{"cells":[{"cell_type":"markdown","source":["#MVD 2023 Nnet seminario. Argimiro Arratia\n","\n","VArios experimentos de análisis de dato con redes neuronales en Pytorch o JAX\n","\n","(En construcción)"],"metadata":{"id":"EYH-GSEJSNVl"}},{"cell_type":"code","source":["import numpy as np\n","import pandas as pd\n","import matplotlib.pyplot as plt\n","import seaborn as sns\n","import statsmodels.api as sm\n","#from sklearn.gaussian_process import GaussianProcessRegressor\n","#from sklearn.gaussian_process.kernels import RBF, ConstantKernel as C\n","from sklearn import metrics\n","from sklearn.preprocessing import StandardScaler\n","from tabulate import tabulate\n","from sklearn.metrics import mean_squared_error\n","from time import time\n","\n","import torch\n","import torch.nn as nn\n","import torch.optim as optim\n","\n","def nrmse(y_actual, y_predicted):\n","    return np.sqrt(mean_squared_error(y_actual, y_predicted) / np.mean(np.sum((y_actual - np.mean(y_actual)) ** 2)))\n","\n","def pcorrect(y_actual, y_predicted):\n","    return (1 - nrmse(y_actual, y_predicted)) * 100\n"],"metadata":{"id":"bUhMfIR0N1HS"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# Read data\n","sp500 = pd.read_csv('SP500_shiller.csv', index_col=0, parse_dates=True)\n","data = sp500.loc['1930-01-01':'2012-12-31']\n","\n","# ... Data preprocessing (tau, target, sp500PE) ...\n","# Frequency of sampling\n","tau = 1 #data is monthly. Try tau=12 (year), tau=3 (quarterly)\n","\n","# Target and Feature as plain Price\n","target = data['SP500']\n","sp500PE = data['P/E10'].dropna() ##feature:P/E MA(10)\n","# sp500PE.head()\n","\n","# Convert Target and Feature as Returns\n","sp500PE = np.log(sp500PE).diff(tau)\n","target = np.log(target).diff(tau)\n","# try non-center (above) /center target (below)\n","target = target - target.mean() ##Center the target\n","target = target.dropna()"],"metadata":{"id":"c7-IvAuPN3ex"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# Model Inputs:\n","# Define matrix of features (each column is a feature)\n","# Features: lags 1,2,3 with (or w/o) PE10 and lags 1,2\n","feat = pd.concat([target.shift(1), target.shift(2), target.shift(3),\n","                 sp500PE, sp500PE.shift(1), sp500PE.shift(2),\n","                  # add other features here,\n","                  target], axis=1)\n","feat.columns = ['lag.1', 'lag.2', 'lag.3',\n","                'PE10', 'PE10.1', 'PE10.2',\n","                # names of other features,\n","                'TARGET']\n","feat = feat.dropna()\n","\n","# Divide data into training (75%) and testing (25%).\n","T = len(feat)\n","p = 0.75\n","trainindex = round(p*T)\n"],"metadata":{"id":"L0KUSUECOGhS"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["feat.head()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":238},"id":"FRbJr426eoLz","executionInfo":{"status":"ok","timestamp":1693751887031,"user_tz":180,"elapsed":445,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"af5e0c3a-d509-4b9b-89da-c17d317777d6"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["               lag.1     lag.2     lag.3      PE10    PE10.1    PE10.2  \\\n","Date                                                                     \n","1930-05-01  0.057355  0.032814  0.056557 -0.061036  0.049584  0.036865   \n","1930-06-01 -0.065761  0.057355  0.032814 -0.105772 -0.061036  0.049584   \n","1930-07-01 -0.110771 -0.065761  0.057355 -0.014740 -0.105772 -0.061036   \n","1930-08-01 -0.025810 -0.110771 -0.065761 -0.011669 -0.014740 -0.105772   \n","1930-09-01 -0.017107 -0.025810 -0.110771 -0.010857 -0.011669 -0.014740   \n","\n","              TARGET  \n","Date                  \n","1930-05-01 -0.065761  \n","1930-06-01 -0.110771  \n","1930-07-01 -0.025810  \n","1930-08-01 -0.017107  \n","1930-09-01 -0.004684  "],"text/html":["\n","  <div id=\"df-8d6e166f-318c-4f9d-a697-96ce554d08eb\" class=\"colab-df-container\">\n","    <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>lag.1</th>\n","      <th>lag.2</th>\n","      <th>lag.3</th>\n","      <th>PE10</th>\n","      <th>PE10.1</th>\n","      <th>PE10.2</th>\n","      <th>TARGET</th>\n","    </tr>\n","    <tr>\n","      <th>Date</th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>1930-05-01</th>\n","      <td>0.057355</td>\n","      <td>0.032814</td>\n","      <td>0.056557</td>\n","      <td>-0.061036</td>\n","      <td>0.049584</td>\n","      <td>0.036865</td>\n","      <td>-0.065761</td>\n","    </tr>\n","    <tr>\n","      <th>1930-06-01</th>\n","      <td>-0.065761</td>\n","      <td>0.057355</td>\n","      <td>0.032814</td>\n","      <td>-0.105772</td>\n","      <td>-0.061036</td>\n","      <td>0.049584</td>\n","      <td>-0.110771</td>\n","    </tr>\n","    <tr>\n","      <th>1930-07-01</th>\n","      <td>-0.110771</td>\n","      <td>-0.065761</td>\n","      <td>0.057355</td>\n","      <td>-0.014740</td>\n","      <td>-0.105772</td>\n","      <td>-0.061036</td>\n","      <td>-0.025810</td>\n","    </tr>\n","    <tr>\n","      <th>1930-08-01</th>\n","      <td>-0.025810</td>\n","      <td>-0.110771</td>\n","      <td>-0.065761</td>\n","      <td>-0.011669</td>\n","      <td>-0.014740</td>\n","      <td>-0.105772</td>\n","      <td>-0.017107</td>\n","    </tr>\n","    <tr>\n","      <th>1930-09-01</th>\n","      <td>-0.017107</td>\n","      <td>-0.025810</td>\n","      <td>-0.110771</td>\n","      <td>-0.010857</td>\n","      <td>-0.011669</td>\n","      <td>-0.014740</td>\n","      <td>-0.004684</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","    <div class=\"colab-df-buttons\">\n","\n","  <div class=\"colab-df-container\">\n","    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8d6e166f-318c-4f9d-a697-96ce554d08eb')\"\n","            title=\"Convert this dataframe to an interactive table.\"\n","            style=\"display:none;\">\n","\n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n","    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n","  </svg>\n","    </button>\n","\n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    .colab-df-buttons div {\n","      margin-bottom: 4px;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","    <script>\n","      const buttonEl =\n","        document.querySelector('#df-8d6e166f-318c-4f9d-a697-96ce554d08eb button.colab-df-convert');\n","      buttonEl.style.display =\n","        google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","      async function convertToInteractive(key) {\n","        const element = document.querySelector('#df-8d6e166f-318c-4f9d-a697-96ce554d08eb');\n","        const dataTable =\n","          await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                    [key], {});\n","        if (!dataTable) return;\n","\n","        const docLinkHtml = 'Like what you see? Visit the ' +\n","          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","          + ' to learn more about interactive tables.';\n","        element.innerHTML = '';\n","        dataTable['output_type'] = 'display_data';\n","        await google.colab.output.renderOutput(dataTable, element);\n","        const docLink = document.createElement('div');\n","        docLink.innerHTML = docLinkHtml;\n","        element.appendChild(docLink);\n","      }\n","    </script>\n","  </div>\n","\n","\n","<div id=\"df-00c31f31-c815-4bda-bd8e-44585de1f801\">\n","  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-00c31f31-c815-4bda-bd8e-44585de1f801')\"\n","            title=\"Suggest charts.\"\n","            style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","     width=\"24px\">\n","    <g>\n","        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n","    </g>\n","</svg>\n","  </button>\n","\n","<style>\n","  .colab-df-quickchart {\n","      --bg-color: #E8F0FE;\n","      --fill-color: #1967D2;\n","      --hover-bg-color: #E2EBFA;\n","      --hover-fill-color: #174EA6;\n","      --disabled-fill-color: #AAA;\n","      --disabled-bg-color: #DDD;\n","  }\n","\n","  [theme=dark] .colab-df-quickchart {\n","      --bg-color: #3B4455;\n","      --fill-color: #D2E3FC;\n","      --hover-bg-color: #434B5C;\n","      --hover-fill-color: #FFFFFF;\n","      --disabled-bg-color: #3B4455;\n","      --disabled-fill-color: #666;\n","  }\n","\n","  .colab-df-quickchart {\n","    background-color: var(--bg-color);\n","    border: none;\n","    border-radius: 50%;\n","    cursor: pointer;\n","    display: none;\n","    fill: var(--fill-color);\n","    height: 32px;\n","    padding: 0;\n","    width: 32px;\n","  }\n","\n","  .colab-df-quickchart:hover {\n","    background-color: var(--hover-bg-color);\n","    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n","    fill: var(--button-hover-fill-color);\n","  }\n","\n","  .colab-df-quickchart-complete:disabled,\n","  .colab-df-quickchart-complete:disabled:hover {\n","    background-color: var(--disabled-bg-color);\n","    fill: var(--disabled-fill-color);\n","    box-shadow: none;\n","  }\n","\n","  .colab-df-spinner {\n","    border: 2px solid var(--fill-color);\n","    border-color: transparent;\n","    border-bottom-color: var(--fill-color);\n","    animation:\n","      spin 1s steps(1) infinite;\n","  }\n","\n","  @keyframes spin {\n","    0% {\n","      border-color: transparent;\n","      border-bottom-color: var(--fill-color);\n","      border-left-color: var(--fill-color);\n","    }\n","    20% {\n","      border-color: transparent;\n","      border-left-color: var(--fill-color);\n","      border-top-color: var(--fill-color);\n","    }\n","    30% {\n","      border-color: transparent;\n","      border-left-color: var(--fill-color);\n","      border-top-color: var(--fill-color);\n","      border-right-color: var(--fill-color);\n","    }\n","    40% {\n","      border-color: transparent;\n","      