AIoT_LSTM.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/yuting89830/Crowd_counting/blob/main/AIoT_LSTM.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0tmVjDFiqpuP"
      },
      "source": [
        "Thanks to https://machinelearningmastery.com/lstm-for-time-series-prediction-in-pytorch/"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "id": "RMLb6kGS-z97",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "37e366ef-8463-4138-e4ce-f5c2e0e01644"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yMNHFqNoVIWZ",
        "outputId": "2de7f556-1ec6-407e-a0ed-a8d0e44eae7f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Index(['Time', 'A metric'], dtype='object')\n",
            "Using device=device(type='cuda', index=0).\n"
          ]
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim\n",
        "import torch.utils.data as data\n",
        "\n",
        "df = pd.read_csv('/content/drive/MyDrive/colab用檔案/grafana_data_process.csv', engine='python')\n",
        "print(df.columns)\n",
        "# interpolation for \"Time\"\n",
        "df['Time'] = pd.to_datetime(df['Time'], yearfirst=True)\n",
        "for dd in df['Time'].index:\n",
        "  if dd%2 == 1:\n",
        "    df['Time'].iloc[dd].values = df['Time'].iloc[dd] + pd.DateOffset(minutes=15)\n",
        "\n",
        "timeseries = df[[\"A metric\"]].values.astype('float32')\n",
        "\n",
        "# train-test split for time series\n",
        "train_size = int(len(timeseries) * 0.67)\n",
        "test_size = len(timeseries) - train_size\n",
        "train, test = timeseries[:train_size], timeseries[train_size:]\n",
        "\n",
        "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
        "print(f'Using {device=}.')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "id": "U7BPn9RzVlBD"
      },
      "outputs": [],
      "source": [
        "def create_dataset(dataset, lookback):\n",
        "    \"\"\"Transform a time series into a prediction dataset\n",
        "\n",
        "    Args:\n",
        "        dataset: A numpy array of time series, first dimension is the time steps\n",
        "        lookback: Size of window for prediction\n",
        "    \"\"\"\n",
        "    X, y = [], []\n",
        "    for i in range(len(dataset)-lookback):\n",
        "        feature = dataset[i:i+lookback]\n",
        "        target = dataset[i+1:i+lookback+1]\n",
        "        X.append(feature)\n",
        "        y.append(target)\n",
        "    return torch.tensor(X), torch.tensor(y)\n",
        "\n",
        "# lookback = 4\n",
        "lookback = 64\n",
        "X_train, y_train = create_dataset(train, lookback=lookback)\n",
        "X_test, y_test = create_dataset(test, lookback=lookback)\n",
        "X_train, y_train = X_train.to(device), y_train.to(device)\n",
        "X_test, y_test = X_test.to(device), y_test.to(device)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "egBwRGkeVv4o",
        "outputId": "b0e0b8c5-40b8-4c19-e6cd-8280ac1b7556"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "simpleLSTM(\n",
              "  (lstm): LSTM(1, 50, batch_first=True)\n",
              "  (linear): Linear(in_features=50, out_features=1, bias=True)\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ],
      "source": [
        "class simpleLSTM(nn.Module):\n",
        "    def __init__(self):\n",
        "        super().__init__()\n",
        "        self.lstm = nn.LSTM(input_size=1, hidden_size=50, num_layers=1, batch_first=True)\n",
        "        self.linear = nn.Linear(50, 1)\n",
        "    def forward(self, x):\n",
        "        x, _ = self.lstm(x)\n",
        "        x = self.linear(x)\n",
        "        return x\n",
        "\n",
        "model = simpleLSTM()\n",
        "optimizer = optim.Adam(model.parameters())\n",
        "loss_fn = nn.MSELoss()\n",
        "loader = data.DataLoader(data.TensorDataset(X_train, y_train), shuffle=True, batch_size=256)\n",
        "\n",
        "model.to(device)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tBHLNEBHV_hH",
        "outputId": "3b6f01ba-f5d2-485c-c402-118e83e6b245"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 0: train RMSE 36.8715, test RMSE 51.7772\n",
