{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "be0a4427",
   "metadata": {},
   "source": [
    "# Neural Architecture Search for Graph Neural Networks\n",
    "\n",
    "In this tutorial, we will design neural architecture search spaces for graph neural networks, specifically for message passing neural networks. \n",
    "\n",
    "For related papers, please check https://ieeexplore.ieee.org/abstract/document/9378060.\n",
    "\n",
    "The search space have multiple input tensors, including node features, edge features, edge pairs (source and target node indices) and node masks (number of nodes before zero-padding). The output of this search space is customizable. In this tutorial, we use an example from the QM7 dataset, which has a scalar output. The QM7 dataset is from the Deepchem library.\n",
    "\n",
    "There are two main variable nodes, namely `mpnn_cell` and `gather_cell`:\n",
    "* `mpnn_cell` is a message passing layer with a varierty of activation, aggregation, update functions, etc. \n",
    "* `gather_cell` is global graph gather layer with a variety of global pooling functions.\n",
    "\n",
    "We also adopted skip-connection in the search space; that is, the output of the *n-1* layer is the direct input to the *n+1* layer. Users can modify the number of `mpnn_cell` and maximum skip-connection distance to control the flexibility of skip-connection.\n",
    "\n",
    "We used random search and aging evolution (regularized evolution) to conduct architecture search. In the paper, we found aging evolution has good scalability. We also showed that the best architecture from the search outperforms the *moleculenet* benchmarks. Users are more than welcome to furthur modify the search space to boost the performance.\n",
    "\n",
    "## Install Deepchem and RDKit\n",
    "\n",
    "We need [DeepChem](https://deepchem.io) for the benchmark datasets. [RDKit](https://www.rdkit.org) is also required to convert molecule smile string to a graph representation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d64db5f6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: deepchem in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (2.5.0)\n",
      "Requirement already satisfied: joblib in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (from deepchem) (1.1.0)\n",
      "Requirement already satisfied: scipy in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (from deepchem) (1.7.2)\n",
      "Requirement already satisfied: pandas in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (from deepchem) (1.3.4)\n",
      "Requirement already satisfied: numpy in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (from deepchem) (1.21.4)\n",
      "Requirement already satisfied: scikit-learn in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (from deepchem) (1.0.1)\n",
      "Requirement already satisfied: python-dateutil>=2.7.3 in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (from pandas->deepchem) (2.8.2)\n",
      "Requirement already satisfied: pytz>=2017.3 in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (from pandas->deepchem) (2021.3)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (from scikit-learn->deepchem) (3.0.0)\n",
      "Requirement already satisfied: six>=1.5 in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (from python-dateutil>=2.7.3->pandas->deepchem) (1.15.0)\n",
      "Collecting package metadata (current_repodata.json): done\n",
      "Solving environment: done\n",
      "\n",
      "\n",
      "==> WARNING: A newer version of conda exists. <==\n",
      "  current version: 4.10.3\n",
      "  latest version: 4.11.0\n",
      "\n",
      "Please update conda by running\n",
      "\n",
      "    $ conda update -n base conda\n",
      "\n",
      "\n",
      "\n",
      "# All requested packages already installed.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "!pip install deepchem\n",
    "!conda install -c rdkit rdkit -y"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f26b8bd1",
   "metadata": {},
   "source": [
    "## Imports and GPU Detection <a class=\"anchor\" id=\"imports-and-gpu-detection\"></a>\n",
    "\n",
    "<div class=\"alert alert-warning\">\n",
    "\n",
    "<b>Warning</b>\n",
    "    \n",
    "By design asyncio does not allow nested event loops. Jupyter is using Tornado which already starts an event loop. Therefore the following patch is required to run this tutorial.\n",
    "    \n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "bb4474f0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: nest_asyncio in /Users/romainegele/miniforge3/envs/dh-arm/lib/python3.9/site-packages (1.5.1)\r\n"
     ]
    }
   ],
   "source": [
    "!pip install nest_asyncio\n",
    "\n",
    "import nest_asyncio\n",
    "nest_asyncio.apply()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "99a9ee9a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import os\n",
    "import pathlib\n",
    "import shutil\n",
    "\n",
    "!export TF_CPP_MIN_LOG_LEVEL=3\n",
    "!export TF_XLA_FLAGS=--tf_xla_enable_xla_devices\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38664b59",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-info\">\n",
    "    \n",
    "<b>Note</b>\n",
    "    \n",
    "The `TF_CPP_MIN_LOG_LEVEL` can be used to avoid the logging of Tensorflow *DEBUG*, *INFO* and *WARNING* statements.\n",
    "    \n",
    "</div>\n",