border-right-color: var(--fill-color);\n","      border-top-color: var(--fill-color);\n","    }\n","    60% {\n","      border-color: transparent;\n","      border-right-color: var(--fill-color);\n","    }\n","    80% {\n","      border-color: transparent;\n","      border-right-color: var(--fill-color);\n","      border-bottom-color: var(--fill-color);\n","    }\n","    90% {\n","      border-color: transparent;\n","      border-bottom-color: var(--fill-color);\n","    }\n","  }\n","</style>\n","\n","  <script>\n","    async function quickchart(key) {\n","      const quickchartButtonEl =\n","        document.querySelector('#' + key + ' button');\n","      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n","      quickchartButtonEl.classList.add('colab-df-spinner');\n","      try {\n","        const charts = await google.colab.kernel.invokeFunction(\n","            'suggestCharts', [key], {});\n","      } catch (error) {\n","        console.error('Error during call to suggestCharts:', error);\n","      }\n","      quickchartButtonEl.classList.remove('colab-df-spinner');\n","      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n","    }\n","    (() => {\n","      let quickchartButtonEl =\n","        document.querySelector('#df-00c31f31-c815-4bda-bd8e-44585de1f801 button');\n","      quickchartButtonEl.style.display =\n","        google.colab.kernel.accessAllowed ? 'block' : 'none';\n","    })();\n","  </script>\n","</div>\n","    </div>\n","  </div>\n"]},"metadata":{},"execution_count":6}]},{"cell_type":"code","source":["# Split the data into features and target\n","training_features = feat.iloc[:trainindex, :-1].values\n","training_target = feat.iloc[:trainindex, -1].values\n","testing_features = feat.iloc[trainindex:, :-1].values\n","testing_target = feat.iloc[trainindex:, -1].values\n","\n","# Standardize features (optional but often beneficial for neural networks)\n","from sklearn.preprocessing import StandardScaler\n","#scaler = StandardScaler()\n","#training_features = scaler.fit_transform(training_features)\n","#testing_features = scaler.transform(testing_features)\n"],"metadata":{"id":"y-XMlhHrUNx6"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["display(training_features)\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":256},"id":"guy91cEExT4z","executionInfo":{"status":"ok","timestamp":1693751964101,"user_tz":180,"elapsed":344,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"12643201-e04e-4e56-cab5-282eba806bbf"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["array([[ 0.05735463,  0.03281436,  0.05655682, -0.06103589,  0.04958382,\n","         0.03686481],\n","       [-0.06576116,  0.05735463,  0.03281436, -0.10577195, -0.06103589,\n","         0.04958382],\n","       [-0.11077123, -0.06576116,  0.05735463, -0.01474002, -0.10577195,\n","        -0.06103589],\n","       ...,\n","       [ 0.06435537,  0.00248555, -0.00668774, -0.009657  ,  0.06964342,\n","         0.00816776],\n","       [-0.01269917,  0.06435537,  0.00248555, -0.01544035, -0.009657  ,\n","         0.06964342],\n","       [-0.0168876 , -0.01269917,  0.06435537,  0.00103681, -0.01544035,\n","        -0.009657  ]])"]},"metadata":{}}]},{"cell_type":"code","source":["display(training_target)"],"metadata":{"id":"UMthunl_2LgJ"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# Convert data to PyTorch tensors\n","training_features = torch.tensor(training_features, dtype=torch.float32)\n","training_target = torch.tensor(training_target, dtype=torch.float32)\n","testing_features = torch.tensor(testing_features, dtype=torch.float32)\n","testing_target = torch.tensor(testing_target, dtype=torch.float32)\n"],"metadata":{"id":"hMkbxwnhv5Vt"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["training_features.shape[1]\n","#training_features"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"DzzqMjNYOa1C","executionInfo":{"status":"ok","timestamp":1693752009718,"user_tz":180,"elapsed":403,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"0247414a-d94d-4740-a588-440c0e66b64c"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["6"]},"metadata":{},"execution_count":11}]},{"cell_type":"code","source":["training_features"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"xu4tYWNcf66-","executionInfo":{"status":"ok","timestamp":1693752341068,"user_tz":180,"elapsed":350,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"18ec2e0a-820e-46e8-d296-9e37c0f1ceaf"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["tensor([[ 0.0574,  0.0328,  0.0566, -0.0610,  0.0496,  0.0369],\n","        [-0.0658,  0.0574,  0.0328, -0.1058, -0.0610,  0.0496],\n","        [-0.1108, -0.0658,  0.0574, -0.0147, -0.1058, -0.0610],\n","        ...,\n","        [ 0.0644,  0.0025, -0.0067, -0.0097,  0.0696,  0.0082],\n","        [-0.0127,  0.0644,  0.0025, -0.0154, -0.0097,  0.0696],\n","        [-0.0169, -0.0127,  