltVk5RVz63d/ob/F8OCnWElQX180N+e2otcIXlFYxOTbwqmyqI3Jajet7Ft+umkmYMYpLJ9zJ5r4DyfkzOaBL0fql2j+se2J//jrpH8RHegLNqF6n8eL3fz/GN918PXr04Morrwx4zN/+9jf279+Py+UiJiaGmJiYeo8LD/cddTZjxgy+/fZbtm7d6i3r27dvqwUKlV5FwgUJhHQPQW1UY8ow+ezP+lcWtoLgJCptiAQKccJLiEzFWqzgdoBaV1d+4MCBNqsTikLlE/8k+pnXAVCheIMEgGrr11jPeYUQo+epYHJMBB/l1yW+HBzeeNK6lJ6+CzidMeyvTB58CQC6Whc5YXXt8489eoijZST0Yca0V3zKeiQNaPS+wZSens7FF1/s9+F+WM+ePdmzp67p7si5EQ258MILfbY1Gg0XX3wxgwcPZvbs2QDNSlV/TFTQ/83+De9XFJ6PSuLUCE/wqHS5+EdBPksswevXAgkU4gQ3efClXDDmJs9GKXy19iVCk92UlZWxa9euNq2bfd1KiqZNImLGgxgnTuazxDO4rHAuADtDurHX7ORsTzJWzoiP5JNB3dljrWVcVBixusaTDRpCdGSOTCRrrWcUz+EgAdDTqeH6KgPzTA7Wf3YP2fme5g2tRsfwHqcSFRrHuaOu87tmeXXrZFjo3r07l1xyCUaj0W/fpEmT6NmzJykpKahUKnbt2sWnn35az1U8zjvvPIxGIzqdjm7dujXYn3HkutnJycnNfxNN0P3B7gH3X/Wbm1PdniBh02ooi4vkL+Emlm3a4l27PTk8gfP7Tsakq/u32pi/jUX71zS5HhIoxAltQPpon+1e8SN5c87DbVQbf0pVBbaVizFOnMxdff7FPzNnYFdrcas0LA3x/ZA8NTaCUxtpbjpaev8Yb6A4WrRbzXSLgfTMs/m5qhRzTTn/vGgWSdEND3v+eNGzDe4LloSEBP7617/6lYeGhnLFFVeQkpLiU967d28efvhhvxFMAAaDoclPB0cObkhOTubkk09m8eLFAc5ovtDuJr8VxJNWWpi+U4va6mZsVd0j8IauiZSFhmBL6MIDU8/AYvfk8I4NiSJKHcYYRyaJShQubTXFvWC3o5orrt9OU0igECe09XsXkZkyxLu9M7dpqS3UcQloklJwbP/DswRiCwm7+W5CL73Ku12r8XTS3pwWT2ao/7fpY9VrVBJqjYqSg54O6I2/HfQ7ZmyfMxnb58xGr/XYp1dRXJXX6HHNVV/f0dVXXx2wWUmtVtf79HEsYmN9l3M95ZRTGDduHB9//LF3nYpgidZoeC8tjV5PuygLg1fPU7O7i4pnnqqli3cQg+9TY3moCWd4JI7YJDRAhNHzpOHATTFV/GhYz/W1k6lIW0hl6mLCLE3/vZVAIU5oy7b/yOb9yxjX50wsNjPr9ixs9BxtZl9iZ9WldCg8dWjwKqTVYjr9XDSpGegHj0DXx7d9uleIkU8Gdyc1SMkD1WoVvUYm0evPich71hdhLqs95utsz1nTKkECfJt9hgwZwvnnn98qfQZqtZr77ruPZ5+te2rS6/Vcd911rF+/nuzsbLZt24Y7CF8cJoeF0cvgCWwx1TB1g8K1PzjpotU1eI551wFMwxqfZW8Lyz/m+kigECc8c005v278pPED/xRy4V+adT9ViGdOQsQ9D6FJTffZp8vsW98pAGwa19+70l5LSeoe4RcouvSOxu3y/fAr2FPps63VtHyKdPB8WI8fP967nZqa2nody3iat2bMmMHPP//Mzp11a90MHz6c4cOHM2LEiKBMyjMeNa9l7E6Fo6e9qXQ6FIcDm1qLXaNj+d49rNyZxZjRo4mKr1tHY8AAzwCD5cuX89j8x+iZkMy9j+lI7tL0eUIq5TjzA1RVVXlXfhLiRKJJ60rcB996t5v6RKFJSSP2/a9RBfhW2JDtEwa0yuJG9honu9cWYq2yk9w9ktS+0X4fxEXZVXz51DqfsrfnPcqm/UtbvH4mk4n777/fu33PPfc0OOKppe3cuZPPPvNPmV5YWMjMmTObde2/xcdzT0xsg/v77NhOyRtv8OH3a3h1qGcQQn/Vfqw/PsnibYV+/Rp6vd6vj8ZohNpaqKys9Omor48ECiGOh9FI+A1/x/r1bFz5vin1Td1MRI6NxJplrVtSVqMl8beGF+rRH/FhrFFBN5OB//RKRa9WMTg8BE0rfmtujKIovH7L795tt+JmydbvqHVYWbVrLiVVLbcAUkhICPfddx/gGep69dVXt9i9mqK2tpZDhw7x/vvv+5QfOHCAjz766LiaodRqNS9Mv4ypTchQe8ukGRyITOEi9RKe19cFJ9VjVQHO8iWBoh3RqnXotIY2m7XamjJ0Oh5OTGKwycQThYf4vqrpv7THyqDVoNNosNjsft+i2oQG+rzYB21E3bd/W4GNmnwFx7QvPIsgHaUps6jbm9U/7mPdnAN+5TklWfz365tb7L6hoaHce++9gGf2/GWXXdZi9zoWDoeDf//7337lP//8M9u3b0dRFGw2Gy5X/atEarVawsPDOeuss+jZsycqt5uu+w8QXVFOdHEJcRUV9Z736uCL+LnbWDYabiRaVffZEuxAIX0UrSAlpht3nvs8YcZIysyFPPzJ5W1dpRZ1X3wCY0M97fBPJae0WKAYltGFy0cPAcCtKPz3l0WUVrftgjIao8YnSAAYkg0YkqE26z7MvZ7z2TcqsmPlUDpMcdUflsOMLfvl8chmsGCPNGoOnU7Hgw8+yJNPPulTftZZZ3HWWWcBnnUtfvnlFyoqKjAajeTn5xMXF0evXr0YPny4z3mKWs3+Ht05nJRl088/Y9u5kxdSfFOt3Lr5G75ZN5c7zv4LH8e81WLvT54oWsF1Ux5iWI9TvNu3z5oMQLg+lOtGXEyEIYx3139FXlXnSF/8WXoGg0x1aQb67doZ4Ojjo1Gr+e/F/kM27/tyTtsuu66BAe/UPztZcStsv2Mvul6DiH5+ls++idFhnJ8QzRUpDbdLtxe11Q7eubf+/oiPfv8vq3fPa7F7h4eHM2PGDJ8yvV6PSqVCpVKh0+kICwsjKiqK+Ph4+vfvT2JiYovV52iKovDBBx+0yKz+t956i9rCQpb1zPQpP/z3FW2EG4frmbnOTuUxZPSQJ4p2YtP+pT6BAkCr1rDi5s+JMnl+QL3ju3HlF/e2Qe2CQ6OBuDgtxY4orisu5vW4OEaFhDCztKRF7vfAWZPqLR/VLY1NBwsIM+ipqq3F4Wq5OQ71csGB5w/Q9Z6u/vsUUGprsW/d5LdrSXk1S8qrGRsVRvcQ/4R27Ulxju+iPct3zKHWbmVr9kqyCjY3cFZwmM1mbDabT9K/Iztpa2trMZvNFBQUsGPHDpYsWcLpp5/O2LGeNOo1NTUUFNTfh9KlS5d6kwkeC5VKxTXXXENFRQUrV66kqqoKl8vF7t27m3R+nz596NWrF0OHDqWyspKXXnrJuy81NZU1+flcmn2ALzK6AvB9Zd3os/Ja+O9y/0mFwSBPFK2kW2I/bjr9SWYvfo4t2SuY1H0MH17yjM8xTy2axeurZ7dRDY9faKiat99JZVn8dL5SeZrVzG+9iv3Td3EG+V4atYqHzplMmLHxP+gah4M3F68mp6yy0WODLfmqZGJP9X06KJ5TTOGXnqdGTUZ34t772u+8BSN70z/M5FfenuTsKOOHlzd5t//5wYVU17bev7Fer+eiiy7yfv4cfpowGAyEhoai0finL0lPT+fgQf/JhEcbP348p512WtDrXFpayquvvtrg/qlTpzJ48GBCQ32bIlevXs0vv/wCePo71q5teEDE8ZIninZkf+F2/vlhXbKxgir/nDj/OuWmDhkohg03ER+v9QQJRUGTXU3MiMspWf0Hzn3rGr/AMZjct6dfkLjyqZf4+F93+R1r0um4bORgnv11id++YOoWpeIfEwy4FXhyiY18s0Lx98VoQ7VEjvZ8mBV9X0TRt3U/c233TL/r3J6e0O6DBEBUom/Cwaev/oYHP55OhaVlnh6PZrfbA+Zu0mq1pKSkcO2113rLmhIkwDPXoH///n5pQJorNjaWhx56iDVr1rBs2TKftOXgyX57dJAA2LevLrV7MFc6PFaycFEb2Vmyj9dW+QaF1Tkt+9jeUqIiPb9G6coB1KU2dLuq0BTVknjJo0Bwh3WmRUf5bN81+zsSu/ek