            "Epoch 10: train RMSE 31.3872, test RMSE 44.7713\n",
            "Epoch 20: train RMSE 28.7193, test RMSE 40.9885\n",
            "Epoch 30: train RMSE 26.6409, test RMSE 37.9487\n",
            "Epoch 40: train RMSE 25.0064, test RMSE 35.6197\n",
            "Epoch 50: train RMSE 23.6309, test RMSE 33.7789\n",
            "Epoch 60: train RMSE 22.4297, test RMSE 32.1981\n",
            "Epoch 70: train RMSE 21.4011, test RMSE 30.9128\n",
            "Epoch 80: train RMSE 20.4877, test RMSE 29.8375\n",
            "Epoch 90: train RMSE 19.7201, test RMSE 28.9664\n",
            "Epoch 100: train RMSE 19.0529, test RMSE 28.3068\n",
            "Epoch 110: train RMSE 18.4835, test RMSE 27.8195\n",
            "Epoch 120: train RMSE 17.9796, test RMSE 27.3322\n",
            "Epoch 130: train RMSE 17.5307, test RMSE 27.0427\n",
            "Epoch 140: train RMSE 17.1265, test RMSE 26.8053\n",
            "Epoch 150: train RMSE 16.7544, test RMSE 26.7001\n",
            "Epoch 160: train RMSE 16.4085, test RMSE 26.5549\n",
            "Epoch 170: train RMSE 16.1076, test RMSE 26.4883\n",
            "Epoch 180: train RMSE 15.8207, test RMSE 26.4429\n",
            "Epoch 190: train RMSE 15.5900, test RMSE 26.4326\n",
            "Epoch 200: train RMSE 15.3804, test RMSE 26.4735\n",
            "Epoch 210: train RMSE 15.1407, test RMSE 26.5269\n",
            "Epoch 220: train RMSE 14.9659, test RMSE 26.5984\n",
            "Epoch 230: train RMSE 14.8277, test RMSE 26.5681\n",
            "Epoch 240: train RMSE 14.7383, test RMSE 26.5923\n",
            "Epoch 250: train RMSE 14.5322, test RMSE 26.6325\n",
            "Epoch 260: train RMSE 14.4631, test RMSE 26.5433\n",
            "Epoch 270: train RMSE 14.3852, test RMSE 26.5195\n",
            "Epoch 280: train RMSE 14.2608, test RMSE 26.6128\n",
            "Epoch 290: train RMSE 15.5238, test RMSE 26.8530\n",
            "Epoch 300: train RMSE 14.9093, test RMSE 26.1806\n",
            "Epoch 310: train RMSE 14.6194, test RMSE 26.1254\n",
            "Epoch 320: train RMSE 14.4475, test RMSE 26.1929\n",
            "Epoch 330: train RMSE 14.3203, test RMSE 26.2130\n",
            "Epoch 340: train RMSE 14.2127, test RMSE 26.2529\n",
            "Epoch 350: train RMSE 14.1314, test RMSE 26.3126\n",
            "Epoch 360: train RMSE 14.0350, test RMSE 26.3448\n",
            "Epoch 370: train RMSE 13.9603, test RMSE 26.3751\n",
            "Epoch 380: train RMSE 13.8958, test RMSE 26.4508\n",
            "Epoch 390: train RMSE 13.8208, test RMSE 26.5079\n",
            "Epoch 400: train RMSE 13.7651, test RMSE 26.5515\n",
            "Epoch 410: train RMSE 13.6865, test RMSE 26.5979\n",
            "Epoch 420: train RMSE 13.6152, test RMSE 26.6082\n",
            "Epoch 430: train RMSE 13.5511, test RMSE 26.6786\n",
            "Epoch 440: train RMSE 13.4832, test RMSE 26.7356\n",
            "Epoch 450: train RMSE 13.4248, test RMSE 26.7708\n",
            "Epoch 460: train RMSE 13.3406, test RMSE 26.8195\n",
            "Epoch 470: train RMSE 13.2439, test RMSE 26.9235\n",
            "Epoch 480: train RMSE 13.1824, test RMSE 26.9426\n",
            "Epoch 490: train RMSE 13.0830, test RMSE 26.9787\n",
            "Epoch 500: train RMSE 12.9960, test RMSE 27.0591\n",
            "Epoch 510: train RMSE 12.9141, test RMSE 27.1341\n",
            "Epoch 520: train RMSE 12.8352, test RMSE 27.2299\n",
            "Epoch 530: train RMSE 12.7572, test RMSE 27.2691\n",
            "Epoch 540: train RMSE 12.6715, test RMSE 27.3232\n",
            "Epoch 550: train RMSE 12.7863, test RMSE 27.4534\n",
            "Epoch 560: train RMSE 12.5596, test RMSE 27.4428\n",
            "Epoch 570: train RMSE 12.3942, test RMSE 27.6716\n",
            "Epoch 580: train RMSE 12.3530, test RMSE 27.7171\n",
            "Epoch 590: train RMSE 12.2727, test RMSE 27.8466\n",
            "Epoch 600: train RMSE 12.1597, test RMSE 27.9067\n",
            "Epoch 610: train RMSE 12.0937, test RMSE 28.0684\n",
            "Epoch 620: train RMSE 12.1082, test RMSE 28.2294\n",
