--- a
+++ b/LSTM - Experiments/LSTM_Experiments.ipynb
@@ -0,0 +1 @@
+{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[],"toc_visible":true},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"code","execution_count":null,"metadata":{"id":"7zKdtJlIFhOn"},"outputs":[],"source":["import matplotlib.pyplot as plt\n","import pandas as pd\n","import torch\n","import torch.nn as nn\n","\n","import numpy as np\n","import torch.optim as optim\n","import torch.utils.data as data\n","from sklearn.model_selection import train_test_split\n","from sklearn.preprocessing import MinMaxScaler, LabelEncoder\n","import ast\n","from torch.utils.data import DataLoader, TensorDataset\n","from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n","import matplotlib.pyplot as plt"]},{"cell_type":"code","source":["dataset = pd.read_csv('labeled_dataset.csv')"],"metadata":{"id":"MeNmCEtPFzFo"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["dataset.head(15)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":519},"id":"P9w7w8WWF0Nd","executionInfo":{"status":"ok","timestamp":1692311813192,"user_tz":300,"elapsed":388,"user":{"displayName":"César Mosqueira","userId":"11705195256143475621"}},"outputId":"fde1231c-6d0f-44f8-b547-99975dec11e2"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["           video  group  frame  \\\n","0   video_15.mp4      1      5   \n","1   video_15.mp4      1     11   \n","2   video_15.mp4      1     17   \n","3   video_15.mp4      1     23   \n","4   video_15.mp4      1     29   \n","5   video_15.mp4      1     35   \n","6   video_15.mp4      1     41   \n","7   video_15.mp4      1     47   \n","8   video_15.mp4      1     53   \n","9   video_15.mp4      1     59   \n","10  video_15.mp4      2     71   \n","11  video_15.mp4      2     77   \n","12  video_15.mp4      2     83   \n","13  video_15.mp4      2     89   \n","14  video_15.mp4      2     95   \n","\n","                                            landmarks Label  \n","0   [     334.75      178.55     0.98386      339....   bad  \n","1   [     329.95      181.47     0.99063      334....   bad  \n","2   [      329.7      182.92     0.99079       334...   bad  \n","3   [     329.32      187.55     0.98055      334....   bad  \n","4   [     331.31      194.96       0.985      335....   bad  \n","5   [     326.15      199.31     0.98249      330....   bad  \n","6   [     317.77      207.25     0.97823       321...   bad  \n","7   [     311.16      194.05     0.97995      315....   bad  \n","8   [     308.41      194.28     0.98219      312....   bad  \n","9   [     318.15      184.29     0.98273       323...   bad  \n","10  [     321.08         126     0.99838      332....   bad  \n","11  [     311.02      128.68     0.99853      321....   bad  \n","12  [      301.6       139.2     0.99845      311....   bad  \n","13  [     294.57      158.13      0.9981      303....   bad  \n","14  [     294.98      166.88     0.99792      303....   bad  "],"text/html":["\n","  <div id=\"df-ef560272-0547-4bdb-ade2-64bf1f701d7d\" 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>video</th>\n","      <th>group</th>\n","      <th>frame</th>\n","      <th>landmarks</th>\n","      <th>Label</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>video_15.mp4</td>\n","      <td>1</td>\n","      <td>5</td>\n","      <td>[     334.75      178.55     0.98386      339....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>video_15.mp4</td>\n","      <td>1</td>\n","      <td>11</td>\n","      <td>[     329.95      181.47     0.99063      334....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>video_15.mp4</td>\n","      <td>1</td>\n","      <td>17</td>\n","      <td>[      329.7      182.92     0.99079       334...</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>video_15.mp4</td>\n","      <td>1</td>\n","      <td>23</td>\n","      <td>[     329.32      187.55     0.98055      334....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>video_15.mp4</td>\n","      <td>1</td>\n","      <td>29</td>\n","      <td>[     331.31      194.96       0.985      335....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>5</th>\n","      <td>video_15.mp4</td>\n","      <td>1</td>\n","      <td>35</td>\n","      <td>[     326.15      199.31     0.98249      330....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>6</th>\n","      <td>video_15.mp4</td>\n","      <td>1</td>\n","      <td>41</td>\n","      <td>[     317.77      207.25     0.97823       321...</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>7</th>\n","      <td>video_15.mp4</td>\n","      <td>1</td>\n","      <td>47</td>\n","      <td>[     311.16      194.05     0.97995      315....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>8</th>\n","      <td>video_15.mp4</td>\n","      <td>1</td>\n","      <td>53</td>\n","      <td>[     308.41      194.28     0.98219      312....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>9</th>\n","      <td>video_15.mp4</td>\n","      <td>1</td>\n","      <td>59</td>\n","      <td>[     318.15      184.29     0.98273       323...</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>10</th>\n","      <td>video_15.mp4</td>\n","      <td>2</td>\n","      <td>71</td>\n","      <td>[     321.08         126     0.99838      332....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>11</th>\n","      <td>video_15.mp4</td>\n","      <td>2</td>\n","      <td>77</td>\n","      <td>[     311.02      128.68     0.99853      321....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>12</th>\n","      <td>video_15.mp4</td>\n","      <td>2</td>\n","      <td>83</td>\n","      <td>[      301.6       139.2     0.99845      311....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>13</th>\n","      <td>video_15.mp4</td>\n","      <td>2</td>\n","      <td>89</td>\n","      <td>[     294.57      158.13      0.9981      303....</td>\n","      <td>bad</td>\n","    </tr>\n","    <tr>\n","      <th>14</th>\n","      <td>video_15.mp4</td>\n","      <td>2</td>\n","      <td>95</td>\n","      <td>[     294.98      166.88     0.99792      303....