70n1N0UlRoaTFNmy4+zfPd/ETcP13DJCT94Mz72clU5y3shh2/Xb2HnXTp8goY6OIfIB307Pm9LiebBHcD+cWkp4jJGhU30nCj555edcOuHONqqRL6fTycGDB5k1axa1tf6zzEtKSli2bBlLlixhyZIlPgsVgWeN7PryQjWXRqNh7Nix3HvvvUybNs1bPmLEiHr7URwOh09iSofDEfQ6NdVxNz1VVlYSFRUV5OqceFLCE7howOn0iu3K00ve7JAd2jodfPV1VxRgVu71bN1dt7ZDzkvTIYgDV68eN5zMRE+Tzg2vvYcx1JNm2+1ykb11E7VmMxqtll9nvuw954Pl68kqKq33esFQ+U/fQBT5tLmBIz003XsS83Ld7N1+oSa+H9azVWcYB8Oan/axab7vyKNnvrmV4srcBs5oG927dyc6Opry8nIOHjyI0+nfIBoXF8f111/v3Z4yZQojR45szWr6sVgsvPJKXRr3F198EZst+OtOVFRUNNqNcNyBIjc3t97VoYQQQnQcOTk5pKamBjzmuAOF2+0mPz+f8PDwDvdNSAghTnSKomA2m0lJSWl0ffjjDhRCCCFODNKZLYQQIiAJFEIIIQKSQCGEECIgCRRCCCECkkAhhBAiIAkUQgghApJAIYQQIiAJFEIIIQKSQCGEECIgCRRCCCECkkAhhBAiIAkUQgghAvp/UQhijM/SzQUAAAAASUVORK5CYII=",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "for i in [0.1, 0.5, 0.6, 0.7, 0.8]:\n",
+    "    contours = skimage.measure.find_contours(y_pred_batch[0, :, :, 0], i)\n",
+    "\n",
+    "    # Display the image and plot all contours found\n",
+    "    fig, ax = plt.subplots()\n",
+    "    ax.imshow(y_pred_batch[0, :, :, 0], cmap=plt.cm.gray)\n",
+    "\n",
+    "    for contour in contours:\n",
+    "        ax.plot(contour[:, 1], contour[:, 0], linewidth=2)\n",
+    "\n",
+    "    ax.axis(\"image\")\n",
+    "    ax.set_xticks([])\n",
+    "    ax.set_yticks([])\n",
+    "    plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "_vgWauNkJRyH"
+   },
+   "source": [
+    "### Quick look at IceNet's work\n",
+    "```\n",
+    "def arr_to_ice_edge_arr(arr, thresh, land_mask, region_mask):\n",
+    "\n",
+    "    '''\n",
+    "    Credit IceNet https://github.com/tom-andersson/icenet-paper/blob/79ab77c452088d805514d0ba2f3ad86581945954/icenet/utils.py#L1808\n",
+    "    Compute a boolean mask with True over ice edge contour grid cells using\n",
+    "    matplotlib.pyplot.contour and an input threshold to define the ice edge\n",
+    "    (e.g. 0.15 for the 15% SIC ice edge or 0.5 for SIP forecasts). The contour\n",
+    "    along the coastline is removed using the region mask.\n",
+    "    '''\n",
+    "\n",
+    "    X, Y = np.meshgrid(np.arange(arr.shape[0]), np.arange(arr.shape[1]))\n",
+    "    X = X.T\n",
+    "    Y = Y.T\n",
+    "\n",
+    "    cs = plt.contour(X, Y, arr, [thresh], alpha=0)  # Do not plot on any axes\n",
+    "    x = []\n",
+    "    y = []\n",
+    "    for p in cs.collections[0].get_paths():\n",
+    "        x_i, y_i = p.vertices.T\n",
+    "        x.extend(np.round(x_i))\n",
+    "        y.extend(np.round(y_i))\n",
+    "    x = np.array(x, int)\n",
+    "    y = np.array(y, int)\n",
+    "    ice_edge_arr = np.zeros(arr.shape, dtype=bool)\n",
+    "    ice_edge_arr[x, y] = True\n",
+    "    # Mask out ice edge contour that hugs the coastline\n",
+    "    ice_edge_arr[land_mask] = False\n",
+    "    ice_edge_arr[region_mask == 13] = False\n",
+    "\n",
+    "    return ice_edge_arr\n",
+    "\n",
+    "```"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "43rkP56dJiH4"
+   },
+   "source": [
+    "## Filter based Edge Detection: Sobel prototype\n",
+    "This still really needs a lake and land mask, but calculating the absolute value of the tensorflow sobel edge detection, thresholding, and combining the x, y sobel filter directions yields a very nice looking edge.\n",
+    "\n",
+    "Because we don't have a land & lake mask, this is mostly picking up the edges of the landmasses. We likely want to just evaluate the sea-ice to open ocean boundary and evaluate our ice-edge metrics against that.\n",
+    "\n",
+    "Really nice that this is a differentiable system, so we can embed this into our model with a little work and compute the model outputs the current SIE predictions plus the result of the sobel edge detection.\n",
+    "\n",
+    "By finding a good threshold for this, we can likely just use the same loss function (binary cross entropy). This would mean computing the sobel edge detection for the y_target as well. It's still not utilizing any of the ice-edge metrics that we really care about in the loss function, but is a step towards informing the model how important ice edges are and penalizing it for incorrect edge forecasts."