            "Epoch 630: train RMSE 11.9595, test RMSE 28.2619\n",
            "Epoch 640: train RMSE 11.8982, test RMSE 28.3220\n",
            "Epoch 650: train RMSE 11.8728, test RMSE 28.3529\n",
            "Epoch 660: train RMSE 11.7929, test RMSE 28.4359\n",
            "Epoch 670: train RMSE 11.7507, test RMSE 28.3614\n",
            "Epoch 680: train RMSE 11.6350, test RMSE 28.7154\n",
            "Epoch 690: train RMSE 11.6241, test RMSE 28.8169\n"
          ]
        }
      ],
      "source": [
        "hist_loss, hist_train_rmse, hist_test_rmse = ([] for i in range(3))\n",
        "n_epochs = 700\n",
        "for epoch in range(n_epochs):\n",
        "  model.train()\n",
        "  for X_batch, y_batch in loader:\n",
        "    # print(X_batch.shape)\n",
        "    y_pred = model(X_batch)\n",
        "    loss = loss_fn(y_pred, y_batch)\n",
        "    optimizer.zero_grad()\n",
        "    loss.backward()\n",
        "    optimizer.step()\n",
        "    hist_loss.append(loss.item())\n",
        "  # Validation\n",
        "  if epoch % 10 != 0:\n",
        "    continue\n",
        "  model.eval()\n",
        "  with torch.no_grad():\n",
        "    y_pred = model(X_train)\n",
        "    train_rmse = np.sqrt(loss_fn(y_pred.cpu(), y_train.cpu()))\n",
        "    y_pred = model(X_test)\n",
        "    test_rmse = np.sqrt(loss_fn(y_pred.cpu(), y_test.cpu()))\n",
        "    hist_train_rmse.append(train_rmse)\n",
        "    hist_test_rmse.append(test_rmse)\n",
        "  print(\"Epoch %d: train RMSE %.4f, test RMSE %.4f\" % (epoch, train_rmse, test_rmse))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 676
        },
        "id": "2v9PIvpgjN6a",
        "outputId": "f0f0aa3c-4290-4951-997c-9a96b309e40c"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x800 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "fig, axes = plt.subplots(2,1,figsize=(10, 8))\n",
        "axes[0].plot(hist_loss)\n",
        "axes[0].legend(['MSE loss'])\n",
        "axes[1].plot(hist_train_rmse, label='train RMSE')\n",
        "axes[1].plot(hist_test_rmse, label='test RMSE')\n",
        "axes[1].legend()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "id": "hkC6rpTj-cQa",
        "outputId": "31663c4d-879e-40d8-e9b7-a2d66aeaeea1",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([2829, 64, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ],
      "source": [
        "with torch.no_grad():\n",
        "  ff = model(X_train)\n",
        "\n",
        "ff.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "id": "r0ArzR1NnpN2"
      },
      "outputs": [],
      "source": [
        "further_pred_length: int = 1000\n",
        "\n",
        "with torch.no_grad():\n",
        "  # shift train predictions for plotting\n",
        "  train_plot = np.ones_like(timeseries) * np.nan\n",
        "  y_pred = model(X_train)\n",
        "  y_pred = y_pred[:, -1, :]\n",
        "  train_plot[lookback:train_size] = model(X_train)[:, -1, :].detach().cpu().numpy()\n",
        "  # shift test predictions for plotting\n",
        "  # test_plot = np.ones_like(timeseries) * np.nan\n",
        "  # test_plot[train_size+lookback:len(timeseries)] = model(X_test)[:, -1, :].detach().cpu().numpy()\n",
        "  test_plot = np.ones(len(timeseries) + further_pred_length).reshape(-1, 1) * np.nan\n",
        "  # test_plot[train_size+lookback:len(timeseries) + further_pred_length] = model(X_test)[:, -1, :].detach().cpu().numpy()\n",
        "  test_plot[train_size+lookback:len(timeseries)] = model(X_test)[:, -1, :].detach().cpu().numpy()\n",
        "\n",
        "  mm_x_test = torch.ones(further_pred_length).reshape(-1, 1).to(device)\n",
        "  test_plot[len(timeseries):] = model(mm_x_test).detach().cpu().numpy()\n",
        "\n",
        "  # test_plot[train_size+lookback+2000:len(timeseries)] = model(X_test)[2000:, -1, :].detach().cpu().numpy()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "id": "zwA6bh46-cQb",
        "outputId": "0fb1ebcd-e22f-4a63-9254-9f1e153b08d1",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([1362, 64, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 29