    "\n",
    "<div class=\"alert alert-info\">\n",
    "    \n",
    "<b>Note</b>\n",
    "    \n",
    "The following can be used to detect if **GPU** devices are available on the current host.\n",
    "    \n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "da5e0602",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No GPU available\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "available_gpus = tf.config.list_physical_devices(\"GPU\")\n",
    "n_gpus = len(available_gpus)\n",
    "\n",
    "if n_gpus > 1:\n",
    "    n_gpus -= 1\n",
    "is_gpu_available = n_gpus > 0\n",
    "\n",
    "if is_gpu_available:\n",
    "    print(f\"{n_gpus} GPU{'s are' if n_gpus > 1 else ' is'} available.\")\n",
    "else:\n",
    "    print(\"No GPU available\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c997c83",
   "metadata": {},
   "source": [
    "## Start Ray\n",
    "\n",
    "We launch the Ray run-time depending on the detected local ressources. If GPU(s) is(are) detected then 1 worker is started for each GPU. If not, then only 1 worker is started. You can start more workers by setting `num_cpus=1` to a value greater than 1.\n",
    "\n",
    "<div class=\"alert alert-warning\">\n",
    "\n",
    "<b>Warning</b>\n",
    "    \n",
    "In the case of GPUs it is important to follow this scheme to avoid multiple processes (Ray workers vs current process) to lock the same GPU.\n",
    "    \n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b522cd81",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-12-13 11:47:14,280\tINFO services.py:1338 -- View the Ray dashboard at \u001b[1m\u001b[32mhttp://127.0.0.1:8265\u001b[39m\u001b[22m\n"
     ]
    }
   ],
   "source": [
    "import ray\n",
    "\n",
    "log_to_driver = False\n",
    "\n",
    "if not(ray.is_initialized()):\n",
    "    if is_gpu_available:\n",
    "        ray.init(num_cpus=n_gpus, num_gpus=n_gpus, log_to_driver=log_to_driver)            \n",
    "    else:\n",
    "        ray.init(num_cpus=1, log_to_driver=log_to_driver)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c702c644",
   "metadata": {},
   "source": [
    "## Reformatting Graph Dataset <a class=\"anchor\" id=\"a-synthetic-dataset\"></a>\n",
    "\n",
    "Now, we will start by reformatting a benchmark QM7 from molecule-net. We will first generate data for a **training set** (used for estimation) and a **testing set** (used to evaluate the final performance). Then the training set will be sub-divided in a new **training set** (used to estimate the neural network weights) and **validation set** (used to estimate the neural network hyperparameters and architecture).\n",
    "\n",
    "The data is converted from a deepchem weave object to a list containing node features, edge features, edge pairs, node masks and GCN attention coefficients. You may check the details here https://ieeexplore.ieee.org/abstract/document/9378060"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9499cd94",
   "metadata": {},
   "outputs": [],
   "source": [
    "from deepchem.molnet import load_qm7\n",
    "from deephyper.contrib.mpnn import get_all_mol_feat\n",
    "\n",
    "\n",
    "def load_data(test_only=0, verbose=0):\n",
    "    _, (train_data, valid_data, test_data), _ = load_qm7(featurizer='Weave',\n",
    "                                                         splitter='random')\n",
    "\n",
    "    max_node, max_edge = [], [] # used to zero-pad the node and edge features to maximum dimension\n",
    "\n",
    "    for data in [train_data, valid_data, test_data]:\n",
    "        x = data.X\n",
    "        size = len(x) # number of molecules in a dataset\n",
    "        max_node.append(np.max([x[i].nodes.shape[0] for i in range(size)]))\n",
    "        max_edge.append(np.max([x[i].pairs.shape[0] for i in range(size)]))\n",
    "    max_node = np.max(max_node)\n",
    "    max_edge = np.max(max_edge)\n",
    "\n",
    "    x_train, y_train = get_all_mol_feat(train_data, max_node, max_edge)\n",
    "    x_valid, y_valid = get_all_mol_feat(valid_data, max_node, max_edge)\n",
    "    x_test, y_test = get_all_mol_feat(test_data, max_node, max_edge)\n",
    "        \n",
    "    if verbose:\n",
    "        print(f'x_train shape: {[x.shape for x in x_train]}')\n",
    "        print(f'y_train shape: {y_train.shape}')\n",
    "        print(f'x_valid shape: {[x.shape for x in x_valid]}')\n",
    "        print(f'y_valid shape: {y_valid.shape}')\n",
    "        print(f'x_test shape: {[x.shape for x in x_test]}')\n",
    "        print(f'y_test shape: {y_test.shape}')\n",
    "        \n",
    "    if test_only:\n",
    "        return (x_test, y_test)\n",
    "    else:\n",
    "        return (x_train, y_train), (x_valid, y_valid)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92bbf261",
   "metadata": {},
   "source": [
    "Then the code to split the training data in a new **training set** and a **validation set** corresponds to:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "16d780cb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape: [(5470, 22, 75), (5470, 506, 2), (5470, 506, 14), (5470, 22), (5470, 506)]\n",
      "y_train shape: (5470, 1)\n",
      "x_valid shape: [(684, 22, 75), (684, 506, 2), (684, 506, 14), (684, 22), (684, 506)]\n",
      "y_valid shape: (684, 1)\n",
      "x_test shape: [(684, 22, 75), (684, 506, 2), (684, 506, 14), (684, 22), (684, 506)]\n",