+   ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 70,
+   "metadata": {
+    "id": "kITYFkLOboED"
+   },
+   "outputs": [],
+   "source": [
+    "truth = tf.convert_to_tensor(X_batch.T)\n",
+    "sobel = tf.image.sobel_edges(truth)\n",
+    "sobel_y = np.asarray(sobel[:, :, :, 0, 0])  # sobel in y-direction\n",
+    "sobel_x = np.asarray(sobel[:, :, :, 0, 1])  # sobel in x-direction\n",
+    "ideal_sobel = (np.abs(sobel_y[0]) > 1) + (np.abs(sobel_x[0]) > 1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 71,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "TensorShape([3, 2000, 2000, 1])"
+      ]
+     },
+     "execution_count": 71,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "truth.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 72,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Perform Sobel edge detection\n",
+    "sobel_edges = tf.image.sobel_edges(truth)\n",
+    "activated = Activation(\"sigmoid\")(\n",
+    "    tf.square(sobel_edges[:, :, :, :, 0]) + tf.square(sobel_edges[:, :, :, :, 1])\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 73,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "TensorShape([3, 2000, 2000, 1, 2])"
+      ]
+     },
+     "execution_count": 73,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sobel_edges.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 74,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
     },
-    "id": "Kj7i0uI5BuBE",
-    "outputId": "08bc159c-0f70-4b65-cd55-a3ec504f20bf"
+    "id": "oI-eO64OpCNk",
+    "outputId": "f400794e-21f9-406b-9c19-c997cc7d50f6"
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "TensorShape([3, 2000, 2000, 1])"
+      ]
+     },
+     "execution_count": 74,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
    "source": [
-    "np.all(0 <= y_pred_batch) & np.all(y_pred_batch <= 1)"
+    "activated.shape"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 75,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.image.AxesImage at 0x7580e37c3850>"
+      ]
+     },
+     "execution_count": 75,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow((activated[0, :, :, 0]))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 76,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/",
      "height": 452
     },
-    "id": "O7WR-dFdBLg4",
-    "outputId": "de4a5c78-68db-40e1-e23f-ebbfc8bdc6eb"
+    "id": "QtpiP0Rfpybb",
+    "outputId": "9a47c31c-a41e-4c84-9fce-052c432a2435"
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.image.AxesImage at 0x7580e3adccd0>"
+      ]
+     },
+     "execution_count": 76,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
    "source": [
-    "plt.imshow(y_pred_batch[0, :, :, 0])"
+    "plt.imshow(np.abs(sobel_y[0]))"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 77,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/",
-     "height": 211
+     "height": 452
     },
-    "id": "_1dQwATjBex_",
-    "outputId": "7c05ae92-4efe-4a51-e554-c96d046278e5"
+    "id": "nIwyKXDaqKLA",
+    "outputId": "cefd7a1e-6e0b-4923-83ca-28199e3eddc6"
    },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.image.AxesImage at 0x7580e3e91820>"
+      ]
+     },
+     "execution_count": 77,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow(sobel_x[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 78,
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "H_N6faw1rEeY",
+    "outputId": "89384e11-8c34-4a43-c52d-f9e82e32fa02"
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([-8., -6., -4., -3., -2., -1.,  0.,  1.,  2.,  3.,  4.],\n",
+       "      dtype=float16)"
+      ]
+     },
+     "execution_count": 78,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "np.unique(sobel_x)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 79,
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/",
+     "height": 452
+    },
+    "id": "Kg1kQS51q7km",
+    "outputId": "e8eba00f-6587-424c-e261-72514aae6ee4"
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.image.AxesImage at 0x7580e3eaabe0>"
+      ]
+     },
+     "execution_count": 79,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow((np.abs(sobel_y[0]) > 1) + (np.abs(sobel_x[0]) > 1))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 84,
+   "metadata": {},
    "outputs": [],
    "source": [
-    "import numpy as np\n",
-    "from scipy import ndimage\n",
-    "from scipy.ndimage import sobel, binary_erosion, label\n",
+    "from numba import jit\n",
     "\n",
     "\n",
-    "def find_sea_ice_edges(image):\n",
+    "@jit(nopython=True)\n",
+    "def average_ice_edge_displacement(observed_edges, model_edges):\n",
     "    \"\"\"\n",
-    "    Finds edges in the sea ice class (assumed to be labeled as 1) of a binary image.\n",
+    "    Calculate the average ice edge displacement (D_AVG_IE) between observed and model ice edges.\n",
+    "    Credit: Validation metrics for ice edge position forecasts, Melsom et al., 2019.\n",
     "\n",
-    "    Args:\n",
-    "    - image: A numpy array of shape (height, width), where pixels are 0 (no ice) or 1 (ice).\n",
+    "    Parameters:\n",
+    "    - observed_edges: numpy array of shape (N, 2), where N is the number of observed ice edge points,\n",
+    "      and each point is represented by its (x, y) coordinates.\n",
+    "    - model_edges: numpy array of shape (M, 2), where M is the number of model ice edge points,\n",
+    "      and each point is represented by its (x, y) coordinates.\n",
     "\n",
     "    Returns:\n",
-    "    - A list of coordinates for the contiguous edge pixels.\n",
+    "    - D_AVG_IE: The average displacement between the observed and model ice edges.\n",
+    "    \"\"\"\n",
+    "\n",
+    "    # Initialize lists to store minimum distances for each point\n",
+    "    observed_to_model_distances = []\n",
+    "    model_to_observed_distances = []\n",
+    "\n",
+    "    # Calculate distances from each observed point to the nearest model point\n",
+    "    for obs_point in observed_edges:\n",
+    "        distances = np.sqrt(np.sum((model_edges - obs_point) ** 2, axis=1))\n",
+    "        observed_to_model_distances.append(np.min(distances))\n",
+    "\n",
+    "    # Calculate distances from each model point to the nearest observed point\n",
+    "    for model_point in model_edges:\n",
+    "        distances = np.sqrt(np.sum((observed_edges - model_point) ** 2, axis=1))\n",
+    "        model_to_observed_distances.append(np.min(distances))\n",
+    "\n",
+    "    # Calculate the average displacement\n",
+    "    avg_displacement = (\n",
+    "        sum(observed_to_model_distances) / len(observed_to_model_distances)\n",
+    "        + sum(model_to_observed_distances) / len(model_to_observed_distances)\n",
+    "    ) / 2\n",
+    "\n",
+    "    return avg_displacement"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 91,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "@jit(nopython=True)\n",
+    "def root_mean_square_ice_edge_displacement(observed_edges, model_edges):\n",
     "    \"\"\"\n",
-    "    # Apply the Sobel filter to detect edges\n",
-    "    sx = sobel(image, axis=0, mode=\"constant\")\n",
-    "    sy = sobel(image, axis=1, mode=\"constant\")\n",
-    "    sobel_mag = np.hypot(sx, sy)\n",
+    "    Calculate the root mean square ice edge displacement (D_RMS_IE) between observed and model ice edges.\n",
     "\n",
-    "    # Threshold the Sobel magnitude to get a binary edge map\n",
-    "    edge_map = sobel_mag > np.mean(sobel_mag)\n",
+    "    Parameters:\n",
+    "    - observed_edges: numpy array of shape (N, 2), where N is the number of observed ice edge points,\n",
+    "      and each point is represented by its (x, y) coordinates.\n",
+    "    - model_edges: numpy array of shape (M, 2), where M is the number of model ice edge points,\n",
+    "      and each point is represented by its (x, y) coordinates.\n",
     "\n",
-    "    # Optionally, perform erosion to thin out the edges\n",
-    "    edge_map_thin = binary_erosion(edge_map)\n",
+    "    Returns:\n",
+    "    - D_RMS_IE: The root mean square displacement between the observed and model ice edges.\n",
+    "    \"\"\"\n",
     "\n",
-    "    # Find connected components in the thinned edge map\n",
-    "    labeled_array, num_features = label(edge_map_thin)\n",
+    "    # Initialize lists to store distances for each point\n",
+    "    observed_to_model_distances = []\n",
+    "    model_to_observed_distances = []\n",
     "\n",
-    "    # Extract the coordinates of the edge pixels\n",
-    "    edge_indices = np.argwhere(labeled_array > 0)\n",
+    "    # Calculate distances from each observed point to the nearest model predicted point\n",
+    "    for obs_point in observed_edges:\n",