</td>\n","      <td>bad</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-ef560272-0547-4bdb-ade2-64bf1f701d7d')\"\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-ef560272-0547-4bdb-ade2-64bf1f701d7d 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-ef560272-0547-4bdb-ade2-64bf1f701d7d');\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-8448df95-760d-4350-a6bb-82553d08a229\">\n","  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-8448df95-760d-4350-a6bb-82553d08a229')\"\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","    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-quickchart: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","  [theme=dark] .colab-df-quickchart {\n","    background-color: #3B4455;\n","    fill: #D2E3FC;\n","  }\n","\n","  [theme=dark] .colab-df-quickchart: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","    async function quickchart(key) {\n","      const charts = await google.colab.kernel.invokeFunction(\n","          'suggestCharts', [key], {});\n","    }\n","    (() => {\n","      let quickchartButtonEl =\n","        document.querySelector('#df-8448df95-760d-4350-a6bb-82553d08a229 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":7}]},{"cell_type":"code","source":["def is_float(num):\n","    try:\n","        float(num)\n","        return True\n","    except ValueError:\n","        return False"],"metadata":{"id":"qxOmR6ioF1cR"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["dataset['landmarks'] = dataset['landmarks'].apply(lambda arr: np.array([float(n) for n in arr.split() if is_float(n)]))"],"metadata":{"id":"P9smPfHIF2qM"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["dataset['Label'].value_counts()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"EUaaKCMlF4SV","executionInfo":{"status":"ok","timestamp":1692311817060,"user_tz":300,"elapsed":3,"user":{"displayName":"César Mosqueira","userId":"11705195256143475621"}},"outputId":"a3113cb5-4c84-4081-809c-0cca9c575d4c"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["bad     3444\n","good    1205\n","Name: Label, dtype: int64"]},"metadata":{},"execution_count":10}]},{"cell_type":"code","source":["# Group the data by 'video' and 'group'\n","grouped_data = dataset.groupby(['video', 'group'])\n","\n","# Define the sequence length\n","sequence_length = 10\n","\n","# Create lists to store the sequences and labels\n","sequences = []\n","labels = []\n","\n","# Iterate over each group\n","for group, data in grouped_data:\n","    landmarks = data['landmarks'].tolist()\n","    group_labels = data['Label'].tolist()\n","\n","    # Create sequences of landmarks\n","    for i in range(len(landmarks) - sequence_length + 1):\n","        sequence = landmarks[i:i+sequence_length]\n","        sequences.append(sequence)\n","        labels.append(group_labels[i+sequence_length-1])"],"metadata":{"id":"A0zDHdCnF5tN"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["sequences = np.array(sequences)\n","\n","scaler = MinMaxScaler()\n","normalized_sequences = np.zeros_like(sequences)\n","\n","for i in range(sequences.shape[0]):\n","    for j in range(sequences.shape[1]):\n","        # Flatten the landmarks for each set within the sequence\n","        landmarks_flattened = np.reshape(sequences[i, j], (-1, 1))\n","        # Normalize the landmarks\n","        landmarks_normalized = scaler.fit_transform(landmarks_flattened)\n","        # Reshape the normalized landmarks back to the original shape\n","        normalized_landmarks = np.reshape(landmarks_normalized, sequences[i, j].shape)\n","        # Update the normalized landmarks in the sequences array\n","        normalized_sequences[i, j] = normalized_landmarks"],"metadata":{"id":"wGMZpHdEF8Z2"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["label_encoder = LabelEncoder()\n","labels_encoded = label_encoder.fit_transform(labels)"],"metadata":{"id":"6SVR-ictF98k"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["label_encoder.transform(['good'])"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"O9qvWDJ3F_SC","executionInfo":{"status":"ok","timestamp":1692311835736,"user_tz":300,"elapsed":2,"user":{"displayName":"César Mosqueira","userId":"11705195256143475621"}},"outputId":"7d541e84-4852-4e37-81a8-117a9cc9cd6e"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([1])"]},"metadata":{},"execution_count":14}]},{"cell_type":"code","source":["train_X, test_X, train_y, test_y = train_test_split(normalized_sequences, labels_encoded, test_size=0.2, shuffle=True)"],"metadata":{"id":"10be1q-rGAV1"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["print(train_X.shape)\n","print(train_y.shape)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Hox_iyMwGBq6","executionInfo":{"status":"ok","timestamp":1692311839481,"user_tz":300,"elapsed":2,"user":{"displayName":"César Mosqueira","userId":"11705195256143475621"}},"outputId":"d0a747f3-6383-4034-b00c-4533cc8f337c"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["(355, 10, 50)\n","(355,)\n"]}]},{"cell_type":"code","source":["train_X_tensor = torch.Tensor(train_X)\n","train_y_tensor = torch.Tensor(train_y)"],"metadata":{"id":"kan0J9GVGCrs"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["# First iteration of LSTM"],"metadata":{"id":"m6iKzio8JNSL"}},{"cell_type":"code","source":["train_dataset = TensorDataset(train_X_tensor, train_y_tensor)\n","train_dataloader = DataLoader(train_dataset, batch_size=32, shuffle=True)"],"metadata":{"id":"msOwbxKyGDvQ"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["class LSTMModel(nn.Module):\n","    def __init__(self, input_size, hidden_size, num_classes):\n","        super(LSTMModel, self).__init__()\n","        self.lstm = nn.LSTM(input_size, hidden_size, batch_first=True)\n","        self.dropout = nn.Dropout(0.2)\n","        self.fc1 = nn.Linear(hidden_size, 32)\n","        self.fc2 = nn.Linear(32, num_classes)\n","        self.relu = nn.ReLU()\n","\n","    def forward(self, x):\n","        _, (h_n, _) = self.lstm(x)\n","        x = self.dropout(h_n[-1])\n","        x = self.fc1(x)\n","        x = self.fc2(x)\n","        x = self.relu(x)\n","        return x"],"metadata":{"id":"whZxvNyLGFL5"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["input_size = train_X.shape[2]\n","hidden_size = 256\n","num_classes = 1\n","num_epochs = 30\n","learning_rate = 0.00001\n","\n","# Instantiate the model\n","model = LSTMModel(input_size, hidden_size, num_classes)"],"metadata":{"id":"Fck9hSdRGKS7"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# Define the loss function and optimizer\n","criterion = nn.BCELoss()\n","optimizer = optim.SGD(model.parameters(), lr=learning_rate)"],"metadata":{"id":"QlpoZwz1GKyq"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["train_X.shape"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"6HZW1eFwGL3D","executionInfo":{"status":"ok","timestamp":1692295580820,"user_tz":300,"elapsed":11,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"229b83bd-8e51-4560-eb55-2eacb4fb641d"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(355, 