      "y_test shape: (684, 1)\n"
     ]
    }
   ],
   "source": [
    "# ignore warning/errors when downloading the dataset for the first time\n",
    "(x, y), (vx, vy) = load_data(verbose=1)\n",
    "(tx , ty) = load_data(test_only=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73734b1c",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-info\">\n",
    "    \n",
    "<b>Note</b>\n",
    "    \n",
    "When it is possible to factorize the two previous function into one, DeepHyper interface requires a function which returns `(train_inputs, train_outputs), (valid_inputs, valid_outputs)`.\n",
    "    \n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ceb5938",
   "metadata": {},
   "source": [
    "## Baseline Neural Network <a class=\"anchor\" id=\"baseline-neural-network\"></a>\n",
    "\n",
    "Let us define a baseline neural network based on a regular multi-layer perceptron architecture which learn the mean estimate and minimise the mean squared error."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "79a60573",
   "metadata": {},
   "outputs": [],
   "source": [
    "import deephyper.layers as dhl\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "996814d7",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-12-13 11:47:19.988355: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)\n",
      "2021-12-13 11:47:19.988498: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "43/43 [==============================] - 5s 106ms/step - loss: 1.0011 - val_loss: 1.0239\n",
      "Epoch 2/5\n",
      "43/43 [==============================] - 4s 102ms/step - loss: 0.9441 - val_loss: 0.9451\n",
      "Epoch 3/5\n",
      "43/43 [==============================] - 4s 105ms/step - loss: 0.9161 - val_loss: 0.9458\n",
      "Epoch 4/5\n",
      "43/43 [==============================] - 5s 105ms/step - loss: 0.8961 - val_loss: 0.9464\n",
      "Epoch 5/5\n",
      "43/43 [==============================] - 5s 108ms/step - loss: 0.9018 - val_loss: 0.9554\n"
     ]
    }
   ],
   "source": [
    "def gnn_default_model(shape):\n",
    "    node_ = tf.keras.layers.Input(shape[0])\n",
    "    adj_ = tf.keras.layers.Input(shape[1], dtype=tf.int32)\n",
    "    edge_ = tf.keras.layers.Input(shape[2])\n",
    "    mask_ = tf.keras.layers.Input(shape[3])\n",
    "    degree_ = tf.keras.layers.Input(shape[4])\n",
    "\n",
    "    input_ = [node_, adj_, edge_, mask_, degree_]\n",
    "\n",
    "    node = dhl.SparseMPNN(state_dim=32,\n",
    "                       T=1,\n",
    "                       aggr_method='max',\n",
    "                       attn_method='const',\n",
    "                       update_method='gru',\n",
    "                       attn_head=1,\n",
    "                       activation='elu')([node_, adj_, edge_, mask_, degree_])\n",
    "\n",
    "    node = dhl.GlobalAttentionPool(128)(node)\n",
    "    node = tf.keras.layers.Flatten()(node)\n",
    "    node = tf.keras.layers.Dense(32, activation='relu')(node)\n",
    "\n",
    "    output = tf.keras.layers.Dense(1, activation='linear')(node)\n",
    "\n",
    "    model = tf.keras.Model(input_, output)\n",
    "    \n",
    "    return model\n",
    "\n",
    "\n",
    "shape = [item.shape[1:] for item in x]\n",
    "model = gnn_default_model(shape)\n",
    "\n",
    "model.compile(optimizer=\"adam\", loss='mse')\n",
    "mc = ModelCheckpoint('gnn_model.h5', monitor='val_loss', mode='min', save_weights_only=True)\n",
    "\n",
    "history = model.fit(x, y, \n",
    "                    epochs=5, \n",
    "                    batch_size=128, \n",
    "                    validation_data=(vx, vy), \n",
    "                    verbose=1, \n",
    "                    callbacks=[mc]).history"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ee1d2d0",
   "metadata": {},
   "source": [
    "We can do a vizualisation of our learning curves to make sure the training and validation loss decrease correctly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "993871d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x355.995 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "width = 8\n",
    "height = width/1.618\n",
    "plt.figure(figsize=(width, height))\n",
    "\n",
    "plt.plot(history[\"loss\"], label=\"training\")\n",
    "plt.plot(history[\"val_loss\"], label=\"validation\")\n",
    "\n",
    "plt.xlabel(\"Epochs\")\n",
    "plt.ylabel(\"MSE\")\n",
    "\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8c375374",
   "metadata": {},
   "source": [
    "Also, let us look at the prediction on the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "11b53fe8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 324x324 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.load_weights('gnn_model.h5')\n",
    "pred_ty = model.predict(tx)\n",
    "\n",
    "width = 4.5\n",
    "height = width\n",
    "plt.figure(figsize=(width, height))\n",
    "\n",
    "vmin, vmax = -4, 4\n",
    "plt.scatter(ty, pred_ty)\n",
    "plt.plot([vmin, vmax], [vmin, vmax], 'k--')\n",
    "plt.xlim([vmin, vmax])\n",
    "plt.ylim([vmin, vmax])\n",
    "plt.xlabel('True')\n",
    "plt.ylabel('Predicted')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8743319",
   "metadata": {},
   "source": [
    "## Define the Neural Architecture Search Space <a class=\"anchor\" id=\"define-the-neural-architecture-search-space\"></a>\n",
    "\n",