+    "        distances = np.sqrt(np.sum((model_edges - obs_point) ** 2, axis=1))\n",
+    "        observed_to_model_distances.append(np.min(distances) ** 2)\n",
     "\n",
-    "    # Optionally, return the labeled array for visualization or further analysis\n",
-    "    return edge_indices, labeled_array\n",
+    "    # Calculate distances from each model point to the nearest observed point\n",
+    "    for model_point in model_edges:\n",
+    "        distances = np.sqrt(np.sum((observed_edges - model_point) ** 2, axis=1))\n",
+    "        model_to_observed_distances.append(np.min(distances) ** 2)\n",
     "\n",
+    "    # Calculate the root mean square displacement\n",
+    "    rms_displacement = np.sqrt(\n",
+    "        (sum(observed_to_model_distances) + sum(model_to_observed_distances))\n",
+    "        / (len(observed_to_model_distances) + len(model_to_observed_distances))\n",
+    "    )\n",
     "\n",
-    "# Example usage\n",
-    "# Assuming `predictions` is a numpy array from your model with shape (height, width) and binary values\n",
-    "predictions = y_pred_batch[0, :, :, 0]  # Dummy data for demonstration\n",
-    "edge_indices, labeled_edges = find_sea_ice_edges(predictions)\n",
+    "    return rms_displacement"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 92,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "edges = {}\n",
+    "for i in range(activated.shape[0]):\n",
+    "    edges[i] = np.where(activated[i, :, :, 0] >= 1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 94,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.03760836743969396 0.8457719932868671\n",
+      "0.0 0.0\n"
+     ]
+    }
+   ],
+   "source": [
+    "for i in range(1, activated.shape[0]):\n",
+    "    x, y = edges[i][0], edges[i][1]\n",
+    "    current_edges = np.stack((x, y), axis=-1)\n",
     "\n",
-    "# edge_indices contains the coordinates of all edge pixels\n",
-    "# labeled_edges is the labeled edge map, useful for visualization or further analysis"
+    "    x_prev, y_prev = edges[i - 1][0], edges[i - 1][1]\n",
+    "    previous_edges = np.stack((x_prev, y_prev), axis=-1)\n",
+    "\n",
+    "    avg_displacement = average_ice_edge_displacement(current_edges, previous_edges)\n",
+    "    rms_displacement = root_mean_square_ice_edge_displacement(\n",
+    "        current_edges, previous_edges\n",
+    "    )\n",
+    "    print(avg_displacement, rms_displacement)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 96,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(26434, 2)"
+      ]
+     },
+     "execution_count": 96,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "current_edges.shape"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "qspqpmthx4uC"
+   },
+   "source": [
+    "## Prototyping the sobel edge detection layer\n",
+    "Very early prototype, no promises that there aren't bugs."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {
+    "id": "42C80tIRrRI5"
+   },
+   "outputs": [],
+   "source": [
+    "import tensorflow as tf\n",
+    "from tensorflow.keras.layers import (\n",
+    "    Conv2D,\n",
+    "    MaxPooling2D,\n",
+    "    concatenate,\n",
+    "    Conv2DTranspose,\n",
+    "    Input,\n",
+    "    Layer,\n",
+    "    Activation,\n",
+    ")\n",
+    "from tensorflow.keras.models import Model\n",
+    "\n",
+    "\n",
+    "class SobelEdgeLayer(Layer):\n",
+    "    def __init__(self, threshold=1, **kwargs):\n",
+    "        super(SobelEdgeLayer, self).__init__(**kwargs)\n",
+    "        self.threshold = threshold\n",
+    "\n",
+    "    def call(self, inputs):\n",
+    "        # Perform Sobel edge detection\n",
+    "        sobel_edges = tf.image.sobel_edges(inputs)\n",
+    "\n",
+    "        # Extract edges for x and y directions and apply sigmoid activation\n",
+    "        activated = Activation(\"sigmoid\")(\n",
+    "            tf.square(sobel_edges[:, :, :, :, 0])\n",
+    "            + tf.square(sobel_edges[:, :, :, :, 1])\n",
+    "        )\n",
+    "        return activated\n",
+    "\n",
+    "    def compute_output_shape(self, input_shape):\n",
+    "        return input_shape\n",
+    "\n",
+    "\n",
+    "def simple_unet_with_sobel(input_shape=(8000 * 2, 8000 * 2, 5)):\n",
+    "    inputs = Input(input_shape)\n",
+    "    # Downsample\n",
+    "    c1 = Conv2D(16, (3, 3), activation=\"relu\", padding=\"same\")(inputs)\n",
+    "    p1 = MaxPooling2D((2, 2))(c1)\n",
+    "    c2 = Conv2D(32, (3, 3), activation=\"relu\", padding=\"same\")(p1)\n",
+    "    p2 = MaxPooling2D((2, 2))(c2)\n",
+    "    # Bottleneck\n",
+    "    b = Conv2D(64, (3, 3), activation=\"relu\", padding=\"same\")(p2)\n",
+    "    # Upsample\n",
+    "    u1 = Conv2DTranspose(32, (2, 2), strides=(2, 2), padding=\"same\")(b)\n",
+    "    u1 = concatenate([u1, c2])\n",
+    "    u2 = Conv2DTranspose(16, (2, 2), strides=(2, 2), padding=\"same\")(u1)\n",
+    "    u2 = concatenate([u2, c1])\n",
+    "\n",
+    "    # Class prediction layer\n",
+    "    class_pred = Conv2D(1, (1, 1), activation=\"sigmoid\")(u2)\n",
+    "\n",
+    "    # Sobel edge layer (applied on class prediction)\n",
+    "    sobel_edges = SobelEdgeLayer()(class_pred)\n",
+    "\n",
+    "    # Concatenate class prediction and Sobel edge detection results\n",
+    "    # outputs = concatenate([class_pred, sobel_edges], axis=-1)\n",
+    "\n",
+    "    model = Model(inputs=[inputs], outputs=[sobel_edges])\n",
+    "    return model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 58,
+   "metadata": {
+    "id": "8ZnJuVY_x767"
+   },
+   "outputs": [],
+   "source": [
+    "def apply_sobel_edge_detection(image, threshold=1):\n",
+    "    \"\"\"perform Sobel edge detection on an (optionally batched) image\"\"\"\n",
+    "    # Ensure image is a float32 tensor for TensorFlow operations\n",
+    "    image = tf.convert_to_tensor(image, dtype=tf.float32)\n",
+    "    image = tf.expand_dims(image, axis=0)  # Add batch dimension\n",
+    "    print(image.get_shape())\n",
+    "    # Perform Sobel edge detection\n",
+    "    sobel = tf.image.sobel_edges(image)\n",
+    "    sobel_x = sobel[..., 0]\n",
+    "    sobel_y = sobel[..., 1]\n",
+    "\n",
+    "    # Apply absolute value, threshold, and sum the results of x and y edges\n",
+    "    edges_x = tf.cast(tf.square(sobel_x) > threshold, tf.float32)\n",
+    "    edges_y = tf.cast(tf.square(sobel_y) > threshold, tf.float32)\n",
+    "    edges = (\n",
+    "        edges_x + edges_y\n",
+    "    )  # tf.reduce_sum(tf.concat([edges_x, edges_y], axis=-1), axis=-1, keepdims=True)\n",
+    "\n",
+    "    return edges\n",
+    "\n",
+    "\n",
+    "class SobelGenerator(AllDataGenerator):\n",
+    "    \"\"\"\n",
+    "    Generator for Keras training to allow multiprocessing and training on batches with only the\n",
+    "    batch itself being loaded into memory.\n",
+    "    \"\"\"\n",
+    "\n",
+    "    # Existing __init__, years_from_filenames, days_from_filenames, _get_data_ids, __len__, and on_epoch_end methods\n",
+    "\n",
+    "    def __getitem__(self, index):\n",
+    "        \"\"\"Generate one batch of data\"\"\"\n",
+    "        # Collect data IDs for this batch number\n",
+    "        batch_data_ids = self.data_IDs[\n",
+    "            index * self.batch_size : (index + 1) * self.batch_size\n",
+    "        ]\n",
+    "\n",
+    "        # Generate data\n",
+    "        X, y = self._data_generation(batch_data_ids)\n",
+    "\n",
+    "        return X.astype(\"float16\"), y.astype(\"float32\")\n",
+    "\n",
+    "    def _data_generation(self, batch_data_ids):\n",
+    "        \"\"\"Generates data containing batch_size samples\"\"\"\n",
+    "        X = np.empty((self.batch_size, *self.dim), dtype=\"float16\")\n",
+    "        y = np.empty((self.batch_size, self.dim[0], self.dim[1], 1), dtype=\"float32\")\n",
+    "\n",
+    "        for i, (year, day) in enumerate(batch_data_ids):\n",
+    "            # Load a 5-day chunk as the input\n",
+    "            X[i,] = self.load_n_day_chunk(i, self.dim[2])\n",
+    "            # Load the next day as the target\n",
+    "            target_image = load_target_sie_data(\n",
+    "                self.years[i + self.dim[2]], self.days[i + self.dim[2]]\n",
+    "            )\n",
+    "            print(target_image.shape)\n",
+    "            # y[i, :, :, :1] = np.expand_dims(target_image, axis=-1)\n",
+    "            # Apply Sobel edge detection to the target and concatenate it to y\n",
+    "            sobel_edges = apply_sobel_edge_detection(target_image)\n",
+    "            y[\n",
+    "                i\n",
+    "            ] = sobel_edges.numpy()  # Make sure to convert the TF tensor to NumPy array\n",
+    "\n",