        }
      ],
      "source": [
        "X_test.shape\n",
        "# X_test[2000:, -1, :].shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "id": "X6KbJb3I-cQb",
        "outputId": "b241e70b-3f3f-4fa7-8107-2ba120cb319d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([5319, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ],
      "source": [
        "# mm_x_test = torch.full_like(X_test.cpu(), 1).to(device)\n",
        "mm_x_test = torch.ones(len(timeseries) + further_pred_length).reshape(-1, 1).to(device)\n",
        "\n",
        "model(mm_x_test).shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "id": "9DDcTEfe-cQc",
        "outputId": "5a77c180-710e-495a-839c-d4c3051b83c2",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(5319, 1)"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ],
      "source": [
        "test_plot.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 428
        },
        "id": "JQooy2YSWDT8",
        "outputId": "3c68d655-b548-4943-ebb3-c9355368d5a0"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x400 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# plot\n",
        "fig, ax = plt.subplots(figsize=(10, 4))\n",
        "fig.suptitle('[LSTM] Crowd Forecasting')\n",
        "ax.set_title(f'RMSE={hist_test_rmse[-1]:.2f}, Length of Future prediced: {further_pred_length}')\n",
        "ax.set_xlabel('Time points')\n",
        "ax.set_ylabel('number of people')\n",
        "\n",
        "ax.plot(timeseries, label='GT')\n",
        "ax.plot(train_plot,  label='train pred', alpha=.8)\n",
        "ax.plot(test_plot, c='y', label='test pred (unseen sample)', alpha=.8)\n",
        "\n",
        "# ax.set_xticklabels(df['Time'])\n",
        "# plt.xticks(rotation=30)\n",
        "ax.legend()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pP0gYbdtrXY5"
      },
      "source": [
        "---\n",
        "\n",
        "\n",
        "# Seq2Seq 有BUG"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CHfKSe3pao-h",
        "outputId": "8631253d-4b18-4850-b729-31d9eff55a89"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-33-2b1511a8ab5a>:17: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  df['Time'].iloc[dd] = df['Time'].iloc[dd] + pd.DateOffset(minutes=15)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Using device=device(type='cuda', index=0).\n"
          ]
        }
      ],
      "source": [
        "import random\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim\n",
        "import torch.utils.data as data\n",
        "\n",
        "df = pd.read_csv('/content/drive/MyDrive/colab用檔案/grafana_data_export.csv', engine='python',header=1)\n",
        "\n",
        "# interpolation for \"Time\"\n",
        "df['Time'] = pd.to_datetime(df['Time'], yearfirst=True)\n",
        "for dd in df['Time'].index:\n",
        "  if dd%2 == 1:\n",
        "    df['Time'].iloc[dd] = df['Time'].iloc[dd] + pd.DateOffset(minutes=15)\n",
        "\n",
        "timeseries = df[[\"A metric\"]].values.astype('float32')\n",
        "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
        "print(f'Using {device=}.')\n",
        "train_ratio = 0.7\n",
        "n_past = 20"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "metadata": {
        "id": "pJGqo3tlrWTK",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "83800bf8-d2cc-4f23-8ff8-fb02e39961da"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "torch.Size([1820, 20, 1])\n",
            "torch.Size([1820, 2])\n"
          ]
        }
      ],
      "source": [
        "def preprocess(data_trend, train_ratio, n_past):\n",
        "\n",
        "    scaler = StandardScaler()\n",
        "    data_trend = scaler.fit_transform(data_trend)\n",
        "\n",
        "    train_ind = int(len(data_trend)*train_ratio)\n",
        "    train_data = data_trend[:train_ind]\n",
        "    val_data = data_trend[train_ind:]\n",
        "\n",
        "    # convert our train data into a pytorch tensor\n",
        "    X_train, Y_train = create_sequences(train_data, n_past)\n",
        "    X_val, Y_val = create_sequences(val_data, n_past)\n",