0.0644,  0.0010, -0.0154, -0.0097]])"]},"metadata":{},"execution_count":19}]},{"cell_type":"code","source":["training_target.view(-1, 1)\n","#training_target"],"metadata":{"id":"LaI6eE5qP87S"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["len(training_target)\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"sk5yNBoc3BKL","executionInfo":{"status":"ok","timestamp":1693682430811,"user_tz":180,"elapsed":271,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"1aa718d5-af4e-408f-d9f5-27be3abf0328"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["744"]},"metadata":{},"execution_count":43}]},{"cell_type":"code","source":["# Define a 1-layer neural network model with sigmoid or ReLU activation\n","class NeuralNetwork(nn.Module):\n","    def __init__(self, input_size, hidden_size1, output_size):\n","        super(NeuralNetwork, self).__init__()\n","        self.layer1 = nn.Linear(input_size, hidden_size1)\n","        self.layer_out = nn.Linear(hidden_size1, output_size)\n","\n","    def forward(self, x):\n","        x = torch.sigmoid(self.layer1(x))\n","        #x = torch.relu(self.layer1(x))\n","        #x = self.layer_out(x) ##output layer without activation\n","        x = torch.sigmoid(self.layer_out(x))\n","        return x"],"metadata":{"id":"Wv5l2V94UwT9"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# Define a 2-layer neural network model\n","class NeuralNetwork2(nn.Module):\n","    def __init__(self, input_size, hidden_size1, hidden_size2, output_size):\n","        super(NeuralNetwork2, self).__init__()\n","        self.layer1 = nn.Linear(input_size, hidden_size1)\n","        self.layer2 = nn.Linear(hidden_size1, hidden_size2)\n","        self.layer_out = nn.Linear(hidden_size2, output_size)\n","\n","    def forward(self, x):\n","        x = torch.relu(self.layer1(x))\n","        x = torch.relu(self.layer2(x))\n","        x = self.layer_out(x)\n","        #x = torch.sigmoid(self.layer_out(x))\n","        return x\n"],"metadata":{"id":"wE8j4h3Df1JU"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# Define Deep Net of adjustable depth\n","class MLPRegressor(nn.Module):\n","    def __init__(self, input_size, hidden_sizes, output_size):\n","        super(MLPRegressor, self).__init__()\n","        self.input_size = input_size\n","        self.hidden_sizes = hidden_sizes\n","        self.output_size = output_size\n","\n","        # Create the layers\n","        layers = []\n","        sizes = [input_size] + hidden_sizes + [output_size]\n","\n","        for i in range(len(sizes) - 1):\n","            layers.append(nn.Linear(sizes[i], sizes[i+1]))\n","            if i < len(sizes) - 2:\n","                layers.append(nn.ReLU())\n","\n","        self.model = nn.Sequential(*layers)\n","\n","    def forward(self, x):\n","        return self.model(x)"],"metadata":{"id":"acnMgFLpOzSM"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# Model parameters\n","size = 100 #200\n","input_size = training_features.shape[1]\n","#hidden_sizes = [size, size, size]\n","hidden_size1 = size\n","hidden_size2= size\n","output_size = 1\n","weight_decay = 10 ** -2\n","learning_rate = 0.01\n","\n","# Create the model\n","#model = MLPRegressor(input_size, hidden_sizes, output_size)\n","#model = NeuralNetwork(input_size, hidden_size1, output_size)\n","model = NeuralNetwork2(input_size, hidden_size1, hidden_size2, output_size)\n","criterion = nn.MSELoss()\n","optimizer = optim.Adam(model.parameters(), lr=learning_rate, weight_decay=weight_decay)\n","\n","# Training loop\n","max_iter = 5000 #10000\n","\n","start = time()\n","for iteration in range(max_iter):\n","    #loss = criterion(outputs, training_target)\n","\n","   # Forward pass\n","    outputs = model(training_features)\n","    loss = criterion(outputs, training_target.view(-1, 1))\n","\n","    # Backpropagation and optimization\n","    optimizer.zero_grad()\n","    loss.backward()\n","    optimizer.step()\n","\n","    if iteration % 1000 == 0:\n","        print(f\"Iteration {iteration}, Loss: {loss.item()}\")\n","\n","end = time()\n","print(\"Runtime for {0} updates = {1:.1f} seconds\".format(max_iter, end-start))\n"],"metadata":{"id":"bz8hiLQvHznw","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1693754105102,"user_tz":180,"elapsed":145023,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"eb344213-f675-4177-83a8-18c610f5339c"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Iteration 0, Loss: 0.0035902371164411306\n","Iteration 1000, Loss: 0.0022013678681105375\n","Iteration 2000, Loss: 0.0022013678681105375\n","Iteration 3000, Loss: 0.0022013678681105375\n","Iteration 4000, Loss: 0.0022013678681105375\n","Runtime for 5000 updates = 144.6 seconds\n"]}]},{"cell_type":"code","source":["# Evaluate the model on testing data\n","#with torch.no_grad():\n","#    pred_nnet1 = model(testing_features).numpy()\n","\n","# ... Rest of the code ...