10, 50)"]},"metadata":{},"execution_count":69}]},{"cell_type":"code","source":["model"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"U0MaHSm9GM4W","executionInfo":{"status":"ok","timestamp":1692295581791,"user_tz":300,"elapsed":5,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"1c9b1d16-49d3-435c-a40f-79fa7c7f0eae"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["LSTMModel(\n","  (lstm): LSTM(50, 256, batch_first=True)\n","  (dropout): Dropout(p=0.2, inplace=False)\n","  (fc1): Linear(in_features=256, out_features=32, bias=True)\n","  (fc2): Linear(in_features=32, out_features=1, bias=True)\n","  (relu): ReLU()\n",")"]},"metadata":{},"execution_count":70}]},{"cell_type":"code","source":["epoch_accuracy = []\n","epoch_f1_score = []\n","epoch_recall = []\n","epoch_loss = []\n","\n","for epoch in range(num_epochs):\n","    true_labels = []\n","    predicted_labels = []\n","    for inputs, labels in train_dataloader:\n","        # Zero the gradients\n","        optimizer.zero_grad()\n","\n","        # Forward pass\n","        outputs = model(inputs)\n","        predictions = torch.round(torch.sigmoid(outputs.squeeze()))\n","\n","        true_labels.extend(labels.numpy())\n","        predicted_labels.extend(predictions.detach().numpy())\n","\n","        loss = criterion(outputs.squeeze(), labels)\n","\n","        # Backward pass and optimization\n","        loss.backward()\n","        optimizer.step()\n","\n","    # Calculate metrics\n","    accuracy = accuracy_score(true_labels, predicted_labels)\n","    f1 = f1_score(true_labels, predicted_labels)\n","    recall = recall_score(true_labels, predicted_labels)\n","\n","    # Store metrics for each epoch\n","    epoch_accuracy.append(accuracy)\n","    epoch_f1_score.append(f1)\n","    epoch_recall.append(recall)\n","    epoch_loss.append(loss.item())\n","    # Print the metrics for every epoch\n","    print(f'Epoch {epoch+1}/{num_epochs}, Loss: {loss.item()}, Accuracy: {accuracy}, F1 Score: {f1}, Recall: {recall}')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"vZfo2EV7GN8H","executionInfo":{"status":"ok","timestamp":1692295594209,"user_tz":300,"elapsed":7890,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"7a2199eb-6b69-443a-89d1-cf71269c8f49"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/30, Loss: 3.30371356010437, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 2/30, Loss: 0.0640532448887825, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 3/30, Loss: 0.9332453608512878, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 4/30, Loss: 0.9942008852958679, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 5/30, Loss: 2.844069480895996, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 6/30, Loss: 1.0282055139541626, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 7/30, Loss: 0.07186160236597061, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 8/30, Loss: 0.9316868185997009, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 9/30, Loss: 0.07316235452890396, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 10/30, Loss: 0.06532102078199387, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 11/30, Loss: 0.05990065634250641, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 12/30, Loss: 0.9599320292472839, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 13/30, Loss: 1.0268315076828003, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 14/30, Loss: 1.7993221282958984, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 15/30, Loss: 0.9321920275688171, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 16/30, Loss: 0.0565328486263752, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 17/30, Loss: 1.7288103103637695, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 18/30, Loss: 1.0498160123825073, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 19/30, Loss: 0.057037562131881714, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 20/30, Loss: 0.08153844624757767, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 21/30, Loss: 0.9557292461395264, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 22/30, Loss: 0.9206430315971375, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 23/30, Loss: 1.7758244276046753, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 24/30, Loss: 0.8312497138977051, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 25/30, Loss: 1.63068687915802, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 26/30, Loss: 0.8575438857078552, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 27/30, Loss: 0.09230203181505203, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 28/30, Loss: 0.8663346767425537, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 29/30, Loss: 0.07764268666505814, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 30/30, Loss: 1.0366336107254028, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n"]}]},{"cell_type":"markdown","source":["# Seconrd Iteration of LSTM"],"metadata":{"id":"6nEKstr5JTxk"}},{"cell_type":"code","source":["train_dataset = TensorDataset(train_X_tensor, train_y_tensor)\n","train_dataloader = DataLoader(train_dataset, batch_size=32, shuffle=True)"],"metadata":{"id":"tMcRmy1mKEB_"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["class ImprovedLSTMModel(nn.Module):\n","    def __init__(self, input_size, hidden_size, num_classes, num_layers=1, bidirectional=True, dropout_rate=0.2):\n","        super(ImprovedLSTMModel, self).