        "\n",
        "    return X_train, Y_train, X_val, Y_val, scaler\n",
        "\n",
        "\n",
        "def create_sequences(data, n_past):\n",
        "    X,Y = [],[]\n",
        "    L = len(data)\n",
        "    # for i in range(L-(n_past+5)):\n",
        "    for i in range(L-(n_past+2)):\n",
        "        X.append(data[i:i+n_past])\n",
        "        # Y.append(data[i+n_past:i+n_past+5][:,3])\n",
        "        Y.append(data[i+n_past:i+n_past+2][:, 0])  # 這裡原本是交易量的樣子，改成人數(0)\n",
        "        # Y.append(data[i+n_past:i+n_past+5][:, 0])  # 這裡原本是交易量的樣子，改成人數(0)\n",
        "\n",
        "    return torch.Tensor(np.array(X)), torch.Tensor(np.array(Y))\n",
        "\n",
        "data = df[[c for c in df.columns if c not in ['Time', 'Adj Close']]].values\n",
        "X_train, Y_train, X_val, Y_val, scaler = preprocess(data, train_ratio, n_past)\n",
        "\n",
        "batch_size = 128#32\n",
        "train_set = torch.utils.data.TensorDataset(X_train, Y_train)\n",
        "train_loader = torch.utils.data.DataLoader(train_set, batch_size=batch_size, shuffle=False)\n",
        "\n",
        "val_set = torch.utils.data.TensorDataset(X_val, Y_val)\n",
        "val_loader = torch.utils.data.DataLoader(val_set, batch_size=batch_size, shuffle=False)\n",
        "\n",
        "print(X_train.shape) # torch.Size([624, 20, 5])\n",
        "print(Y_train.shape) # torch.Size([624, 5])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "metadata": {
        "id": "WvWVlFe2q1Hp"
      },
      "outputs": [],
      "source": [
        "class Encoder(nn.Module):\n",
        "    def __init__(self, input_size=5, embedding_size=36, hidden_size=128, n_layers=1, dropout=0.3):\n",
        "        super().__init__()\n",
        "        self.hidden_size = hidden_size\n",
        "        self.n_layers = n_layers\n",
        "\n",
        "        self.lstm = nn.LSTM(input_size, hidden_size, n_layers)\n",
        "\n",
        "    def forward(self, x):\n",
        "        # x: input batch data, size: [input_seq_len, batch_size, feature_size]\n",
        "        output, (hidden, cell) = self.lstm(x)\n",
        "        return hidden, cell\n",
        "\n",
        "\n",
        "class Decoder(nn.Module):\n",
        "    def __init__(self, output_size=5, embedding_size=36, hidden_size=128, n_layers=1, dropout=0.3):\n",
        "        super().__init__()\n",
        "        self.output_size = output_size\n",
        "        self.hidden_size = hidden_size\n",
        "        self.n_layers = n_layers\n",
        "\n",
        "        self.lstm = nn.LSTM(output_size, hidden_size, n_layers)\n",
        "        self.linear = nn.Linear(hidden_size, output_size)\n",
        "\n",
        "    def forward(self, x, hidden, cell):\n",
        "        \"\"\"\n",
        "        x size = [batch_size, feature_size]\n",
        "        --> x only has two dimensions since the input is batch of last coordinate of observed trajectory\n",
        "        so the sequence length has been removed.\n",
        "        \"\"\"\n",
        "        # add sequence dimension to x, to allow use of nn.LSTM\n",
        "        x = x.unsqueeze(0)  # -->[1, batch_size, feature_size]\n",
        "\n",
        "        output, (hidden, cell) = self.lstm(x, (hidden, cell))\n",
        "        prediction = self.linear(output)\n",
        "\n",
        "        return prediction, hidden, cell"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "metadata": {
        "id": "niZUsEW9q6Ak",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0a69f5eb-646f-4f52-aea4-7008692da0e8"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Seq2Seq(\n",
              "  (encoder): Encoder(\n",
              "    (lstm): LSTM(2, 128, num_layers=2)\n",
              "  )\n",
              "  (decoder): Decoder(\n",
              "    (lstm): LSTM(2, 128, num_layers=2)\n",
              "    (linear): Linear(in_features=128, out_features=2, bias=True)\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 36
        }
      ],
      "source": [
        "class Seq2Seq(nn.Module):\n",