\n","# After training, you can use the trained model for predictions\n","with torch.no_grad():\n","    pred_nnet1 = model(testing_features)\n","\n","pred_nnet1 = pred_nnet1.numpy().flatten()  # Convert predictions to NumPy array\n","\n","# Evaluate the model\n","accuracy = pcorrect(pred_nnet1, testing_target.numpy().flatten())\n","print(f'Model Accuracy: {accuracy:.2f}%')\n","\n"],"metadata":{"id":"lP4g9fHJlP60","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1693687694734,"user_tz":180,"elapsed":314,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"6d0aa770-bc4c-495f-efd7-33f516f09eb2"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Model Accuracy: -inf%\n"]},{"output_type":"stream","name":"stderr","text":["<ipython-input-76-8f106156bf07>:19: RuntimeWarning: divide by zero encountered in float_scalars\n","  return np.sqrt(mean_squared_error(y_actual, y_predicted) / np.mean(np.sum((y_actual - np.mean(y_actual)) ** 2)))\n"]}]},{"cell_type":"code","source":["pred_nnet1"],"metadata":{"id":"IXxM8GteoBdx"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# Plot the results of nnetMul\n","plt.figure(figsize=(12, 8))\n","plt.plot(testing_target.numpy().flatten(), label='true', linestyle='--')\n","plt.plot(pred_nnet1, label='pred_nnetMul')\n","plt.legend()\n","plt.show()\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":544},"id":"TZlDp34wnkwl","executionInfo":{"status":"ok","timestamp":1693687709866,"user_tz":180,"elapsed":855,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"50c03909-4381-49f1-86f8-1aa308942fd5"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1200x800 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","source":[],"metadata":{"id":"RInLgODMN0HE"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["#Nnet with JAX\n"],"metadata":{"id":"tUxWtSW6cHDY"}},{"cell_type":"code","source":["import numpy as np\n","import pandas as pd\n","import matplotlib.pyplot as plt\n","import seaborn as sns\n","import statsmodels.api as sm\n","import jax\n","import jax.numpy as jnp\n","from jax import grad, jit, vmap\n","\n","# Custom implementation of NRMSE and PCorrect\n","def nrmse(y_actual, y_predicted):\n","    return jnp.sqrt(jnp.mean(jnp.square(y_actual - y_predicted)) / jnp.mean(jnp.square(y_actual - jnp.mean(y_actual))))\n","\n","def pcorrect(y_actual, y_predicted):\n","    return (1 - nrmse(y_actual, y_predicted)) * 100\n","\n","# Read data\n","sp500 = pd.read_csv('SP500_shiller.csv', index_col=0, parse_dates=True)\n","# Select data from 1925 to 2012\n","data = sp500.loc['1930-01-01':'2012-12-31']\n","data['SP500'].plot()\n","\n","# Frequency of sampling\n","tau = 1  # data is monthly. Try tau=12 (year), tau=3 (quarterly)\n","\n","# Option 1. Target and Feature as plain Price\n","target = data['SP500']\n","sp500PE = data['P/E10']  # feature: P/E MA(10)\n","\n","# Align time series data and compute features\n","aligned_data = pd.concat([target, sp500PE], axis=1).dropna()\n","target = aligned_data['SP500']\n","sp500PE = aligned_data['P/E10']\n","\n","# Option 2. Target and Feature as Returns\n","sp500PE = jnp.log(jnp.array(sp500PE))\n","sp500PE = jnp.diff(sp500PE, n=tau)  # Calculate differences\n","target = jnp.log(jnp.array(target))\n","target = jnp.diff(target, n=tau)  # Calculate differences\n","# try non-center (above) /center target (below)\n","target = target - jnp.mean(target)  # Center the target\n","\n","# Ensure all arrays have the same length\n","min_length = min(len(target), len(sp500PE))\n","target = target[:min_length]\n","sp500PE = sp500PE[:min_length]"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":451},"id":"mZZ7JojTqN_a","executionInfo":{"status":"ok","timestamp":1693671027019,"user_tz":180,"elapsed":1010,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"7e5cb395-8e44-4eac-c665-db7281a2b897"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","source":["# Model Inputs:\n","# Define matrix of features (each column is a feature)\n","# Features: lags 1,2,3 with (or w/o) PE10 and lags 1,2\n","feat = jnp.column_stack([\n","    target[:-3], target[1:-2], target[2:-1],\n","    sp500PE[2:-1], sp500PE[1:-2], sp500PE[:-3],\n","    # add other features here,\n","    target[3:]\n","])"],"metadata":{"id":"S_li4ov6s0PR"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["feat"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"AbEq_6WRqR44","executionInfo":{"status":"ok","timestamp":1693671889453,"user_tz":180,"elapsed":408,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"d98555c5-6e19-4909-a6fa-a95922d0423c"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["Array([[ 0.05655699,  0.03281432,  0.0573545 , ...,  0.03686476,\n","         0.06044006, -0.06576104],\n","       [ 0.03281432,  0.0573545 , -0.06576104, ...,  0.04958391,\n","         0.03686476, -0.11077113],\n","       [ 0.0573545 , -0.06576104, -0.11077113, ..., -0.06103611,\n","         