    "The neural architecture search space is composed of discrete decision variables. For each decision variable we choose among a list of possible operation to perform (e.g., fully connected, ReLU). To define this search space, it is necessary to use two classes:\n",
    "\n",
    "* `KSearchSpace` (for Keras Search Space): represents a directed acyclic graph (DAG) in which each node represents a chosen operation. It represents the possible neural networks that can be created.\n",
    "* `SpaceFactory`: is a utilitiy class used to pack the logic of a search space definition and share it with others.\n",
    "\n",
    "Then, inside a `KSearchSpace` we will have two types of nodes:\n",
    "* `VariableNode`: corresponds to discrete decision variables and are used to define a list of possible operation.\n",
    "* `ConstantNode`: corresponds to fixed operation in the search space (e.g., input/outputs)\n",
    "\n",
    "Finally, it is possible to reuse any `tf.keras.layers` to define a `KSearchSpace`. However, it is important to wrap each layer in an `operation` to perform a lazy memory allocation of tensors."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42fd4eb3",
   "metadata": {},
   "source": [
    "We implement the constructor `__init__` and `build` method of the `RegressionSpace` a subclass of `KSearchSpace`. The `__init__` method interface is:\n",
    "\n",
    "```python\n",
    "def __init__(self, input_shape, output_shape, **kwargs):\n",
    "    ...\n",
    "```\n",
    "\n",
    "for the `build` method the interface is:\n",
    "\n",
    "```python\n",
    "def build(self):\n",
    "    ...\n",
    "    return self\n",
    "```\n",
    "\n",
    "where:\n",
    "* `input_shape` corresponds to a tuple or a list of tuple indicating the shapes of inputs tensors.\n",
    "* `output_shape` corresponds to the same but of output_tensors.\n",
    "* `**kwargs` denotes that any other key word argument can be defined by the user."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2f99f031",
   "metadata": {},
   "outputs": [],
   "source": [
    "import collections, itertools\n",
    "\n",
    "from deephyper.nas import KSearchSpace\n",
    "from deephyper.nas.node import ConstantNode, VariableNode\n",
    "from deephyper.nas.operation import operation, Zero, Connect, AddByProjecting, Identity\n",
    "\n",
    "# define operations\n",
    "Flatten = operation(tf.keras.layers.Flatten)\n",
    "Dense = operation(tf.keras.layers.Dense)\n",
    "\n",
    "SparseMPNN = operation(dhl.SparseMPNN)\n",
    "GlobalAttentionPool = operation(dhl.GlobalAttentionPool)\n",
    "GlobalAttentionSumPool = operation(dhl.GlobalAttentionSumPool)\n",
    "GlobalAvgPool = operation(dhl.GlobalAvgPool)\n",
    "GlobalMaxPool = operation(dhl.GlobalMaxPool)\n",
    "GlobalSumPool = operation(dhl.GlobalSumPool)\n",
    "\n",
    "class MPNNSpace(KSearchSpace):\n",
    "\n",
    "    def __init__(self, input_shape, output_shape, seed=None, num_layers=3):\n",
    "        super().__init__(input_shape, output_shape, seed=seed)\n",
    "        self.num_layers = 3\n",
    "\n",
    "    def build(self):\n",
    "        \n",
    "        source = prev_input = self.input_nodes[0]\n",
    "        prev_input1 = self.input_nodes[1]\n",
    "        prev_input2 = self.input_nodes[2]\n",
    "        prev_input3 = self.input_nodes[3]\n",
    "        prev_input4 = self.input_nodes[4]\n",
    "        \n",
    "        anchor_points = collections.deque([source], maxlen=3)\n",
    "        \n",
    "        for _ in range(self.num_layers):\n",
    "            graph_attn_cell = VariableNode()\n",
    "            self.mpnn_cell(graph_attn_cell)\n",
    "            self.connect(prev_input, graph_attn_cell)\n",
    "            self.connect(prev_input1, graph_attn_cell)\n",
    "            self.connect(prev_input2, graph_attn_cell)\n",
    "            self.connect(prev_input3, graph_attn_cell)\n",
    "            self.connect(prev_input4, graph_attn_cell)\n",
    "\n",
    "            merge = ConstantNode()\n",
    "            merge.set_op(AddByProjecting(self, [graph_attn_cell], activation=\"relu\"))\n",
    "            \n",
    "            for node in anchor_points:\n",
    "                skipco = VariableNode()\n",
    "                skipco.add_op(Zero())\n",
    "                skipco.add_op(Connect(self, node))\n",
    "                self.connect(skipco, merge)\n",
    "\n",
    "            prev_input = merge\n",
    "\n",
    "            anchor_points.append(prev_input)\n",
    "            \n",
    "        global_pooling_node = VariableNode()\n",
    "        self.gather_cell(global_pooling_node)\n",
    "        self.connect(prev_input, global_pooling_node)\n",
    "        prev_input = global_pooling_node\n",
    "\n",
    "        flatten_node = ConstantNode(Flatten())\n",
    "        self.connect(prev_input, flatten_node)\n",
    "        \n",
    "        # Output\n",
    "        output = ConstantNode(Dense(self.output_shape[0]))\n",
    "        self.connect(flatten_node, output)\n",
    "        \n",
    "        return self\n",
    "                \n",
    "    def mpnn_cell(self, node):\n",
    "        state_dims = [4, 8, 16, 32]\n",
    "        Ts = [1, 2, 3, 4]\n",
    "        attn_methods = [\"const\", \"gat\", \"sym-gat\", \"linear\", \"gen-linear\", \"cos\"]\n",
    "        attn_heads = [1, 2, 4, 6]\n",