        "    def __init__(self, encoder, decoder, device):\n",
        "        super().__init__()\n",
        "        self.encoder = encoder\n",
        "        self.decoder = decoder\n",
        "        self.device = device\n",
        "\n",
        "        assert encoder.hidden_size == decoder.hidden_size, \\\n",
        "            \"Hidden dimensions of encoder and decoder must be equal!\"\n",
        "        assert encoder.n_layers == decoder.n_layers, \\\n",
        "            \"Encoder and decoder must have equal number of layers!\"\n",
        "\n",
        "    def forward(self, x, y, teacher_forcing_ratio=0.5):\n",
        "        # print(f'[DEBUG] {x.shape=}')\n",
        "        x = x.permute(1,0,2) #我們的 dataloader是 [batch, seq, dim]\n",
        "        # print(f'[DEBUG] {y.shape=}')\n",
        "        # y = y.permute(1,0,2) #但為了方便操作LSTM，直接把它擺成[seq,batch,dim]，output再把它擺回來\n",
        "        y = y.permute(1,0) #但為了方便操作LSTM，直接把它擺成[seq,batch,dim]，output再把它擺回來\n",
        "        \"\"\"\n",
        "        x = [input_seq_len, batch_size, feature_size]\n",
        "        y = [target_seq_len, batch_size, feature_size]\n",
        "        \"\"\"\n",
        "        batch_size = x.shape[1]\n",
        "        target_len = y.shape[0]\n",
        "\n",
        "        # tensor to store decoder outputs of each time step\n",
        "        outputs = torch.zeros(y.shape).to(self.device)\n",
        "\n",
        "        hidden, cell = self.encoder(x)\n",
        "        decoder_input = x[-1, :, :] # first input to decoder is last of x\n",
        "\n",
        "        for i in range(target_len):\n",
        "            output, hidden, cell = self.decoder(decoder_input, hidden, cell)\n",
        "            # place predictions in a tensor holding predictions for each time step\n",
        "            # print(f'{output[-1, -1, :].shape = }')\n",
        "            # print(f'{output[-1, :, :].shape = }')\n",
        "            print(f'{output[-1, -1, :] = }')\n",
        "            print(f'{output[-1, :, :] = }')\n",
        "            # print(f'{i = }')\n",
        "            print(f'{output = }')\n",
        "            print(f'{outputs = }')\n",
        "            print(f'{outputs[i] = }')\n",
        "            print(f'{outputs[:, i] = }')\n",
        "            # outputs[i] = torch.squeeze(output[-1, -1, :], 0)\n",
        "            # outputs[i] = torch.squeeze(output[-1, :, :], 0)\n",
        "            # outputs[i] = torch.squeeze(output, 0)\n",
        "            # outputs[:, i] = output[-1, -1, :]\n",
        "            outputs[i] = output[-1, -1, :]\n",
        "\n",
        "            teacher_forcing = random.random() < teacher_forcing_ratio\n",
        "            # output is the same shape as decorder input-->[batch_size, feature_size]\n",
        "            # so we use output directly as input or use true lable depending on teacher_forcing flag\n",
        "            decoder_input = y[i] if teacher_forcing else torch.squeeze(output,0)\n",
        "\n",
        "        # return outputs.permute(1,0,2)\n",
        "        return outputs.permute(1,0)\n",
        "\n",
        "\n",
        "myEncoder = Encoder(input_size=2, embedding_size=36, hidden_size=128, n_layers=2, dropout=0.3)\n",
        "myDecoder = Decoder(output_size=2, embedding_size=36, hidden_size=128, n_layers=2, dropout=0.3)\n",
        "\n",
        "model = Seq2Seq(myEncoder, myDecoder, device)\n",
        "model.to(device)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 37,
      "metadata": {
        "id": "pLtodE0aq7P3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 409
        },
        "outputId": "ef4cb5cc-f1d9-4296-835f-51ea4c1708be"
      },
      "outputs": [
        {
          "output_type": "error",