+    "        return X, y"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 59,
    "metadata": {
-    "id": "iCg6CNxpd8Cv"
+    "id": "tg5yllday9C-"
    },
    "outputs": [],
+   "source": [
+    "def combined_loss(y_true, y_pred, beta=0.5):\n",
+    "    # Assume y_true and y_pred are both of shape (batch_size, height, width, channels)\n",
+    "    # and that the last two channels of y_pred are the Sobel edge detection results.\n",
+    "    # Also assume that y_true is binary (0 or 1) for the class predictions.\n",
+    "    class_true = y_true[..., :1]  # Extract the class prediction part from y_true\n",
+    "    class_pred = y_pred[..., :1]  # Extract the class prediction part from y_pred\n",
+    "\n",
+    "    edge_pred = y_pred[..., 1:]  # Extract the edge detection part from y_pred\n",
+    "\n",
+    "    # Binary cross-entropy for the class predictions\n",
+    "    bce_loss = tf.keras.losses.binary_crossentropy(class_true, class_pred)\n",
+    "\n",
+    "    # MSE for the edge detection part (assuming no ground truth for edges, using predicted edges as target)\n",
+    "    mse_loss = tf.keras.losses.mean_squared_error(class_true, edge_pred)\n",
+    "\n",
+    "    # Combine the losses\n",
+    "    combined = (1 - beta) * bce_loss + beta * mse_loss\n",
+    "    return combined"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "bLBbED1SyLIq",
+    "outputId": "ba4b553b-3025-460e-ce1b-859c40fb908c"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(2000, 2000)\n",
+      "(1, 2000, 2000)\n"
+     ]
+    },
+    {
+     "ename": "InvalidArgumentError",
+     "evalue": "The first dimension of paddings must be the rank of inputs[4,2], [1,2000,2000] [Op:MirrorPad]",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mInvalidArgumentError\u001b[0m                      Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[60], line 33\u001b[0m\n\u001b[1;32m     28\u001b[0m model\u001b[38;5;241m.\u001b[39mcompile(\n\u001b[1;32m     29\u001b[0m     optimizer\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124madam\u001b[39m\u001b[38;5;124m\"\u001b[39m, loss\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mbinary_crossentropy\u001b[39m\u001b[38;5;124m\"\u001b[39m, metrics\u001b[38;5;241m=\u001b[39m[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124maccuracy\u001b[39m\u001b[38;5;124m\"\u001b[39m]\n\u001b[1;32m     30\u001b[0m )  \u001b[38;5;66;03m#'binary_crossentropy', metrics=['accuracy']) 'sparse_categorical_crossentropy'\u001b[39;00m\n\u001b[1;32m     32\u001b[0m \u001b[38;5;66;03m# Train the model\u001b[39;00m\n\u001b[0;32m---> 33\u001b[0m history \u001b[38;5;241m=\u001b[39m \u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m     34\u001b[0m \u001b[43m    \u001b[49m\u001b[43msobel_train_generator\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     35\u001b[0m \u001b[43m    \u001b[49m\u001b[43mepochs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m     36\u001b[0m \u001b[43m    \u001b[49m\u001b[43mvalidation_data\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m \u001b[49m\u001b[43msobel_test_generator\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     37\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;66;43;03m# Only for training on CPU\u001b[39;49;00m\n\u001b[1;32m     38\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;66;43;03m# use_multiprocessing=True,\u001b[39;49;00m\n\u001b[1;32m     39\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;66;43;03m# workers=2,\u001b[39;49;00m\n\u001b[1;32m     40\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcallbacks\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m[\u001b[49m\u001b[43mcheckpoint_callback\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mearly_stopping_callback\u001b[49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     41\u001b[0m \u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/keras/engine/training.py:1134\u001b[0m, in \u001b[0;36mModel.fit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m   1128\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_cluster_coordinator \u001b[38;5;241m=\u001b[39m tf\u001b[38;5;241m.\u001b[39mdistribute\u001b[38;5;241m.\u001b[39mexperimental\u001b[38;5;241m.\u001b[39mcoordinator\u001b[38;5;241m.\u001b[39mClusterCoordinator(\n\u001b[1;32m   1129\u001b[0m       \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdistribute_strategy)\n\u001b[1;32m   1131\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdistribute_strategy\u001b[38;5;241m.\u001b[39mscope(), \\\n\u001b[1;32m   1132\u001b[0m      training_utils\u001b[38;5;241m.\u001b[39mRespectCompiledTrainableState(\u001b[38;5;28mself\u001b[39m):\n\u001b[1;32m   1133\u001b[0m   \u001b[38;5;66;03m# Creates a `tf.data.Dataset` and handles batch and epoch iteration.\u001b[39;00m\n\u001b[0;32m-> 1134\u001b[0m   data_handler \u001b[38;5;241m=\u001b[39m \u001b[43mdata_adapter\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_data_handler\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1135\u001b[0m \u001b[43m      \u001b[49m\u001b[43mx\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mx\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1136\u001b[0m \u001b[43m      \u001b[49m\u001b[43my\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43my\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1137\u001b[0m \u001b[43m      \u001b[49m\u001b[43msample_weight\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43msample_weight\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1138\u001b[0m \u001b[43m      \u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mbatch_size\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1139\u001b[0m \u001b[43m      \u001b[49m\u001b[43msteps_per_epoch\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43msteps_per_epoch\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1140\u001b[0m \u001b[43m      \u001b[49m\u001b[43minitial_epoch\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43minitial_epoch\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1141\u001b[0m \u001b[43m      \u001b[49m\u001b[43mepochs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mepochs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1142\u001b[0m \u001b[43m      \u001b[49m\u001b[43mshuffle\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mshuffle\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1143\u001b[0m \u001b[43m      \u001b[49m\u001b[43mclass_weight\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mclass_weight\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1144\u001b[0m \u001b[43m      \u001b[49m\u001b[43mmax_queue_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mmax_queue_size\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1145\u001b[0m \u001b[43m      \u001b[49m\u001b[43mworkers\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mworkers\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1146\u001b[0m \u001b[43m      \u001b[49m\u001b[43muse_multiprocessing\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43muse_multiprocessing\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1147\u001b[0m \u001b[43m      \u001b[49m\u001b[43mmodel\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1148\u001b[0m \u001b[43m      \u001b[49m\u001b[43msteps_per_execution\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_steps_per_execution\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1150\u001b[0m   \u001b[38;5;66;03m# Container that configures and calls `tf.keras.Callback`s.\u001b[39;00m\n\u001b[1;32m   1151\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(callbacks, callbacks_module\u001b[38;5;241m.\u001b[39mCallbackList):\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/keras/engine/data_adapter.py:1383\u001b[0m, in \u001b[0;36mget_data_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m   1381\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mgetattr\u001b[39m(kwargs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodel\u001b[39m\u001b[38;5;124m\"\u001b[39m], \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m_cluster_coordinator\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m):\n\u001b[1;32m   1382\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m _ClusterCoordinatorDataHandler(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[0;32m-> 1383\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mDataHandler\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/keras/engine/data_adapter.py:1138\u001b[0m, in \u001b[0;36mDataHandler.__init__\u001b[0;34m(self, x, y, sample_weight, batch_size, steps_per_epoch, initial_epoch, epochs, shuffle, class_weight, max_queue_size, workers, use_multiprocessing, model, steps_per_execution, distribute)\u001b[0m\n\u001b[1;32m   1135\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_steps_per_execution_value \u001b[38;5;241m=\u001b[39m steps_per_execution\u001b[38;5;241m.