    "        aggr_methods = [\"max\", \"mean\", \"sum\"]\n",
    "        update_methods = [\"gru\", \"mlp\"]\n",
    "        activations = [tf.keras.activations.sigmoid,\n",
    "                       tf.keras.activations.tanh,\n",
    "                       tf.keras.activations.relu,\n",
    "                       tf.keras.activations.linear,\n",
    "                       tf.keras.activations.elu,\n",
    "                       tf.keras.activations.softplus,\n",
    "                       tf.nn.leaky_relu,\n",
    "                       tf.nn.relu6]\n",
    "\n",
    "        for hp in itertools.product(state_dims,\n",
    "                                    Ts,\n",
    "                                    attn_methods,\n",
    "                                    attn_heads,\n",
    "                                    aggr_methods,\n",
    "                                    update_methods,\n",
    "                                    activations):\n",
    "            (state_dim, T, attn_method, attn_head, aggr_method, update_method, activation) = hp\n",
    "            \n",
    "            node.add_op(\n",
    "                SparseMPNN(\n",
    "                    state_dim=state_dim,\n",
    "                    T=T,\n",
    "                    attn_method=attn_method,\n",
    "                    attn_head=attn_head,\n",
    "                    aggr_method=aggr_method,\n",
    "                    update_method=update_method,\n",
    "                    activation=activation\n",
    "                )\n",
    "            )\n",
    "            \n",
    "            \n",
    "    def gather_cell(self, node):\n",
    "        for functions in [GlobalSumPool, GlobalMaxPool, GlobalAvgPool]:\n",
    "            for axis in [-1, -2]:\n",
    "                node.add_op(functions(axis=axis))\n",
    "        node.add_op(Flatten())\n",
    "        \n",
    "        for state_dim in [16, 32, 64]:\n",
    "            node.add_op(GlobalAttentionPool(state_dim=state_dim))\n",
    "        node.add_op(GlobalAttentionSumPool())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "26483b5a",
   "metadata": {},
   "source": [
    "Let us visualize a few randomly sampled neural architecture from this search space."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "10b57103",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[(0, 18431), (0, 1), (0, 18431), (0, 1), (0, 1), (0, 18431), (0, 1), (0, 1), (0, 1), (0, 10)]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.image as mpimg\n",
    "from tensorflow.keras.utils import plot_model\n",
    "\n",
    "input_shape = [item.shape[1:] for item in x]\n",
    "output_shape = y.shape[1:]\n",
    "shapes = dict(input_shape=input_shape, output_shape=output_shape)\n",
    "space = MPNNSpace(**shapes).build()\n",
    "\n",
    "print(space.choices())\n",
    "    \n",
    "\n",
    "images = []\n",
    "plt.figure(figsize=(15,15))\n",
    "for i in range(4):\n",
    "    \n",
    "    model = space.sample()\n",
    "    plt.subplot(2,2,i+1)\n",
    "    plot_model(model, \"random_model.png\", show_shapes=False, show_layer_names=False)\n",
    "    image = mpimg.imread(\"random_model.png\")\n",
    "    plt.imshow(image)\n",
    "    plt.axis('off')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "265a3a44",
   "metadata": {},
   "source": [
    "## Define the Neural Architecture Optimization Problem <a class=\"anchor\" id=\"define-the-neural-architecture-search-optimization-problem\"></a>\n",
    "\n",
    "In order to define a neural architecture search problem we have to use the `NaProblem` class. This class gives access to different method for the user to customize the training settings of neural networks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "ea447c6a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Problem is:\n",
       "    - search space   : __main__.MPNNSpace\n",
       "    - data loading   : __main__.load_data\n",
       "    - preprocessing  : None\n",
       "    - hyperparameters: \n",
       "        * verbose: 0\n",
       "        * batch_size: 128\n",
       "        * learning_rate: 0.001\n",
       "        * optimizer: adam\n",
       "        * num_epochs: 1\n",
       "    - loss           : mse\n",
       "    - metrics        : \n",
       "    - objective      : -val_loss"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from deephyper.problem import NaProblem\n",
    "\n",
    "problem = NaProblem()\n",
    "\n",
    "# Bind a function which returns (train_input, train_output), (valid_input, valid_output)\n",
    "problem.load_data(load_data)\n",
    "\n",
    "# Bind a function which returns a search space and give some arguments for the `build` method\n",
    "problem.search_space(MPNNSpace, num_layers=2)\n",
    "\n",
    "# Define a set of fixed hyperparameters for all trained neural networks\n",
    "problem.hyperparameters(\n",
    "    batch_size=128,\n",
    "    learning_rate=1e-3,\n",
    "    optimizer=\"adam\",\n",
    "    num_epochs=1, # lower fidelity\n",
    ")\n",
    "\n",
    "# Define the loss to minimize\n",
    "problem.loss(\"mse\")\n",
    "\n",
    "# Define complementary metrics\n",
    "problem.metrics([])\n",
    "\n",
    "# Define the maximized objective. Here we take the negative of the validation loss.\n",
    "problem.objective(\"-val_loss\")\n",
    "\n",