__init__()\n","        self.lstm = nn.LSTM(input_size, hidden_size, num_layers=num_layers, batch_first=True, bidirectional=bidirectional)\n","        self.dropout = nn.Dropout(dropout_rate)\n","\n","        # Calculate the correct input size for the first linear layer based on bidirectional LSTM\n","        fc1_input_size = hidden_size * (2 if bidirectional else 1) * num_layers\n","        self.fc1 = nn.Linear(fc1_input_size, 128)  # Use 128 units for the first dense layer\n","        self.fc2 = nn.Linear(128, num_classes)\n","        self.relu = nn.ReLU()\n","\n","    def forward(self, x):\n","        _, (h_n, _) = self.lstm(x)\n","        h_n_concat = h_n.permute(1, 0, 2).contiguous().view(h_n.shape[1], -1)  # Reshape h_n for linear layer\n","        x = self.dropout(h_n_concat)\n","        x = self.fc1(x)\n","        x = self.fc2(x)\n","        x = self.relu(x)\n","        return x"],"metadata":{"id":"n1qUiIfSJXXp"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["input_size = train_X.shape[2]\n","hidden_size = 256\n","num_classes = 1\n","num_layers = 2\n","bidirectional = True\n","dropout_rate = 0.2\n","learning_rate = 0.00001\n","\n","# Instantiate the improved model\n","model = ImprovedLSTMModel(input_size, hidden_size, num_classes, num_layers, bidirectional, dropout_rate)\n","\n","# Define the loss function and optimizer\n","criterion = nn.BCELoss()\n","optimizer = optim.SGD(model.parameters(), lr=learning_rate)"],"metadata":{"id":"ZpxWbsioJg0b"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["model"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"LEYM9BBfJpJx","executionInfo":{"status":"ok","timestamp":1692295679473,"user_tz":300,"elapsed":958,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"c9fbd8e6-44f3-44ed-f1cc-2355396a258b"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["ImprovedLSTMModel(\n","  (lstm): LSTM(50, 256, num_layers=2, batch_first=True, bidirectional=True)\n","  (dropout): Dropout(p=0.2, inplace=False)\n","  (fc1): Linear(in_features=1024, out_features=128, bias=True)\n","  (fc2): Linear(in_features=128, out_features=1, bias=True)\n","  (relu): ReLU()\n",")"]},"metadata":{},"execution_count":75}]},{"cell_type":"code","source":["epoch_accuracy = []\n","epoch_f1_score = []\n","epoch_recall = []\n","epoch_loss = []\n","\n","for epoch in range(num_epochs):\n","    true_labels = []\n","    predicted_labels = []\n","    for inputs, labels in train_dataloader:\n","        # Zero the gradients\n","        optimizer.zero_grad()\n","\n","        # Forward pass\n","        outputs = model(inputs)\n","        predictions = torch.round(torch.sigmoid(outputs.squeeze()))\n","\n","        true_labels.extend(labels.numpy())\n","        predicted_labels.extend(predictions.detach().numpy())\n","\n","        loss = criterion(outputs.squeeze(), labels)\n","\n","        # Backward pass and optimization\n","        loss.backward()\n","        optimizer.step()\n","\n","    # Calculate metrics\n","    accuracy = accuracy_score(true_labels, predicted_labels)\n","    f1 = f1_score(true_labels, predicted_labels)\n","    recall = recall_score(true_labels, predicted_labels)\n","\n","    # Store metrics for each epoch\n","    epoch_accuracy.append(accuracy)\n","    epoch_f1_score.append(f1)\n","    epoch_recall.append(recall)\n","    epoch_loss.append(loss.item())\n","    # Print the metrics for every epoch\n","    print(f'Epoch {epoch+1}/{num_epochs}, Loss: {loss.item()}, Accuracy: {accuracy}, F1 Score: {f1}, Recall: {recall}')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"XEIU1OidKGTR","executionInfo":{"status":"ok","timestamp":1692295728092,"user_tz":300,"elapsed":47843,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"4deb738f-129d-4e5d-e38b-c83aaa3e611d"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/30, Loss: 0.14196285605430603, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 2/30, Loss: 0.7603593468666077, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 3/30, Loss: 0.1483137458562851, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 4/30, Loss: 0.1442556381225586, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 5/30, Loss: 2.0081534385681152, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 6/30, Loss: 2.103062391281128, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 7/30, Loss: 0.7612888216972351, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 8/30, Loss: 0.7601293921470642, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 9/30, Loss: 0.15363752841949463, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 10/30, Loss: 0.1545637995004654, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 11/30, Loss: 0.14706026017665863, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 12/30, Loss: 0.7576022744178772, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 13/30, Loss: 0.7605314254760742, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 14/30, Loss: 1.419185996055603, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 15/30, Loss: 1.3716388940811157, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 16/30, Loss: 1.329178810119629, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 17/30, Loss: 0.743213951587677, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 18/30, Loss: 0.8101215362548828, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 19/30, Loss: 0.1562623828649521, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 20/30, Loss: 0.1567939668893814, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 21/30, Loss: 1.3488202095031738, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 22/30, Loss: 0.1470808982849121, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 23/30, Loss: 0.1572873443365097, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 24/30, Loss: 0.14903366565704346, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 25/30, Loss: 0.7512195706367493, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 26/30, Loss: 0.711514949798584, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 27/30, Loss: 0.7488934993743896, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 28/30, Loss: 0.15341293811798096, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 29/30, Loss: 0.7657694816589355, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 30/30, Loss: 0.15584520995616913, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n"]}]},{"cell_type":"markdown","source":["# Thrird iteration of LSTM"],"metadata":{"id":"qcdyA9WGMwUw"}},{"cell_type":"code","source":["train_dataset = TensorDataset(train_X_tensor, train_y_tensor)\n","train_dataloader = DataLoader(train_dataset, batch_size=16, shuffle=True)"],"metadata":{"id":"VqEEGw2oKL5a"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["class LSTMModel(nn.Module):\n","    def __init__(self, input_size, hidden_size, num_classes):\n","        super(LSTMModel, self).