\u001b[39mnumpy()\u001b[38;5;241m.\u001b[39mitem()\n\u001b[1;32m   1137\u001b[0m adapter_cls \u001b[38;5;241m=\u001b[39m select_data_adapter(x, y)\n\u001b[0;32m-> 1138\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_adapter \u001b[38;5;241m=\u001b[39m \u001b[43madapter_cls\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1139\u001b[0m \u001b[43m    \u001b[49m\u001b[43mx\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1140\u001b[0m \u001b[43m    \u001b[49m\u001b[43my\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1141\u001b[0m \u001b[43m    \u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mbatch_size\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1142\u001b[0m \u001b[43m    \u001b[49m\u001b[43msteps\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43msteps_per_epoch\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1143\u001b[0m \u001b[43m    \u001b[49m\u001b[43mepochs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mepochs\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m-\u001b[39;49m\u001b[43m \u001b[49m\u001b[43minitial_epoch\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1144\u001b[0m \u001b[43m    \u001b[49m\u001b[43msample_weights\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43msample_weight\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1145\u001b[0m \u001b[43m    \u001b[49m\u001b[43mshuffle\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mshuffle\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1146\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmax_queue_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mmax_queue_size\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1147\u001b[0m \u001b[43m    \u001b[49m\u001b[43mworkers\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mworkers\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1148\u001b[0m \u001b[43m    \u001b[49m\u001b[43muse_multiprocessing\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43muse_multiprocessing\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1149\u001b[0m \u001b[43m    \u001b[49m\u001b[43mdistribution_strategy\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtf\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdistribute\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_strategy\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1150\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmodel\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mmodel\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1152\u001b[0m strategy \u001b[38;5;241m=\u001b[39m tf\u001b[38;5;241m.\u001b[39mdistribute\u001b[38;5;241m.\u001b[39mget_strategy()\n\u001b[1;32m   1154\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_current_step \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/keras/engine/data_adapter.py:917\u001b[0m, in \u001b[0;36mKerasSequenceAdapter.__init__\u001b[0;34m(self, x, y, sample_weights, shuffle, workers, use_multiprocessing, max_queue_size, model, **kwargs)\u001b[0m\n\u001b[1;32m    915\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_keras_sequence \u001b[38;5;241m=\u001b[39m x\n\u001b[1;32m    916\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_enqueuer \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[0;32m--> 917\u001b[0m \u001b[38;5;28;43msuper\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mKerasSequenceAdapter\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[38;5;21;43m__init__\u001b[39;49m\u001b[43m(\u001b[49m\n\u001b[1;32m    918\u001b[0m \u001b[43m    \u001b[49m\u001b[43mx\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    919\u001b[0m \u001b[43m    \u001b[49m\u001b[43mshuffle\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# Shuffle is handed in the _make_callable override.\u001b[39;49;00m\n\u001b[1;32m    920\u001b[0m \u001b[43m    \u001b[49m\u001b[43mworkers\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mworkers\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    921\u001b[0m \u001b[43m    \u001b[49m\u001b[43muse_multiprocessing\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43muse_multiprocessing\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    922\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmax_queue_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mmax_queue_size\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    923\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmodel\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    924\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/keras/engine/data_adapter.py:794\u001b[0m, in \u001b[0;36mGeneratorDataAdapter.__init__\u001b[0;34m(self, x, y, sample_weights, workers, use_multiprocessing, max_queue_size, model, **kwargs)\u001b[0m\n\u001b[1;32m    790\u001b[0m \u001b[38;5;28msuper\u001b[39m(GeneratorDataAdapter, \u001b[38;5;28mself\u001b[39m)\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__init__\u001b[39m(x, y, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[1;32m    792\u001b[0m \u001b[38;5;66;03m# Since we have to know the dtype of the python generator when we build the\u001b[39;00m\n\u001b[1;32m    793\u001b[0m \u001b[38;5;66;03m# dataset, we have to look at a batch to infer the structure.\u001b[39;00m\n\u001b[0;32m--> 794\u001b[0m peek, x \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_peek_and_restore\u001b[49m\u001b[43m(\u001b[49m\u001b[43mx\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    795\u001b[0m peek \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_standardize_batch(peek)\n\u001b[1;32m    796\u001b[0m peek \u001b[38;5;241m=\u001b[39m _process_tensorlike(peek)\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/keras/engine/data_adapter.py:928\u001b[0m, in \u001b[0;36mKerasSequenceAdapter._peek_and_restore\u001b[0;34m(x)\u001b[0m\n\u001b[1;32m    926\u001b[0m \u001b[38;5;129m@staticmethod\u001b[39m\n\u001b[1;32m    927\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_peek_and_restore\u001b[39m(x):\n\u001b[0;32m--> 928\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mx\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m]\u001b[49m, x\n",
+      "Cell \u001b[0;32mIn[58], line 32\u001b[0m, in \u001b[0;36mSobelGenerator.__getitem__\u001b[0;34m(self, index)\u001b[0m\n\u001b[1;32m     29\u001b[0m batch_data_ids \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdata_IDs[index \u001b[38;5;241m*\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mbatch_size : (index \u001b[38;5;241m+\u001b[39m \u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mbatch_size]\n\u001b[1;32m     31\u001b[0m \u001b[38;5;66;03m# Generate data\u001b[39;00m\n\u001b[0;32m---> 32\u001b[0m X, y \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_data_generation\u001b[49m\u001b[43m(\u001b[49m\u001b[43mbatch_data_ids\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     34\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m X\u001b[38;5;241m.\u001b[39mastype(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mfloat16\u001b[39m\u001b[38;5;124m\"\u001b[39m), y\u001b[38;5;241m.\u001b[39mastype(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mfloat32\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
+      "Cell \u001b[0;32mIn[58], line 49\u001b[0m, in \u001b[0;36mSobelGenerator._data_generation\u001b[0;34m(self, batch_data_ids)\u001b[0m\n\u001b[1;32m     46\u001b[0m     \u001b[38;5;28mprint\u001b[39m(target_image\u001b[38;5;241m.\u001b[39mshape)\n\u001b[1;32m     47\u001b[0m     \u001b[38;5;66;03m# y[i, :, :, :1] = np.expand_dims(target_image, axis=-1)\u001b[39;00m\n\u001b[1;32m     48\u001b[0m     \u001b[38;5;66;03m# Apply Sobel edge detection to the target and concatenate it to y\u001b[39;00m\n\u001b[0;32m---> 49\u001b[0m     sobel_edges \u001b[38;5;241m=\u001b[39m \u001b[43mapply_sobel_edge_detection\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtarget_image\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     50\u001b[0m     y[i] \u001b[38;5;241m=\u001b[39m sobel_edges\u001b[38;5;241m.\u001b[39mnumpy()  \u001b[38;5;66;03m# Make sure to convert the TF tensor to NumPy array\u001b[39;00m\n\u001b[1;32m     52\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m X, y\n",