    "problem"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea800ab1",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-info\">\n",
    "    \n",
    "<b>Tip</b>\n",
    "    \n",
    "Adding an `EarlyStopping(...)` callback is a good idea to stop the training of your model as soon as it stops to improve.\n",
    "\n",
    "```python\n",
    "...\n",
    "EarlyStopping=dict(monitor=\"val_loss\", mode=\"min\", verbose=0, patience=3)\n",
    "...\n",
    "```\n",
    "    \n",
    "</div>\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4161c7b",
   "metadata": {},
   "source": [
    "## Define the Evaluator Object\n",
    "\n",
    "The `Evaluator` object is responsible of defining the backend used to distribute the function evaluation in DeepHyper."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "e798a7c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "from deephyper.evaluator import Evaluator\n",
    "from deephyper.evaluator.callback import LoggerCallback\n",
    "\n",
    "\n",
    "def get_evaluator(run_function):\n",
    "    \n",
    "    # Default arguments for Ray: 1 worker and 1 worker per evaluation\n",
    "    method_kwargs = {\n",
    "        \"num_cpus\": 1, \n",
    "        \"num_cpus_per_task\": 1,\n",
    "        \"callbacks\": [LoggerCallback()] # To interactively follow the finished evaluations,\n",
    "    }\n",
    "\n",
    "    # If GPU devices are detected then it will create 'n_gpus' workers\n",
    "    # and use 1 worker for each evaluation\n",
    "    if is_gpu_available:\n",
    "        method_kwargs[\"num_cpus\"] = n_gpus\n",
    "        method_kwargs[\"num_gpus\"] = n_gpus\n",
    "        method_kwargs[\"num_cpus_per_task\"] = 1\n",
    "        method_kwargs[\"num_gpus_per_task\"] = 1\n",
    "\n",
    "    evaluator = Evaluator.create(\n",
    "        run_function, \n",
    "        method=\"ray\", \n",
    "        method_kwargs=method_kwargs\n",
    "    )\n",
    "    print(f\"Created new evaluator with {evaluator.num_workers} worker{'s' if evaluator.num_workers > 1 else ''} and config: {method_kwargs}\", )\n",
    "    \n",
    "    return evaluator"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85e31cdd",
   "metadata": {},
   "source": [
    "For neural architecture search a standard training pipeline is provided by the `run_base_trainer` function from the `deephyper.nas.run` module."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "5db3be52",
   "metadata": {},
   "outputs": [],
   "source": [
    "from deephyper.nas.run import run_base_trainer"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "845ab242",
   "metadata": {},
   "source": [
    "## Define and Run the Neural Architecture Search\n",
    "\n",
    "All search algorithms follow a similar interface. A `problem` and `evaluator` object has to be provided to the search then the search can be executed through the `search(max_evals, timeout)` method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "11781742",
   "metadata": {},
   "outputs": [],
   "source": [
    "results = {} # used to store the results of different search algorithms\n",
    "max_evals = 10 # maximum number of iteratins for all searches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "1dc5ff9c",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Created new evaluator with 1 worker and config: {'num_cpus': 1, 'num_cpus_per_task': 1, 'callbacks': [<deephyper.evaluator.callback.LoggerCallback object at 0x293060100>]}\n",
      "[00001] -- best objective: -1.01752 -- received objective: -1.01752\n",
      "[00002] -- best objective: -1.01752 -- received objective: -1.04781\n",
      "[00003] -- best objective: -1.00011 -- received objective: -1.00011\n",
      "[00004] -- best objective: -0.92881 -- received objective: -0.92881\n",
      "[00005] -- best objective: -0.92881 -- received objective: -1.04169\n",
      "[00006] -- best objective: -0.92881 -- received objective: -1.07058\n",
      "[00007] -- best objective: -0.92881 -- received objective: -1.27250\n",
      "[00008] -- best objective: -0.92881 -- received objective: -1.07548\n",
      "[00009] -- best objective: -0.92881 -- received objective: -0.99138\n",
      "[00010] -- best objective: -0.92881 -- received objective: -0.99931\n"
     ]
    }
   ],
   "source": [
    "from deephyper.search.nas import Random\n",
    "\n",
    "random_search = Random(problem, get_evaluator(run_base_trainer), log_dir=\"random_search\")\n",
    "\n",
    "results[\"random\"] = random_search.search(max_evals=max_evals)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df8f3d31",
   "metadata": {},
   "source": [
    "If we look at the dataframe of results for each search we will find it slightly different than the one of hyperparameter search. A new column `arch_seq` corresponds to an embedding for each evaluated architecture. Each integer of an `arch_seq` list corresponds to the choice of a `VariableNode` in our `KSearchSpace`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "5c52c9c3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>arch_seq</th>\n",
       "      <th>id</th>\n",
       "      <th>objective</th>\n",
       "      <th>elapsed_sec</th>\n",
       "      <th>duration</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[13876, 0, 2160, 0, 0, 4762, 0, 1, 1, 1]</td>\n",
       "      <td>1</td>\n",
       "      <td>-1.017522</td>\n",
       "      <td>44.933679</td>\n",