__init__()\n","        self.lstm = nn.LSTM(input_size, hidden_size, batch_first=True)\n","        self.dropout = nn.Dropout(0.2)\n","        self.fc1 = nn.Linear(hidden_size, 32)\n","        self.relu = nn.ReLU6()\n","        self.fc2 = nn.Linear(32, num_classes)\n","        self.relu = nn.ReLU6()\n","\n","    def forward(self, x):\n","        _, (h_n, _) = self.lstm(x)\n","        x = self.dropout(h_n[-1])\n","        x = self.fc1(x)\n","        x = self.fc2(x)\n","        x = self.relu(x)\n","        return x"],"metadata":{"id":"1TAwRu-WVzJf"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["input_size = train_X.shape[2]\n","hidden_size = 256\n","num_classes = 1\n","num_epochs = 20\n","learning_rate = 0.01\n","\n","# Instantiate the model\n","model = LSTMModel(input_size, hidden_size, num_classes)\n","\n","# Define the loss function and optimizer\n","criterion = nn.BCELoss()\n","optimizer = optim.SGD(model.parameters(), lr=learning_rate)"],"metadata":{"id":"hQjiUGqJM9VF"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["model"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"469B6ZaHVu31","executionInfo":{"status":"ok","timestamp":1692295813086,"user_tz":300,"elapsed":7,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"f95e0662-4fc6-4a5f-d63c-55cde09f18cb"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["LSTMModel(\n","  (lstm): LSTM(50, 256, batch_first=True)\n","  (dropout): Dropout(p=0.2, inplace=False)\n","  (fc1): Linear(in_features=256, out_features=32, bias=True)\n","  (relu): ReLU6()\n","  (fc2): Linear(in_features=32, out_features=1, bias=True)\n",")"]},"metadata":{},"execution_count":84}]},{"cell_type":"code","source":["epoch_accuracy = []\n","epoch_f1_score = []\n","epoch_recall = []\n","epoch_loss = []\n","\n","for epoch in range(num_epochs):\n","    true_labels = []\n","    predicted_labels = []\n","    for inputs, labels in train_dataloader:\n","        # Zero the gradients\n","        optimizer.zero_grad()\n","\n","        # Forward pass\n","        outputs = model(inputs)\n","        predictions = torch.round(torch.sigmoid(outputs.squeeze()))\n","\n","        true_labels.extend(labels.numpy())\n","        predicted_labels.extend(predictions.detach().numpy())\n","\n","        loss = criterion(outputs.squeeze(), labels)\n","\n","        # Backward pass and optimization\n","        loss.backward()\n","        optimizer.step()\n","\n","    # Calculate metrics\n","    accuracy = accuracy_score(true_labels, predicted_labels)\n","    f1 = f1_score(true_labels, predicted_labels)\n","    recall = recall_score(true_labels, predicted_labels)\n","\n","    # Store metrics for each epoch\n","    epoch_accuracy.append(accuracy)\n","    epoch_f1_score.append(f1)\n","    epoch_recall.append(recall)\n","    epoch_loss.append(loss.item())\n","    # Print the metrics for every epoch\n","    print(f'Epoch {epoch+1}/{num_epochs}, Loss: {loss.item()}, Accuracy: {accuracy}, F1 Score: {f1}, Recall: {recall}')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"gBn4JcRcVsgq","executionInfo":{"status":"ok","timestamp":1692295819755,"user_tz":300,"elapsed":5947,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"ffb30456-1a91-4c7e-8c82-a75f1aa14851"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/20, Loss: 0.6219484806060791, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 2/20, Loss: 0.6616726517677307, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 3/20, Loss: 0.6715094447135925, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 4/20, Loss: 0.32363080978393555, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 5/20, Loss: 0.6457496285438538, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 6/20, Loss: 0.27271386981010437, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 7/20, Loss: 0.28883811831474304, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 8/20, Loss: 0.3118828535079956, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 9/20, Loss: 0.266092449426651, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 10/20, Loss: 0.29886820912361145, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 11/20, Loss: 0.2565235197544098, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 12/20, Loss: 0.2884318232536316, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 13/20, Loss: 0.22864384949207306, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 14/20, Loss: 1.2777477502822876, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 15/20, Loss: 0.3226184546947479, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 16/20, Loss: 0.944756805896759, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 17/20, Loss: 0.989539623260498, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 18/20, Loss: 0.28581663966178894, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 19/20, Loss: 0.6567023396492004, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n","Epoch 20/20, Loss: 0.5800800919532776, Accuracy: 0.2591549295774648, F1 Score: 0.41163310961968674, Recall: 1.0\n"]}]},{"cell_type":"markdown","source":["# KERAS"],"metadata":{"id":"zzbvFblpAW6l"}},{"cell_type":"markdown","source":["## 1 Iteration"],"metadata":{"id":"f2Jn64iP9f7W"}},{"cell_type":"code","source":["train_dataset = TensorDataset(train_X_tensor, train_y_tensor)\n","train_dataloader = DataLoader(train_dataset, batch_size=16, shuffle=True)"],"metadata":{"id":"G6pUQcBFAcCN"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["import numpy as np\n","from keras.models import Sequential\n","from keras.layers import LSTM, Dense, Dropout"],"metadata":{"id":"GcaXwWAiAbt5"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["num_features = normalized_sequences.shape[2]\n","# Define the LSTM model\n","model = Sequential()\n","model.add(LSTM(units=64, input_shape=(sequence_length, num_features), return_sequences=True))\n","model.add(Dropout(0.2))\n","model.add(LSTM(units=64))\n","model.add(Dropout(0.2))\n","model.add(Dense(units=1, activation='sigmoid'))\n","\n","# Compile the model\n","model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n","\n","# Print the model summary\n","model.summary()\n","\n","# Train the model\n","batch_size = 32\n","epochs = 30\n","history = model.fit(train_X, train_y, batch_size=batch_size, epochs=epochs, validation_split=0.2)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"qQ0CwkQwV2b5","executionInfo":{"status":"ok","timestamp":1692301346068,"user_tz":300,"elapsed":14789,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"4d832068-bd15-49ee-9c04-ca9005602ee3"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Model: \"sequential_2\"\n","_________________________________________________________________\n"," Layer (type)                Output Shape              Param #   \n","=================================================================\n"," lstm_2 (LSTM)               (None, 10, 64)            29440     \n","                                                                 \n"," dropout_2 (Dropout)         (None, 10, 64)            