          "ename": "RuntimeError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mRuntimeError\u001b[0m                              Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-37-f20c7562581e>\u001b[0m in \u001b[0;36m<cell line: 5>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      8\u001b[0m     \u001b[0my\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munsqueeze\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      9\u001b[0m     \u001b[0mpseudo_y\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mones\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 10\u001b[0;31m     \u001b[0mpred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpseudo_y\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mteacher_forcing_ratio\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\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     11\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     12\u001b[0m     \u001b[0;31m# close_price_true.append(scaler.inverse_transform(y)[0,0,dim].item())\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/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1499\u001b[0m                 \u001b[0;32mor\u001b[0m \u001b[0m_global_backward_pre_hooks\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0m_global_backward_hooks\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1500\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\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   1502\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1503\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\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<ipython-input-36-5cc7b544b64e>\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x, y, teacher_forcing_ratio)\u001b[0m\n\u001b[1;32m     27\u001b[0m         \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzeros\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     28\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 29\u001b[0;31m         \u001b[0mhidden\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mencoder\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\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     30\u001b[0m         \u001b[0mdecoder_input\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;31m# first input to decoder is last of x\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     31\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1499\u001b[0m                 \u001b[0;32mor\u001b[0m \u001b[0m_global_backward_pre_hooks\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0m_global_backward_hooks\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1500\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\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   1502\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1503\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\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<ipython-input-35-1cae362a0435>\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m      9\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\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     10\u001b[0m         \u001b[0;31m# x: input batch data, size: [input_seq_len, batch_size, feature_size]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 11\u001b[0;31m         \u001b[0moutput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhidden\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlstm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\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     12\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mhidden\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     13\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1499\u001b[0m                 \u001b[0;32mor\u001b[0m \u001b[0m_global_backward_pre_hooks\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0m_global_backward_hooks\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1500\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\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   1502\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1503\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\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/torch/nn/modules/rnn.