       "      <td>43.074681</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[5937, 0, 16471, 1, 0, 9397, 1, 1, 0, 2]</td>\n",
       "      <td>2</td>\n",
       "      <td>-1.047807</td>\n",
       "      <td>281.613137</td>\n",
       "      <td>236.678758</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[6123, 1, 1777, 1, 0, 9158, 0, 1, 1, 7]</td>\n",
       "      <td>3</td>\n",
       "      <td>-1.000114</td>\n",
       "      <td>339.803210</td>\n",
       "      <td>58.188634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[5796, 1, 18368, 1, 1, 15679, 0, 1, 1, 6]</td>\n",
       "      <td>4</td>\n",
       "      <td>-0.928806</td>\n",
       "      <td>579.374068</td>\n",
       "      <td>239.570209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[7159, 1, 1494, 0, 0, 9234, 0, 1, 1, 4]</td>\n",
       "      <td>5</td>\n",
       "      <td>-1.041694</td>\n",
       "      <td>611.826204</td>\n",
       "      <td>32.451197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[4662, 0, 7258, 0, 0, 2419, 1, 1, 1, 5]</td>\n",
       "      <td>6</td>\n",
       "      <td>-1.070579</td>\n",
       "      <td>658.328865</td>\n",
       "      <td>46.501993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>[5898, 0, 12026, 1, 0, 4872, 1, 0, 1, 7]</td>\n",
       "      <td>7</td>\n",
       "      <td>-1.272499</td>\n",
       "      <td>710.819948</td>\n",
       "      <td>52.490447</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>[6942, 1, 1192, 0, 0, 5577, 0, 1, 1, 5]</td>\n",
       "      <td>8</td>\n",
       "      <td>-1.075484</td>\n",
       "      <td>725.114781</td>\n",
       "      <td>14.294140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>[10451, 1, 12804, 1, 0, 5329, 1, 0, 1, 10]</td>\n",
       "      <td>9</td>\n",
       "      <td>-0.991375</td>\n",
       "      <td>796.194318</td>\n",
       "      <td>71.078856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>[6617, 1, 2963, 0, 0, 6786, 1, 1, 0, 3]</td>\n",
       "      <td>10</td>\n",
       "      <td>-0.999313</td>\n",
       "      <td>824.445577</td>\n",
       "      <td>28.250628</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     arch_seq  id  objective  elapsed_sec  \\\n",
       "0    [13876, 0, 2160, 0, 0, 4762, 0, 1, 1, 1]   1  -1.017522    44.933679   \n",
       "1    [5937, 0, 16471, 1, 0, 9397, 1, 1, 0, 2]   2  -1.047807   281.613137   \n",
       "2     [6123, 1, 1777, 1, 0, 9158, 0, 1, 1, 7]   3  -1.000114   339.803210   \n",
       "3   [5796, 1, 18368, 1, 1, 15679, 0, 1, 1, 6]   4  -0.928806   579.374068   \n",
       "4     [7159, 1, 1494, 0, 0, 9234, 0, 1, 1, 4]   5  -1.041694   611.826204   \n",
       "5     [4662, 0, 7258, 0, 0, 2419, 1, 1, 1, 5]   6  -1.070579   658.328865   \n",
       "6    [5898, 0, 12026, 1, 0, 4872, 1, 0, 1, 7]   7  -1.272499   710.819948   \n",
       "7     [6942, 1, 1192, 0, 0, 5577, 0, 1, 1, 5]   8  -1.075484   725.114781   \n",
       "8  [10451, 1, 12804, 1, 0, 5329, 1, 0, 1, 10]   9  -0.991375   796.194318   \n",
       "9     [6617, 1, 2963, 0, 0, 6786, 1, 1, 0, 3]  10  -0.999313   824.445577   \n",
       "\n",
       "     duration  \n",
       "0   43.074681  \n",
       "1  236.678758  \n",
       "2   58.188634  \n",
       "3  239.570209  \n",
       "4   32.451197  \n",
       "5   46.501993  \n",
       "6   52.490447  \n",
       "7   14.294140  \n",
       "8   71.078856  \n",
       "9   28.250628  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results[\"random\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f0bc94ea",
   "metadata": {},
   "source": [
    "Let us visualize the best architecture found:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "7c32a210",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "i_max = results[\"random\"].objective.argmax()\n",
    "best_score = results[\"random\"].iloc[i_max].objective\n",
    "best_arch_seq = json.loads(results[\"random\"].iloc[i_max].arch_seq)\n",
    "\n",
    "best_model = space.sample(best_arch_seq)\n",
    "plot_model(best_model, show_shapes=True, show_layer_names=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "468e2559",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "43/43 [==============================] - 91s 2s/step - loss: 0.9519 - val_loss: 0.9487\n",
      "Epoch 2/50\n",
      "43/43 [==============================] - 93s 2s/step - loss: 0.8702 - val_loss: 0.8870\n",
      "Epoch 3/50\n",
      "43/43 [==============================] - 85s 2s/step - loss: 0.8189 - val_loss: 0.8192\n",
      "Epoch 4/50\n",
      "43/43 [==============================] - 84s 2s/step - loss: 0.7917 - val_loss: 0.7964\n",
      "Epoch 5/50\n",
      "43/43 [==============================] - 82s 2s/step - loss: 0.7909 - val_loss: 0.7834\n",
      "Epoch 6/50\n",
      "43/43 [==============================] - 84s 2s/step - loss: 0.7709 - val_loss: 0.7724\n",
      "Epoch 7/50\n",
      "43/43 [==============================] - 84s 2s/step - loss: 0.7605 - val_loss: 0.7746\n",
      "Epoch 8/50\n",
      "43/43 [==============================] - 83s 2s/step - loss: 0.7524 - val_loss: 0.7662\n",
      "Epoch 9/50\n",
      "43/43 [==============================] - 84s 2s/step - loss: 0.7487 - val_loss: 0.7601\n",
      "Epoch 10/50\n",
      "43/43 [==============================] - 84s 2s/step - loss: 0.7488 - val_loss: 0.7609\n",
      "Epoch 11/50\n",