0         \n","                                                                 \n"," lstm_3 (LSTM)               (None, 64)                33024     \n","                                                                 \n"," dropout_3 (Dropout)         (None, 64)                0         \n","                                                                 \n"," dense_1 (Dense)             (None, 1)                 65        \n","                                                                 \n","=================================================================\n","Total params: 62,529\n","Trainable params: 62,529\n","Non-trainable params: 0\n","_________________________________________________________________\n","Epoch 1/30\n","9/9 [==============================] - 6s 208ms/step - loss: 0.6018 - accuracy: 0.7430 - val_loss: 0.6794 - val_accuracy: 0.6761\n","Epoch 2/30\n","9/9 [==============================] - 0s 42ms/step - loss: 0.5474 - accuracy: 0.7570 - val_loss: 0.6236 - val_accuracy: 0.6761\n","Epoch 3/30\n","9/9 [==============================] - 0s 45ms/step - loss: 0.5487 - accuracy: 0.7570 - val_loss: 0.6342 - val_accuracy: 0.6761\n","Epoch 4/30\n","9/9 [==============================] - 0s 45ms/step - loss: 0.5318 - accuracy: 0.7570 - val_loss: 0.6125 - val_accuracy: 0.6761\n","Epoch 5/30\n","9/9 [==============================] - 0s 46ms/step - loss: 0.4928 - accuracy: 0.7570 - val_loss: 0.5275 - val_accuracy: 0.7183\n","Epoch 6/30\n","9/9 [==============================] - 0s 42ms/step - loss: 0.5334 - accuracy: 0.7606 - val_loss: 0.5935 - val_accuracy: 0.6761\n","Epoch 7/30\n","9/9 [==============================] - 0s 38ms/step - loss: 0.4558 - accuracy: 0.7852 - val_loss: 0.4841 - val_accuracy: 0.7183\n","Epoch 8/30\n","9/9 [==============================] - 0s 34ms/step - loss: 0.4072 - accuracy: 0.7852 - val_loss: 0.4071 - val_accuracy: 0.7746\n","Epoch 9/30\n","9/9 [==============================] - 0s 43ms/step - loss: 0.3700 - accuracy: 0.7993 - val_loss: 0.3723 - val_accuracy: 0.8028\n","Epoch 10/30\n","9/9 [==============================] - 0s 44ms/step - loss: 0.3615 - accuracy: 0.8063 - val_loss: 0.4775 - val_accuracy: 0.7887\n","Epoch 11/30\n","9/9 [==============================] - 0s 37ms/step - loss: 0.3962 - accuracy: 0.7993 - val_loss: 0.6988 - val_accuracy: 0.6761\n","Epoch 12/30\n","9/9 [==============================] - 0s 27ms/step - loss: 0.4335 - accuracy: 0.7852 - val_loss: 0.4021 - val_accuracy: 0.9014\n","Epoch 13/30\n","9/9 [==============================] - 0s 23ms/step - loss: 0.3833 - accuracy: 0.8204 - val_loss: 0.4010 - val_accuracy: 0.7887\n","Epoch 14/30\n","9/9 [==============================] - 0s 22ms/step - loss: 0.3861 - accuracy: 0.8063 - val_loss: 0.4407 - val_accuracy: 0.7746\n","Epoch 15/30\n","9/9 [==============================] - 0s 23ms/step - loss: 0.3911 - accuracy: 0.8204 - val_loss: 0.3450 - val_accuracy: 0.9014\n","Epoch 16/30\n","9/9 [==============================] - 0s 23ms/step - loss: 0.3624 - accuracy: 0.8486 - val_loss: 0.3330 - val_accuracy: 0.8169\n","Epoch 17/30\n","9/9 [==============================] - 0s 25ms/step - loss: 0.3641 - accuracy: 0.8063 - val_loss: 0.5208 - val_accuracy: 0.7746\n","Epoch 18/30\n","9/9 [==============================] - 0s 22ms/step - loss: 0.3684 - accuracy: 0.8239 - val_loss: 0.3172 - val_accuracy: 0.8310\n","Epoch 19/30\n","9/9 [==============================] - 0s 22ms/step - loss: 0.3354 - accuracy: 0.8345 - val_loss: 0.2897 - val_accuracy: 0.9014\n","Epoch 20/30\n","9/9 [==============================] - 0s 26ms/step - loss: 0.3454 - accuracy: 0.8134 - val_loss: 0.3917 - val_accuracy: 0.8028\n","Epoch 21/30\n","9/9 [==============================] - 0s 22ms/step - loss: 0.3410 - accuracy: 0.8345 - val_loss: 0.3124 - val_accuracy: 0.8451\n","Epoch 22/30\n","9/9 [==============================] - 0s 23ms/step - loss: 0.3278 - accuracy: 0.8345 - val_loss: 0.2882 - val_accuracy: 0.9155\n","Epoch 23/30\n","9/9 [==============================] - 0s 25ms/step - loss: 0.3276 - accuracy: 0.8345 - val_loss: 0.3020 - val_accuracy: 0.8592\n","Epoch 24/30\n","9/9 [==============================] - 0s 22ms/step - loss: 0.3348 - accuracy: 0.8310 - val_loss: 0.2724 - val_accuracy: 0.9155\n","Epoch 25/30\n","9/9 [==============================] - 0s 29ms/step - loss: 0.3508 - accuracy: 0.8275 - val_loss: 0.2979 - val_accuracy: 0.9014\n","Epoch 26/30\n","9/9 [==============================] - 0s 26ms/step - loss: 0.3341 - accuracy: 0.8662 - val_loss: 0.4384 - val_accuracy: 0.8028\n","Epoch 27/30\n","9/9 [==============================] - 0s 24ms/step - loss: 0.3112 - accuracy: 0.8521 - val_loss: 0.2568 - val_accuracy: 0.9296\n","Epoch 28/30\n","9/9 [==============================] - 0s 23ms/step - loss: 0.3373 - accuracy: 0.8486 - val_loss: 0.3282 - val_accuracy: 0.8451\n","Epoch 29/30\n","9/9 [==============================] - 0s 23ms/step - loss: 0.3946 - accuracy: 0.7993 - val_loss: 0.4196 - val_accuracy: 0.7465\n","Epoch 30/30\n","9/9 [==============================] - 0s 22ms/step - loss: 0.3253 - accuracy: 0.8345 - val_loss: 0.3035 - val_accuracy: 0.8451\n"]}]},{"cell_type":"code","source":["# Evaluate the model\n","test_loss, test_accuracy = model.evaluate(test_X, test_y)\n","print(\"Test Loss:\", test_loss)\n","print(\"Test Accuracy:\", test_accuracy)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"mUJzwnRiAjTR","executionInfo":{"status":"ok","timestamp":1692301355734,"user_tz":300,"elapsed":605,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"b2a14763-1112-4427-e14b-bfc11ac33426"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["3/3 [==============================] - 0s 9ms/step - loss: 0.3437 - accuracy: 0.8539\n","Test Loss: 0.3437446355819702\n","Test Accuracy: 0.8539325594902039\n"]}]},{"cell_type":"code","source":["# Plot training and validation metrics\n","plt.figure(figsize=(10, 4))\n","plt.subplot(1, 2, 1)\n","plt.plot(history.history['loss'], label='Training Loss')\n","plt.plot(history.history['val_loss'], label='Validation Loss')\n","plt.xlabel('Epoch')\n","plt.ylabel('Loss')\n","plt.legend()\n","\n","plt.subplot(1, 2, 2)\n","plt.plot(history.history['accuracy'], label='Training Accuracy')\n","plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n","plt.xlabel('Epoch')\n","plt.ylabel('Accuracy')\n","plt.legend()\n","\n","plt.tight_layout()\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":406},"id":"Q2QfFCBcAwAv","executionInfo":{"status":"ok","timestamp":1692301359426,"user_tz":300,"elapsed":969,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"95875b70-9956-49f1-9900-5fecda96acc5"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1000x400 