+      "Cell \u001b[0;32mIn[58], line 8\u001b[0m, in \u001b[0;36mapply_sobel_edge_detection\u001b[0;34m(image, threshold)\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[38;5;28mprint\u001b[39m(image\u001b[38;5;241m.\u001b[39mget_shape())\n\u001b[1;32m      7\u001b[0m \u001b[38;5;66;03m# Perform Sobel edge detection\u001b[39;00m\n\u001b[0;32m----> 8\u001b[0m sobel \u001b[38;5;241m=\u001b[39m \u001b[43mtf\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mimage\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msobel_edges\u001b[49m\u001b[43m(\u001b[49m\u001b[43mimage\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m      9\u001b[0m sobel_x \u001b[38;5;241m=\u001b[39m sobel[\u001b[38;5;241m.\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;241m.\u001b[39m, \u001b[38;5;241m0\u001b[39m]\n\u001b[1;32m     10\u001b[0m sobel_y \u001b[38;5;241m=\u001b[39m sobel[\u001b[38;5;241m.\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;241m.\u001b[39m, \u001b[38;5;241m1\u001b[39m]\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/tensorflow/python/util/dispatch.py:206\u001b[0m, in \u001b[0;36madd_dispatch_support.<locals>.wrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    204\u001b[0m \u001b[38;5;250m\u001b[39m\u001b[38;5;124;03m\"\"\"Call target, and fall back on dispatchers if there is a TypeError.\"\"\"\u001b[39;00m\n\u001b[1;32m    205\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 206\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mtarget\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    207\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m (\u001b[38;5;167;01mTypeError\u001b[39;00m, \u001b[38;5;167;01mValueError\u001b[39;00m):\n\u001b[1;32m    208\u001b[0m   \u001b[38;5;66;03m# Note: convert_to_eager_tensor currently raises a ValueError, not a\u001b[39;00m\n\u001b[1;32m    209\u001b[0m   \u001b[38;5;66;03m# TypeError, when given unexpected types.  So we need to catch both.\u001b[39;00m\n\u001b[1;32m    210\u001b[0m   result \u001b[38;5;241m=\u001b[39m dispatch(wrapper, args, kwargs)\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/tensorflow/python/ops/image_ops_impl.py:4583\u001b[0m, in \u001b[0;36msobel_edges\u001b[0;34m(image)\u001b[0m\n\u001b[1;32m   4581\u001b[0m \u001b[38;5;66;03m# Use depth-wise convolution to calculate edge maps per channel.\u001b[39;00m\n\u001b[1;32m   4582\u001b[0m pad_sizes \u001b[38;5;241m=\u001b[39m [[\u001b[38;5;241m0\u001b[39m, \u001b[38;5;241m0\u001b[39m], [\u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m1\u001b[39m], [\u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m1\u001b[39m], [\u001b[38;5;241m0\u001b[39m, \u001b[38;5;241m0\u001b[39m]]\n\u001b[0;32m-> 4583\u001b[0m padded \u001b[38;5;241m=\u001b[39m \u001b[43marray_ops\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpad\u001b[49m\u001b[43m(\u001b[49m\u001b[43mimage\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mpad_sizes\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmode\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mREFLECT\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m   4585\u001b[0m \u001b[38;5;66;03m# Output tensor has shape [batch_size, h, w, d * num_kernels].\u001b[39;00m\n\u001b[1;32m   4586\u001b[0m strides \u001b[38;5;241m=\u001b[39m [\u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m1\u001b[39m]\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/tensorflow/python/util/dispatch.py:206\u001b[0m, in \u001b[0;36madd_dispatch_support.<locals>.wrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    204\u001b[0m \u001b[38;5;250m\u001b[39m\u001b[38;5;124;03m\"\"\"Call target, and fall back on dispatchers if there is a TypeError.\"\"\"\u001b[39;00m\n\u001b[1;32m    205\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 206\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mtarget\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    207\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m (\u001b[38;5;167;01mTypeError\u001b[39;00m, \u001b[38;5;167;01mValueError\u001b[39;00m):\n\u001b[1;32m    208\u001b[0m   \u001b[38;5;66;03m# Note: convert_to_eager_tensor currently raises a ValueError, not a\u001b[39;00m\n\u001b[1;32m    209\u001b[0m   \u001b[38;5;66;03m# TypeError, when given unexpected types.  So we need to catch both.\u001b[39;00m\n\u001b[1;32m    210\u001b[0m   result \u001b[38;5;241m=\u001b[39m dispatch(wrapper, args, kwargs)\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/tensorflow/python/ops/array_ops.py:3533\u001b[0m, in \u001b[0;36mpad\u001b[0;34m(tensor, paddings, mode, name, constant_values)\u001b[0m\n\u001b[1;32m   3530\u001b[0m     result \u001b[38;5;241m=\u001b[39m gen_array_ops\u001b[38;5;241m.\u001b[39mpad_v2(\n\u001b[1;32m   3531\u001b[0m         tensor, paddings, constant_values, name\u001b[38;5;241m=\u001b[39mname)\n\u001b[1;32m   3532\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m mode \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mREFLECT\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[0;32m-> 3533\u001b[0m   result \u001b[38;5;241m=\u001b[39m \u001b[43mgen_array_ops\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmirror_pad\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   3534\u001b[0m \u001b[43m      \u001b[49m\u001b[43mtensor\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mpaddings\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmode\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mREFLECT\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mname\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mname\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   3535\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m mode \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mSYMMETRIC\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[1;32m   3536\u001b[0m   result \u001b[38;5;241m=\u001b[39m gen_array_ops\u001b[38;5;241m.\u001b[39mmirror_pad(\n\u001b[1;32m   3537\u001b[0m       tensor, paddings, mode\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mSYMMETRIC\u001b[39m\u001b[38;5;124m\"\u001b[39m, name\u001b[38;5;241m=\u001b[39mname)\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/tensorflow/python/ops/gen_array_ops.py:6005\u001b[0m, in \u001b[0;36mmirror_pad\u001b[0;34m(input, paddings, mode, name)\u001b[0m\n\u001b[1;32m   6003\u001b[0m   \u001b[38;5;28;01mpass\u001b[39;00m\n\u001b[1;32m   6004\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 6005\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mmirror_pad_eager_fallback\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   6006\u001b[0m \u001b[43m      \u001b[49m\u001b[38;5;28;43minput\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mpaddings\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmode\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mmode\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mname\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mname\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mctx\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m_ctx\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   6007\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m _core\u001b[38;5;241m.\u001b[39m_SymbolicException:\n\u001b[1;32m   6008\u001b[0m   \u001b[38;5;28;01mpass\u001b[39;00m  \u001b[38;5;66;03m# Add nodes to the TensorFlow graph.\u001b[39;00m\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/tensorflow/python/ops/gen_array_ops.py:6032\u001b[0m, in \u001b[0;36mmirror_pad_eager_fallback\u001b[0;34m(input, paddings, mode, name, ctx)\u001b[0m\n\u001b[1;32m   6030\u001b[0m _inputs_flat \u001b[38;5;241m=\u001b[39m [\u001b[38;5;28minput\u001b[39m, paddings]\n\u001b[1;32m   6031\u001b[0m _attrs \u001b[38;5;241m=\u001b[39m (\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mT\u001b[39m\u001b[38;5;124m\"\u001b[39m, _attr_T, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mTpaddings\u001b[39m\u001b[38;5;124m\"\u001b[39m, _attr_Tpaddings, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmode\u001b[39m\u001b[38;5;124m\"\u001b[39m, mode)\n\u001b[0;32m-> 6032\u001b[0m _result \u001b[38;5;241m=\u001b[39m \u001b[43m_execute\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mexecute\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43mb\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mMirrorPad\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m_inputs_flat\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   6033\u001b[0m \u001b[43m                           \u001b[49m\u001b[43mattrs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m_attrs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mctx\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mctx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mname\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mname\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   6034\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m _execute\u001b[38;5;241m.\u001b[39mmust_record_gradient():\n\u001b[1;32m   6035\u001b[0m   _execute\u001b[38;5;241m.\u001b[39mrecord_gradient(\n\u001b[1;32m   6036\u001b[0m       \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mMirrorPad\u001b[39m\u001b[38;5;124m\"\u001b[39m, _inputs_flat, _attrs, _result)\n",