      "43/43 [==============================] - 83s 2s/step - loss: 0.7414 - val_loss: 0.7677\n",
      "Epoch 12/50\n",
      "43/43 [==============================] - 82s 2s/step - loss: 0.7424 - val_loss: 0.7504\n",
      "Epoch 13/50\n",
      "43/43 [==============================] - 81s 2s/step - loss: 0.7314 - val_loss: 0.7356\n",
      "Epoch 14/50\n",
      "43/43 [==============================] - 81s 2s/step - loss: 0.7280 - val_loss: 0.7458\n",
      "Epoch 15/50\n",
      "43/43 [==============================] - 81s 2s/step - loss: 0.7263 - val_loss: 0.7381\n",
      "Epoch 16/50\n",
      "43/43 [==============================] - 83s 2s/step - loss: 0.7241 - val_loss: 0.7443\n",
      "Epoch 17/50\n",
      "43/43 [==============================] - 83s 2s/step - loss: 0.7212 - val_loss: 0.7343\n",
      "Epoch 18/50\n",
      "43/43 [==============================] - 83s 2s/step - loss: 0.7138 - val_loss: 0.7494\n",
      "Epoch 19/50\n",
      "43/43 [==============================] - 87s 2s/step - loss: 0.7073 - val_loss: 0.7442\n",
      "Epoch 20/50\n",
      "43/43 [==============================] - 84s 2s/step - loss: 0.7147 - val_loss: 0.7518\n",
      "Epoch 21/50\n",
      "43/43 [==============================] - 101s 2s/step - loss: 0.7149 - val_loss: 0.7477\n",
      "Epoch 22/50\n",
      "43/43 [==============================] - 108s 3s/step - loss: 0.7036 - val_loss: 0.7551\n",
      "Epoch 23/50\n",
      "43/43 [==============================] - 96s 2s/step - loss: 0.6998 - val_loss: 0.7245\n",
      "Epoch 24/50\n",
      "43/43 [==============================] - 89s 2s/step - loss: 0.6934 - val_loss: 0.7366\n",
      "Epoch 25/50\n",
      "43/43 [==============================] - 110s 3s/step - loss: 0.6904 - val_loss: 0.7511\n",
      "Epoch 26/50\n",
      "43/43 [==============================] - 90s 2s/step - loss: 0.6902 - val_loss: 0.7313\n",
      "Epoch 27/50\n",
      "43/43 [==============================] - 121s 3s/step - loss: 0.6898 - val_loss: 0.7531\n",
      "Epoch 28/50\n",
      "43/43 [==============================] - 103s 2s/step - loss: 0.6891 - val_loss: 0.7624\n",
      "Epoch 29/50\n",
      "43/43 [==============================] - 96s 2s/step - loss: 0.6809 - val_loss: 0.7437\n",
      "Epoch 30/50\n",
      "43/43 [==============================] - 169s 4s/step - loss: 0.6823 - val_loss: 0.7417\n",
      "Epoch 31/50\n",
      "43/43 [==============================] - 90s 2s/step - loss: 0.6784 - val_loss: 0.7530\n",
      "Epoch 32/50\n",
      "43/43 [==============================] - 142s 2s/step - loss: 0.6726 - val_loss: 0.7621\n",
      "Epoch 33/50\n",
      "43/43 [==============================] - 130s 3s/step - loss: 0.6721 - val_loss: 0.7374\n",
      "Epoch 34/50\n",
      "43/43 [==============================] - 95s 2s/step - loss: 0.6696 - val_loss: 0.7525\n",
      "Epoch 35/50\n",
      "43/43 [==============================] - 125s 3s/step - loss: 0.6667 - val_loss: 0.7607\n",
      "Epoch 36/50\n",
      "43/43 [==============================] - 90s 2s/step - loss: 0.6716 - val_loss: 0.7414\n",
      "Epoch 37/50\n",
      "43/43 [==============================] - 132s 2s/step - loss: 0.6580 - val_loss: 0.7319\n",
      "Epoch 38/50\n",
      "43/43 [==============================] - 129s 3s/step - loss: 0.6563 - val_loss: 0.7527\n",
      "Epoch 39/50\n",
      "43/43 [==============================] - 97s 2s/step - loss: 0.6595 - val_loss: 0.7719\n",
      "Epoch 40/50\n",
      "43/43 [==============================] - 97s 2s/step - loss: 0.6614 - val_loss: 0.7665\n",
      "Epoch 41/50\n",
      "43/43 [==============================] - 83s 2s/step - loss: 0.6496 - val_loss: 0.7620\n",
      "Epoch 42/50\n",
      "20/43 [============>.................] - ETA: 42s - loss: 0.6322"
     ]
    }
   ],
   "source": [
    "best_model.compile(optimizer=\"adam\", loss='mse')\n",
    "mc = ModelCheckpoint('gnn_best_model.h5', monitor='val_loss', mode='min', save_weights_only=True)\n",
    "\n",
    "history_best = best_model.fit(x, y, \n",
    "                    epochs=50, \n",
    "                    batch_size=128, \n",
    "                    validation_data=(vx, vy), \n",
    "                    verbose=1, \n",
    "                    callbacks=[mc]).history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4e7b4ffc",
   "metadata": {},
   "outputs": [],
   "source": [
    "width = 8\n",
    "height = width/1.618\n",
    "plt.figure(figsize=(width, height))\n",
    "\n",
    "plt.plot(history[\"val_loss\"], label=\"Default Model\")\n",
    "plt.plot(history_best[\"val_loss\"], label=\"Best Model\")\n",
    "\n",
    "plt.xlabel(\"Epochs\")\n",
    "plt.ylabel(\"MSE\")\n",
    "\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "de2cb2c8",
   "metadata": {},
   "outputs": [],
   "source": [
    "best_model.load_weights('gnn_best_model.h5')\n",
    "pred_ty = best_model.predict(tx)\n",
    "\n",
    "width = 4.5\n",
    "height = width\n",
    "plt.figure(figsize=(width, height))\n",
    "\n",
    "vmin, vmax = -4, 4\n",
    "plt.scatter(ty, pred_ty)\n",
    "plt.plot([vmin, vmax], [vmin, vmax], 'k--')\n",
    "plt.xlim([vmin, vmax])\n",
    "plt.ylim([vmin, vmax])\n",
    "plt.xlabel('True')\n",
    "plt.ylabel('Predicted')\n",
    "\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "357.0375061035156px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}