with 2 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":["## 2 iteration"],"metadata":{"id":"mTInRUCaDOlN"}},{"cell_type":"code","source":["from keras.optimizers import Adam"],"metadata":{"id":"jXw8MdvnD0jo"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["num_features = normalized_sequences.shape[2]\n","# Define the LSTM model\n","model = Sequential()\n","model.add(LSTM(units=64, input_shape=(sequence_length, num_features), return_sequences=True))\n","model.add(Dropout(0.5))\n","model.add(LSTM(units=64))\n","model.add(Dropout(0.5))\n","model.add(Dense(units=1, activation='sigmoid'))\n","\n","custom_learning_rate = 0.001\n","optimizer = Adam(learning_rate=custom_learning_rate)\n","\n","# Compile the model\n","model.compile(loss='binary_crossentropy', optimizer=optimizer, metrics=['accuracy'])\n","\n","# Print the model summary\n","model.summary()\n","\n","# Train the model\n","batch_size = 32\n","epochs = 60\n","history = model.fit(train_X, train_y, batch_size=batch_size, epochs=epochs, validation_split=0.2, verbose=1)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"NM-mOMJUBG9d","executionInfo":{"status":"ok","timestamp":1692302229233,"user_tz":300,"elapsed":24949,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"b17c4bb3-e63c-4d03-f517-5447fe886c73"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Model: \"sequential_9\"\n","_________________________________________________________________\n"," Layer (type)                Output Shape              Param #   \n","=================================================================\n"," lstm_16 (LSTM)              (None, 10, 64)            29440     \n","                                                                 \n"," dropout_16 (Dropout)        (None, 10, 64)            0         \n","                                                                 \n"," lstm_17 (LSTM)              (None, 64)                33024     \n","                                                                 \n"," dropout_17 (Dropout)        (None, 64)                0         \n","                                                                 \n"," dense_8 (Dense)             (None, 1)                 65        \n","                                                                 \n","=================================================================\n","Total params: 62,529\n","Trainable params: 62,529\n","Non-trainable params: 0\n","_________________________________________________________________\n","Epoch 1/60\n","9/9 [==============================] - 5s 137ms/step - loss: 0.5819 - accuracy: 0.7500 - val_loss: 0.6935 - val_accuracy: 0.6761\n","Epoch 2/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.5605 - accuracy: 0.7570 - val_loss: 0.6287 - val_accuracy: 0.6761\n","Epoch 3/60\n","9/9 [==============================] - 0s 22ms/step - loss: 0.5663 - accuracy: 0.7570 - val_loss: 0.6188 - val_accuracy: 0.6761\n","Epoch 4/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.5395 - accuracy: 0.7570 - val_loss: 0.6263 - val_accuracy: 0.6761\n","Epoch 5/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.5264 - accuracy: 0.7606 - val_loss: 0.5813 - val_accuracy: 0.6761\n","Epoch 6/60\n","9/9 [==============================] - 0s 21ms/step - loss: 0.4948 - accuracy: 0.7570 - val_loss: 0.5450 - val_accuracy: 0.7183\n","Epoch 7/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.4424 - accuracy: 0.7676 - val_loss: 0.5160 - val_accuracy: 0.7465\n","Epoch 8/60\n","9/9 [==============================] - 0s 22ms/step - loss: 0.4390 - accuracy: 0.7958 - val_loss: 0.5872 - val_accuracy: 0.7465\n","Epoch 9/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.4254 - accuracy: 0.7782 - val_loss: 0.5325 - val_accuracy: 0.7465\n","Epoch 10/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.4564 - accuracy: 0.7394 - val_loss: 0.5347 - val_accuracy: 0.7324\n","Epoch 11/60\n","9/9 [==============================] - 0s 22ms/step - loss: 0.4075 - accuracy: 0.7923 - val_loss: 0.3847 - val_accuracy: 0.7887\n","Epoch 12/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.4177 - accuracy: 0.7923 - val_loss: 0.4642 - val_accuracy: 0.7465\n","Epoch 13/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.4260 - accuracy: 0.7923 - val_loss: 0.5127 - val_accuracy: 0.7465\n","Epoch 14/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.3976 - accuracy: 0.7852 - val_loss: 0.3617 - val_accuracy: 0.8169\n","Epoch 15/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.3906 - accuracy: 0.8063 - val_loss: 0.3605 - val_accuracy: 0.8732\n","Epoch 16/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.3821 - accuracy: 0.7958 - val_loss: 0.4222 - val_accuracy: 0.7887\n","Epoch 17/60\n","9/9 [==============================] - 0s 26ms/step - loss: 0.3660 - accuracy: 0.7887 - val_loss: 0.4211 - val_accuracy: 0.7887\n","Epoch 18/60\n","9/9 [==============================] - 0s 22ms/step - loss: 0.3776 - accuracy: 0.7923 - val_loss: 0.3312 - val_accuracy: 0.8732\n","Epoch 19/60\n","9/9 [==============================] - 0s 22ms/step - loss: 0.3680 - accuracy: 0.8134 - val_loss: 0.3497 - val_accuracy: 0.9296\n","Epoch 20/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.3578 - accuracy: 0.8380 - val_loss: 0.5053 - val_accuracy: 0.7606\n","Epoch 21/60\n","9/9 [==============================] - 0s 26ms/step - loss: 0.4147 - accuracy: 0.8063 - val_loss: 0.4343 - val_accuracy: 0.7606\n","Epoch 22/60\n","9/9 [==============================] - 0s 39ms/step - loss: 0.3787 - accuracy: 0.7887 - val_loss: 0.3581 - val_accuracy: 0.8732\n","Epoch 23/60\n","9/9 [==============================] - 0s 41ms/step - loss: 0.3746 - accuracy: 0.8063 - val_loss: 0.4687 - val_accuracy: 0.7606\n","Epoch 24/60\n","9/9 [==============================] - 0s 40ms/step - loss: 0.3559 - accuracy: 0.8239 - val_loss: 0.3165 - val_accuracy: 0.8873\n","Epoch 25/60\n","9/9 [==============================] - 0s 33ms/step - loss: 0.3335 - accuracy: 0.8169 - val_loss: 0.3215 - val_accuracy: 0.8310\n","Epoch 26/60\n","9/9 [==============================] - 0s 34ms/step - loss: 0.3496 - accuracy: 0.8134 - val_loss: 0.4068 - val_accuracy: 0.8028\n","Epoch 27/60\n","9/9 [==============================] - 0s 37ms/step - loss: 0.3751 - accuracy: 0.7993 - val_loss: 0.5081 - val_accuracy: 