0.04958391, -0.02581067],\n","       ...,\n","       [ 0.02285533,  0.02740722,  0.02387863, ...,  0.01934481,\n","         0.02167201, -0.00809044],\n","       [ 0.02740722,  0.02387863, -0.00809044, ...,  0.01760125,\n","         0.01934481, -0.03478856],\n","       [ 0.02387863, -0.00809044, -0.03478856, ..., -0.00968885,\n","         0.01760125,  0.01552206]], dtype=float32)"]},"metadata":{},"execution_count":27}]},{"cell_type":"code","source":["import numpy as np\n","import pandas as pd\n","import matplotlib.pyplot as plt\n","import seaborn as sns\n","import statsmodels.api as sm\n","import jax\n","import jax.numpy as jnp\n","from jax import grad, jit, vmap\n","\n","from tabulate import tabulate\n","\n","\n","# Custom implementation of NRMSE and PCorrect\n","def nrmse(y_actual, y_predicted):\n","    return jnp.sqrt(jnp.mean(jnp.square(y_actual - y_predicted)) / jnp.mean(jnp.square(y_actual - jnp.mean(y_actual))))\n","\n","def pcorrect(y_actual, y_predicted):\n","    return (1 - nrmse(y_actual, y_predicted)) * 100\n","\n","# Read data\n","sp500 = pd.read_csv('SP500_shiller.csv', index_col=0, parse_dates=True)\n","# Select data from 1925 to 2012\n","data = sp500.loc['1930-01-01':'2012-12-31']\n","data['SP500'].plot()\n","\n","# Frequency of sampling\n","tau = 1  # data is monthly. Try tau=12 (year), tau=3 (quarterly)\n","\n","# Option 1. Target and Feature as plain Price\n","target = data['SP500']\n","sp500PE = data['P/E10']  # feature: P/E MA(10)\n","\n","# Align time series data and compute features\n","aligned_data = pd.concat([target, sp500PE], axis=1).dropna()\n","target = aligned_data['SP500']\n","sp500PE = aligned_data['P/E10']\n","\n","# Option 2. Target and Feature as Returns\n","sp500PE = jnp.log(jnp.array(sp500PE))\n","sp500PE = jnp.diff(sp500PE, n=tau)  # Calculate differences\n","target = jnp.log(jnp.array(target))\n","target = jnp.diff(target, n=tau)  # Calculate differences\n","# try non-center (above) /center target (below)\n","target = target - jnp.mean(target)  # Center the target\n","\n","# Ensure all arrays have the same length\n","min_length = min(len(target), len(sp500PE))\n","target = target[:min_length]\n","sp500PE = sp500PE[:min_length]\n","\n","# Model Inputs:\n","# Define matrix of features (each column is a feature)\n","# Features: lags 1,2,3 with (or w/o) PE10 and lags 1,2\n","feat = jnp.column_stack([\n","    target[:-3], target[1:-2], target[2:-1],\n","    sp500PE[2:-1], sp500PE[1:-2], sp500PE[:-3],\n","    # add other features here,\n","    target[3:]\n","])\n","\n","# Divide data into training (75%) and testing (25%).\n","T = len(feat)\n","p = 0.75\n","trainindex = round(p * T)\n","\n","# Process arrays as JAX arrays\n","training_features = feat[:trainindex, :-1]\n","training_target = feat[:trainindex, -1]\n","\n","testing_features = feat[trainindex:, :-1]\n","testing_target = feat[trainindex:, -1]\n","\n","# Define a simple neural network using JAX and NumPy\n","def neural_network(params, x):\n","    for w, b in params:\n","        x = jnp.dot(x, w) + b\n","        x = jnp.maximum(0, x)  # ReLU activation\n","    return x\n","\n","# Initialize network parameters\n","layer_sizes = [training_features.shape[1], 200, 200, 1]\n","rng = jax.random.PRNGKey(0)\n","\n","def init_network_params(layer_sizes, key):\n","    scale = 1e-2\n","    params = []\n","    for n_in, n_out in zip(layer_sizes[:-1], layer_sizes[1:]):\n","        key, subkey = jax.random.split(key)\n","        w = scale * jax.random.normal(subkey, (n_in, n_out))\n","        b = jnp.zeros((n_out,))\n","        params.append((w, b))\n","    return params, key\n","\n","params, key = init_network_params(layer_sizes, rng)\n","\n","# Define loss function\n","def loss(params, inputs, targets):\n","    predictions = neural_network(params, inputs)\n","    return jnp.mean((predictions - targets) ** 2)\n","\n","# Compute gradient of the loss function\n","grad_loss = jit(grad(loss))\n","\n","# Training parameters\n","learning_rate = 0.001\n","num_epochs = 10000\n","\n","# Training loop (Simple Gradient Descent)\n","for epoch in range(num_epochs):\n","    gradient = grad_loss(params, training_features, training_target)\n","    params = [(w - learning_rate * grad_w, b - learning_rate * grad_b) for (w, b), (grad_w, grad_b) in zip(params, gradient)]\n","\n","\n","# Make predictions using the trained neural network\n","pred_nnet1 = neural_network(params, testing_features)\n","\n","print(tabulate([['NnetMulti', pcorrect(pred_nnet1, testing_target)]], headers=['Model', '% outperform mean']))\n","\n","# Evaluation of Results\n","# Convert testing_target to a 1-dimensional array\n","testing_target = jnp.ravel(testing_target)\n","prediction =jnp.ravel(pred_nnet1)\n","# Create the DataFrame\n","df_res = pd.DataFrame({'true': testing_target, 'pred_nnetMul': pred_nnet1})\n","\n","\n","# Plot the results of nnetMul\n","plt.figure(figsize=(12, 