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input, hx)\u001b[0m\n\u001b[1;32m    808\u001b[0m             \u001b[0mhx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpermute_hidden\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msorted_indices\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    809\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 810\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcheck_forward_args\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_sizes\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    811\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mbatch_sizes\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    812\u001b[0m             result = _VF.lstm(input, hx, self._flat_weights, self.bias, self.num_layers,\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/torch/nn/modules/rnn.py\u001b[0m in \u001b[0;36mcheck_forward_args\u001b[0;34m(self, input, hidden, batch_sizes)\u001b[0m\n\u001b[1;32m    728\u001b[0m                            \u001b[0mbatch_sizes\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mOptional\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mTensor\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    729\u001b[0m                            ):\n\u001b[0;32m--> 730\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcheck_input\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_sizes\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    731\u001b[0m         self.check_hidden_size(hidden[0], self.get_expected_hidden_size(input, batch_sizes),\n\u001b[1;32m    732\u001b[0m                                'Expected hidden[0] size {}, got {}')\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/torch/nn/modules/rnn.py\u001b[0m in \u001b[0;36mcheck_input\u001b[0;34m(self, input, batch_sizes)\u001b[0m\n\u001b[1;32m    216\u001b[0m                     expected_input_dim, input.dim()))\n\u001b[1;32m    217\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minput_size\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\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--> 218\u001b[0;31m             raise RuntimeError(\n\u001b[0m\u001b[1;32m    219\u001b[0m                 'input.size(-1) must be equal to input_size. Expected {}, got {}'.format(\n\u001b[1;32m    220\u001b[0m                     self.input_size, input.size(-1)))\n",
            "\u001b[0;31mRuntimeError\u001b[0m: input.size(-1) must be equal to input_size. Expected 2, got 1"
          ]
        }
      ],
      "source": [
        "close_price_true = []\n",
        "close_price_pred = []\n",
        "\n",
        "dim = 1#3\n",
        "for i in range(len(val_set)):\n",
        "    x,y = val_set[i]\n",
        "    x = torch.unsqueeze(x,0)\n",
        "    y = torch.unsqueeze(y,0)\n",
        "    pseudo_y = torch.ones(y.shape)\n",
        "    pred = model(x.to(device), pseudo_y.to(device), teacher_forcing_ratio=0)\n",
        "\n",
        "    # close_price_true.append(scaler.inverse_transform(y)[0,0,dim].item())\n",
        "    close_price_true.append(scaler.inverse_transform(y)[0,dim].item())\n",
        "    close_price_pred.append(scaler.inverse_transform(pred.detach().cpu().numpy())[0,dim])\n",
        "\n",
        "mape = np.mean(np.absolute(np.array(close_price_true) - np.array(close_price_pred))/np.array(close_price_true))\n",
        "mae =  np.mean( np.absolute(np.array(close_price_true) - np.array(close_price_pred)))\n",
        "plt.plot(close_price_true)\n",
        "plt.plot(close_price_pred)\n",
        "plt.title('MAE:{:.3f}, MAPE:{:.3f}'.format(mae, mape))\n",
        "plt.legend(['true', 'predict'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Bw3YPeE4s5dq"
      },
      "outputs": [],
      "source": [
        "# ss = torch.tensor([[[-0.0782, -0.0997]]])\n",
        "ss = torch.tensor([[-0.0782, -0.0997]])\n",
        "print(f'{ss.shape = }')\n",
        "print(f'{ss = }')\n",
        "print(torch.squeeze(ss, 0).shape)\n",
        "print(torch.squeeze(ss, 0))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Aa1MbYk7L67M"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": [],
      "include_colab_link": true
    },
    "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": "3.8.10"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}