0.7606\n","Epoch 28/60\n","9/9 [==============================] - 0s 37ms/step - loss: 0.3561 - accuracy: 0.8415 - val_loss: 0.3360 - val_accuracy: 0.9014\n","Epoch 29/60\n","9/9 [==============================] - 0s 35ms/step - loss: 0.3496 - accuracy: 0.8415 - val_loss: 0.3543 - val_accuracy: 0.8028\n","Epoch 30/60\n","9/9 [==============================] - 0s 40ms/step - loss: 0.3512 - accuracy: 0.8275 - val_loss: 0.4070 - val_accuracy: 0.7887\n","Epoch 31/60\n","9/9 [==============================] - 0s 28ms/step - loss: 0.3476 - accuracy: 0.8134 - val_loss: 0.2978 - val_accuracy: 0.9155\n","Epoch 32/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.3246 - accuracy: 0.8275 - val_loss: 0.2865 - val_accuracy: 0.9296\n","Epoch 33/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.3283 - accuracy: 0.8662 - val_loss: 0.3079 - val_accuracy: 0.8310\n","Epoch 34/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.3164 - accuracy: 0.8556 - val_loss: 0.4392 - val_accuracy: 0.8028\n","Epoch 35/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.3153 - accuracy: 0.8486 - val_loss: 0.3293 - val_accuracy: 0.8310\n","Epoch 36/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.3461 - accuracy: 0.8415 - val_loss: 0.3007 - val_accuracy: 0.9155\n","Epoch 37/60\n","9/9 [==============================] - 0s 22ms/step - loss: 0.3288 - accuracy: 0.8451 - val_loss: 0.3310 - val_accuracy: 0.8310\n","Epoch 38/60\n","9/9 [==============================] - 0s 25ms/step - loss: 0.3282 - accuracy: 0.8310 - val_loss: 0.3204 - val_accuracy: 0.8592\n","Epoch 39/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.3088 - accuracy: 0.8239 - val_loss: 0.4563 - val_accuracy: 0.7887\n","Epoch 40/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.3047 - accuracy: 0.8486 - val_loss: 0.2776 - val_accuracy: 0.9014\n","Epoch 41/60\n","9/9 [==============================] - 0s 22ms/step - loss: 0.3029 - accuracy: 0.8380 - val_loss: 0.2660 - val_accuracy: 0.9155\n","Epoch 42/60\n","9/9 [==============================] - 0s 21ms/step - loss: 0.2952 - accuracy: 0.8380 - val_loss: 0.3441 - val_accuracy: 0.8592\n","Epoch 43/60\n","9/9 [==============================] - 0s 21ms/step - loss: 0.3133 - accuracy: 0.8451 - val_loss: 0.2432 - val_accuracy: 0.9296\n","Epoch 44/60\n","9/9 [==============================] - 0s 25ms/step - loss: 0.3076 - accuracy: 0.8345 - val_loss: 0.3246 - val_accuracy: 0.8451\n","Epoch 45/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.2683 - accuracy: 0.8768 - val_loss: 0.3763 - val_accuracy: 0.8310\n","Epoch 46/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.2869 - accuracy: 0.8592 - val_loss: 0.2761 - val_accuracy: 0.9014\n","Epoch 47/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.2748 - accuracy: 0.8768 - val_loss: 0.2327 - val_accuracy: 0.9296\n","Epoch 48/60\n","9/9 [==============================] - 0s 22ms/step - loss: 0.2771 - accuracy: 0.8627 - val_loss: 0.3769 - val_accuracy: 0.8451\n","Epoch 49/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.2532 - accuracy: 0.8908 - val_loss: 0.3743 - val_accuracy: 0.8592\n","Epoch 50/60\n","9/9 [==============================] - 0s 25ms/step - loss: 0.2605 - accuracy: 0.8838 - val_loss: 0.3267 - val_accuracy: 0.8592\n","Epoch 51/60\n","9/9 [==============================] - 0s 22ms/step - loss: 0.2753 - accuracy: 0.8662 - val_loss: 0.3762 - val_accuracy: 0.8732\n","Epoch 52/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.2641 - accuracy: 0.8627 - val_loss: 0.2680 - val_accuracy: 0.8873\n","Epoch 53/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.2692 - accuracy: 0.8944 - val_loss: 0.2402 - val_accuracy: 0.9014\n","Epoch 54/60\n","9/9 [==============================] - 0s 25ms/step - loss: 0.2545 - accuracy: 0.8627 - val_loss: 0.3228 - val_accuracy: 0.8169\n","Epoch 55/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.3025 - accuracy: 0.8592 - val_loss: 0.2707 - val_accuracy: 0.9155\n","Epoch 56/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.2981 - accuracy: 0.8451 - val_loss: 0.2414 - val_accuracy: 0.9155\n","Epoch 57/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.2904 - accuracy: 0.8592 - val_loss: 0.3366 - val_accuracy: 0.8873\n","Epoch 58/60\n","9/9 [==============================] - 0s 29ms/step - loss: 0.2521 - accuracy: 0.8627 - val_loss: 0.2486 - val_accuracy: 0.9014\n","Epoch 59/60\n","9/9 [==============================] - 0s 23ms/step - loss: 0.2530 - accuracy: 0.8732 - val_loss: 0.3621 - val_accuracy: 0.8732\n","Epoch 60/60\n","9/9 [==============================] - 0s 24ms/step - loss: 0.2533 - accuracy: 0.8838 - val_loss: 0.2578 - val_accuracy: 0.9014\n"]}]},{"cell_type":"code","source":["# Evaluate the model\n","test_loss, test_accuracy = model.evaluate(test_X, test_y)\n","print(\"Test Loss:\", test_loss)\n","print(\"Test Accuracy:\", test_accuracy)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"9-zQxYTLDfbi","executionInfo":{"status":"ok","timestamp":1692302229234,"user_tz":300,"elapsed":10,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"47d2073c-2ee9-420b-8c8e-19d33d346c15"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["3/3 [==============================] - 0s 12ms/step - loss: 0.3524 - accuracy: 0.8764\n","Test Loss: 0.3523918390274048\n","Test Accuracy: 0.8764045238494873\n"]}]},{"cell_type":"code","source":["# Plot training and validation metrics\n","plt.figure(figsize=(10, 4))\n","plt.subplot(1, 2, 1)\n","plt.plot(history.history['loss'], label='Training Loss')\n","plt.plot(history.history['val_loss'], label='Validation Loss')\n","plt.xlabel('Epoch')\n","plt.ylabel('Loss')\n","plt.legend()\n","\n","plt.subplot(1, 2, 2)\n","plt.plot(history.history['accuracy'], label='Training Accuracy')\n","plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n","plt.xlabel('Epoch')\n","plt.ylabel('Accuracy')\n","plt.legend()\n","\n","plt.tight_layout()\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":406},"id":"0MPVbGXzDf47","executionInfo":{"status":"ok","timestamp":1692302233675,"user_tz":300,"elapsed":946,"user":{"displayName":"Francesco Bassino","userId":"12214195385968794219"}},"outputId":"89493b24-7385-4ba2-90c5-295ff425b05f"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1000x400 with 2 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","source":["model.save('lstm_model_v01.h5')"],"metadata":{"id":"kJzUsHTqEbJZ"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":[],"metadata":{"id":"08qQAUfb9YvN"}}]}
\ No newline at end of file