8))\n","plt.plot(df_res['true'], label='true', linestyle='--')\n","plt.plot(df_res['pred_nnetMul'], label='pred_nnetMul')\n","plt.legend()\n","plt.show()\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":914},"id":"RTfKLS1mpi2Y","executionInfo":{"status":"error","timestamp":1693673527150,"user_tz":180,"elapsed":58505,"user":{"displayName":"Argimiro Arratia Quesada","userId":"14219514422153572380"}},"outputId":"5ee9180b-a1fa-4c28-b6cd-4e2df06ec949"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Model        % outperform mean\n","---------  -------------------\n","NnetMulti                 -inf\n"]},{"output_type":"error","ename":"ValueError","evalue":"ignored","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)","\u001b[0;32m<ipython-input-34-835207cd4cd1>\u001b[0m in \u001b[0;36m<cell line: 124>\u001b[0;34m()\u001b[0m\n\u001b[1;32m    122\u001b[0m \u001b[0mprediction\u001b[0m \u001b[0;34m=\u001b[0m\u001b[0mjnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mravel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpred_nnet1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    123\u001b[0m \u001b[0;31m# Create the DataFrame\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 124\u001b[0;31m \u001b[0mdf_res\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDataFrame\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m'true'\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mtesting_target\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'pred_nnetMul'\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mpred_nnet1\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    125\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    126\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.10/dist-packages/pandas/core/frame.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, data, index, columns, dtype, copy)\u001b[0m\n\u001b[1;32m    662\u001b[0m         \u001b[0;32melif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdict\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    663\u001b[0m             \u001b[0;31m# GH#38939 de facto copy defaults to False only in non-dict cases\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 664\u001b[0;31m             \u001b[0mmgr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdict_to_mgr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcolumns\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtyp\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmanager\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    665\u001b[0m         \u001b[0;32melif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mma\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mMaskedArray\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    666\u001b[0m             \u001b[0;32mimport\u001b[0m \u001b[0mnumpy\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mma\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmrecords\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mmrecords\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.10/dist-packages/pandas/core/internals/construction.py\u001b[0m in \u001b[0;36mdict_to_mgr\u001b[0;34m(data, index, columns, dtype, typ, copy)\u001b[0m\n\u001b[1;32m    491\u001b[0m             \u001b[0marrays\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"dtype\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mx\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0marrays\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    492\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 493\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0marrays_to_mgr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marrays\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcolumns\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtyp\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtyp\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconsolidate\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    494\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    495\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.10/dist-packages/pandas/core/internals/construction.py\u001b[0m in \u001b[0;36marrays_to_mgr\u001b[0;34m(arrays, columns, index, dtype, verify_integrity, typ, consolidate)\u001b[0m\n\u001b[1;32m    121\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    122\u001b[0m         \u001b[0;31m# don't force copy because getting jammed in an ndarray anyway\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 123\u001b[0;31m         \u001b[0marrays\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_homogenize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marrays\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    124\u001b[0m         \u001b[0;31m# _homogenize ensures\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    125\u001b[0m         \u001b[0;31m#  - all(len(x) == len(index) for x in arrays)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.10/dist-packages/pandas/core/internals/construction.py\u001b[0m in \u001b[0;36m_homogenize\u001b[0;34m(data, index, dtype)\u001b[0m\n\u001b[1;32m    615\u001b[0m                 \u001b[0mval\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlib\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfast_multiget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mval\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moindex\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_values\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdefault\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnan\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    616\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 617\u001b[0;31m             val = sanitize_array(\n\u001b[0m\u001b[1;32m    618\u001b[0m                 \u001b[0mval\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mraise_cast_failure\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    619\u001b[0m             )\n","\u001b[0;32m/usr/local/lib/python3.10/dist-packages/pandas/core/construction.py\u001b[0m in \u001b[0;36msanitize_array\u001b[0;34m(data, index, dtype, copy, raise_cast_failure, allow_2d)\u001b[0m\n\u001b[1;32m    645\u001b[0m                 \u001b[0msubarr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmaybe_infer_to_datetimelike\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msubarr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    646\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 647\u001b[0;31m     \u001b[0msubarr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_sanitize_ndim\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msubarr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mallow_2d\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mallow_2d\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    648\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    649\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msubarr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndarray\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.10/dist-packages/pandas/core/construction.py\u001b[0m in \u001b[0;36m_sanitize_ndim\u001b[0;34m(result, data, dtype, index, allow_2d)\u001b[0m\n\u001b[1;32m    696\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mallow_2d\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    697\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mresult\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 698\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Data must be 1-dimensional\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    699\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mis_object_dtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mExtensionDtype\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    700\u001b[0m             \u001b[0;31m# i.e. PandasDtype(\"O\")\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;31mValueError\u001b[0m: Data must be 1-dimensional"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":["Defining your own Adam optimizer in JAX"],"metadata":{"id":"5yuRhZ9IkVQ0"}},{"cell_type":"code","source":["import jax.numpy as jnp\n","\n","def adam_optimizer(learning_rate=0.001, beta1=0.9, beta2=0.999, epsilon=1e-7):\n","    def init(params):\n","        state = [{'m': jnp.zeros_like(p), 'v': jnp.zeros_like(p)} for p in params]\n","        return state\n","\n","    def update(i, grads, state):\n","        new_state = []\n","        for s, g in zip(state, grads):\n","            m = beta1 * s['m'] + (1 - beta1) * g\n","            v = beta2 * s['v'] + (1 - beta2) * (g ** 2)\n","            m_hat = m / (1 - beta1 ** (i + 1))\n","            v_hat = v / (1 - beta2 ** (i + 1))\n","            delta = -learning_rate * m_hat / (jnp.sqrt(v_hat) + epsilon)\n","            new_state.append({'m': m, 'v': v})\n","        return new_state, delta\n","\n","    return init, update\n","\n","# Initialize the optimizer\n","init_optimizer, update_optimizer = adam_optimizer(learning_rate=0.001)\n","opt_state = init_optimizer(params)\n","\n","# Training loop\n","for epoch in range(num_epochs):\n","    gradient = grad_loss(get_params(opt_state), training_features, training_target)\n","    opt_state, delta = update_optimizer(epoch, gradient, opt_state)\n","    params = [param + delta_param for param, delta_param in zip(params, delta)]\n","\n","# ... rest of your code ...\n"],"metadata":{"id":"JTQoazE6kTYU"},"execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","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":"3.11.1"},"vscode":{"interpreter":{"hash":"31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6"}},"colab":{"provenance":[],"gpuType":"T4"}},"nbformat":4,"nbformat_minor":0}