+      "File \u001b[0;32m~/code/personal/icedyno/unet_model/.pixi/envs/default/lib/python3.9/site-packages/tensorflow/python/eager/execute.py:59\u001b[0m, in \u001b[0;36mquick_execute\u001b[0;34m(op_name, num_outputs, inputs, attrs, ctx, name)\u001b[0m\n\u001b[1;32m     57\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m     58\u001b[0m   ctx\u001b[38;5;241m.\u001b[39mensure_initialized()\n\u001b[0;32m---> 59\u001b[0m   tensors \u001b[38;5;241m=\u001b[39m \u001b[43mpywrap_tfe\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mTFE_Py_Execute\u001b[49m\u001b[43m(\u001b[49m\u001b[43mctx\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_handle\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdevice_name\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mop_name\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     60\u001b[0m \u001b[43m                                      \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mattrs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnum_outputs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     61\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m core\u001b[38;5;241m.\u001b[39m_NotOkStatusException \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m     62\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n",
+      "\u001b[0;31mInvalidArgumentError\u001b[0m: The first dimension of paddings must be the rank of inputs[4,2], [1,2000,2000] [Op:MirrorPad]"
+     ]
+    }
+   ],
+   "source": [
+    "datetime_string = datetime.datetime.now().strftime(\"%I:%M%p_%B_%d_%Y\")\n",
+    "\n",
+    "# Model checkpoint foldernames now generated by datetime (won't overwrite previous runs)\n",
+    "checkpoint_dir = f\"./model_checkpoints/jbacon/unet_{datetime_string}_{WINDOW_SIZE}km/\"\n",
+    "if not os.path.exists(checkpoint_dir):\n",
+    "    os.makedirs(checkpoint_dir)\n",
+    "checkpoint_path = os.path.join(checkpoint_dir, \"cp-{epoch:04d}.ckpt\")\n",
+    "\n",
+    "checkpoint_callback = ModelCheckpoint(\n",
+    "    filepath=checkpoint_path,\n",
+    "    save_weights_only=False,\n",
+    "    monitor=\"loss\",\n",
+    "    mode=\"min\",\n",
+    "    save_best_only=True,\n",
+    "    verbose=1,\n",
+    ")\n",
+    "\n",
+    "early_stopping_callback = EarlyStopping(\n",
+    "    monitor=\"loss\", patience=10, verbose=1, mode=\"min\"\n",
+    ")\n",
+    "\n",
+    "model = simple_unet_with_sobel(input_shape=dim)  # skip(input_shape=dim)\n",
+    "\n",
+    "# Example of compiling the model with the custom loss\n",
+    "# model.compile(optimizer='adam', loss=combined_loss, metrics=['accuracy'])\n",
+    "sobel_train_generator = SobelGenerator(train_files, batch_size=batch_size, dim=dim)\n",
+    "sobel_test_generator = SobelGenerator(test_files, batch_size=test_batch_size, dim=dim)\n",
+    "model.compile(\n",
+    "    optimizer=\"adam\", loss=\"binary_crossentropy\", metrics=[\"accuracy\"]\n",
+    ")  #'binary_crossentropy', metrics=['accuracy']) 'sparse_categorical_crossentropy'\n",
+    "\n",
+    "# Train the model\n",
+    "history = model.fit(\n",
+    "    sobel_train_generator,\n",
+    "    epochs=1,\n",
+    "    validation_data=sobel_test_generator,\n",
+    "    # Only for training on CPU\n",
+    "    # use_multiprocessing=True,\n",
+    "    # workers=2,\n",
+    "    callbacks=[checkpoint_callback, early_stopping_callback],\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "id": "_RIqv4yLzLN3"
+   },
+   "outputs": [],
+   "source": [
+    "0.7243\n",
+    "\n",
+    "\n",
+    "loss started at like 0.6931"
+   ]
   }
  ],
  "metadata": {
-  "accelerator": "GPU",
   "colab": {
-   "gpuType": "T4",
    "provenance": []
   },
   "kernelspec": {
diff --git a/unet_model/pixi.lock b/unet_model/pixi.lock
index d73f1bd..a22a547 100644
--- a/unet_model/pixi.lock
+++ b/unet_model/pixi.lock
@@ -77,6 +77,7 @@ environments:
       - conda: https://conda.anaconda.org/conda-forge/linux-64/libgomp-13.2.0-h807b86a_5.conda
       - conda: https://conda.anaconda.org/conda-forge/linux-64/libiconv-1.17-hd590300_2.conda
       - conda: https://conda.anaconda.org/conda-forge/linux-64/liblapack-3.9.0-20_linux64_openblas.conda
+      - conda: https://conda.anaconda.org/conda-forge/linux-64/libllvm11-11.1.0-he0ac6c6_5.tar.bz2
       - conda: https://conda.anaconda.org/conda-forge/linux-64/libllvm13-13.0.1-hf817b99_2.tar.bz2
       - conda: https://conda.anaconda.org/conda-forge/linux-64/libnghttp2-1.51.0-hdcd2b5c_0.conda
       - conda: https://conda.anaconda.org/conda-forge/linux-64/libnsl-2.0.1-hd590300_0.conda
@@ -101,6 +102,7 @@ environments:
       - conda: https://conda.anaconda.org/conda-forge/linux-64/libxkbcommon-1.0.3-he3ba5ed_0.tar.bz2
       - conda: https://conda.anaconda.org/conda-forge/linux-64/libxml2-2.9.14-haae042b_4.conda
       - conda: https://conda.anaconda.org/conda-forge/linux-64/libzlib-1.2.13-hd590300_5.conda
+      - conda: https://conda.anaconda.org/conda-forge/linux-64/llvmlite-0.38.1-py39h7d9a04d_0.tar.bz2
       - conda: https://conda.anaconda.org/conda-forge/linux-64/markupsafe-2.1.5-py39hd1e30aa_0.conda
       - conda: https://conda.anaconda.org/conda-forge/linux-64/matplotlib-3.5.3-py39hf3d152e_2.tar.bz2
       - conda: https://conda.anaconda.org/conda-forge/linux-64/matplotlib-base-3.5.3-py39h19d6b11_2.tar.bz2
@@ -111,6 +113,7 @@ environments:
       - conda: https://conda.anaconda.org/conda-forge/linux-64/ncurses-6.4.20240210-h59595ed_0.conda
       - conda: https://conda.anaconda.org/conda-forge/linux-64/nspr-4.35-h27087fc_0.conda
       - conda: https://conda.anaconda.org/conda-forge/linux-64/nss-3.98-h1d7d5a4_0.conda
+      - conda: https://conda.anaconda.org/conda-forge/linux-64/numba-0.55.2-py39h66db6d7_0.tar.bz2
       - conda: https://conda.anaconda.org/conda-forge/linux-64/numpy-1.19.5-py39hd249d9e_3.tar.bz2
       - conda: https://conda.anaconda.org/conda-forge/linux-64/openjpeg-2.5.0-hfec8fc6_2.conda
       - conda: https://conda.anaconda.org/conda-forge/linux-64/openssl-1.1.1w-hd590300_0.conda
@@ -2831,6 +2834,23 @@ packages:
   license_family: BSD
   size: 14350
   timestamp: 1700568424034
+- kind: conda
+  name: libllvm11
+  version: 11.1.0
+  build: he0ac6c6_5
+  build_number: 5
+  subdir: linux-64
+  url: https://conda.anaconda.org/conda-forge/linux-64/libllvm11-11.1.0-he0ac6c6_5.tar.bz2
+  sha256: fe02eb90fb3c9a2eb57c0b653d7630f7df72ee8b3093b0c3d1c34296e4e01134
+  md5: cae79c6fd61cc6823cbebdbb2c16c60e
+  depends:
+  - libgcc-ng >=12
+  - libstdcxx-ng >=12
+  - libzlib >=1.2.13,<1.3.0a0
+  license: Apache-2.0 WITH LLVM-exception
+  license_family: Apache
+  size: 30190183
+  timestamp: 1666875231224
 - kind: conda
   name: libllvm13
   version: 13.0.1
@@ -3223,6 +3243,26 @@ packages:
   license_family: Other
   size: 61588
   timestamp: 1686575217516
+- kind: conda
+  name: llvmlite
+  version: 0.38.1
+  build: py39h7d9a04d_0
+  subdir: linux-64
+  url: https://conda.anaconda.org/conda-forge/linux-64/llvmlite-0.38.1-py39h7d9a04d_0.tar.bz2
+  sha256: 3ab57078aa9fc90f0a7d13685c75919ea4cb72d38e32e25b23d2c7db5c9f2db1
+  md5: 6eed9f0a1f7c42eba45d9debb86232a5
+  depends:
+  - libgcc-ng >=10.3.0
+  - libllvm11 >=11.1.0,<11.2.0a0
+  - libstdcxx-ng >=10.3.0
+  - libzlib >=1.2.11,<1.3.0a0
+  - python >=3.9,<3.10.0a0
+  - python_abi 3.9.* *_cp39
+  - zlib >=1.2.11,<1.3.0a0
+  license: BSD-2-Clause
+  license_family: BSD
+  size: 2416751
+  timestamp: 1653080812454
 - kind: conda
   name: locket
   version: 1.0.0
@@ -3615,6 +3655,33 @@ packages:
   license_family: MOZILLA
   size: 2019716
   timestamp: 1708065114928
+- kind: conda
+  name: numba
+  version: 0.55.2
+  build: py39h66db6d7_0
+  subdir: linux-64
+  url: https://conda.anaconda.org/conda-forge/linux-64/numba-0.55.2-py39h66db6d7_0.tar.bz2
+  sha256: 5ff1cf67bfd2a49b07f63c2b458743d7b17ad6ca5c4d8ef2695d2ae9f1ea7439
+  md5: 82570db680a632f95911c4e0c9da7b76
+  depends:
+  - libgcc-ng >=12
+  - libstdcxx-ng >=12
+  - llvmlite >=0.38.1,<0.39.0a0
+  - numpy >=1.19.5,<2.0a0
+  - python >=3.9,<3.10.0a0
+  - python_abi 3.9.* *_cp39
+  - setuptools
+  constrains:
+  - tbb  2021.*
+  - scipy >=1.0
+  - cuda-python >=11.6
+  - libopenblas !=0.3.6
+  - numpy >=1.18,<1.23
+  - cudatoolkit >=9.2
+  license: BSD-2-Clause
+  license_family: BSD
+  size: 3975083
+  timestamp: 1655473455301
 - kind: conda
   name: numpy
   version: 1.19.5
diff --git a/unet_model/pixi.toml b/unet_model/pixi.toml
index 057c85e..7aa2fbb 100644
--- a/unet_model/pixi.toml
+++ b/unet_model/pixi.toml
@@ -19,3 +19,4 @@ scikit-learn = ">=1.1.2,<1.2"
 xarray = ">=2022.9.0,<2022.10"
 h5netcdf = ">=1.3.0,<1.4"
 scikit-image = ">=0.19.3,<0.20"
+numba = ">=0.55.2,<0.56"