ANN_create_synthetic_data.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hzjoOR-Ax9dA"
      },
      "source": [
        "Reference: Vekuri, H., Tuovinen, JP., Kulmala, L. et al. A widely-used eddy covariance gap-filling method creates systematic bias in carbon balance estimates. Sci Rep 13, 1720 (2023). https://doi.org/10.1038/s41598-023-28827-2\n",
        "code availbility: https://github.com/hvekuri/co2_gapfilling"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "5ij4UANGx8wF"
      },
      "outputs": [],
      "source": [
        "from keras.layers import Dense, Dropout\n",
        "from keras.models import Sequential\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import random\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.utils import shuffle"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DoQ_kLzQqc6R",
        "outputId": "dc9b9ff9-9844-4c8d-93f1-c84719072326"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n",
            "site_data.csv\n",
            "sythetic_data_ANN.csv\n",
            "ANN_site_data.csv\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "<ipython-input-36-cbff4cacccb6>:14: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\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",
            "  data['DOY_sin'] = np.sin((data['Day']-1)*(2*np.pi/12))\n",
            "<ipython-input-36-cbff4cacccb6>:15: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\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",
            "  data['DOY_cos'] = np.cos((data['Day']-1)*(2*np.pi/12))\n",
            "<ipython-input-36-cbff4cacccb6>:16: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\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",
            "  data['Month_sin'] = np.sin((data['Month']-1)*(2*np.pi/12))\n",
            "<ipython-input-36-cbff4cacccb6>:17: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\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",
            "  data['Month_cos'] = np.cos((data['Month']-1)*(2*np.pi/12))\n",
            "<ipython-input-36-cbff4cacccb6>:18: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\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",
            "  data['Hour_sin'] = np.sin(data['Hour']*(2*np.pi/24))\n",
            "<ipython-input-36-cbff4cacccb6>:19: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\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",
            "  data['Hour_cos'] = np.cos(data['Hour']*(2*np.pi/24))\n"
          ]
        }
      ],
      "source": [
        "# mount google drive\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "\n",
        "from pathlib import Path\n",
        "site_data_dir = Path('/content/drive/MyDrive/synthetic_data/')\n",
        "# List all files in the directory\n",
        "file_list = [file for file in site_data_dir.iterdir() if file.is_file()]\n",
        "\n",
        "# Print the list of files\n",
        "for file in file_list:\n",
        "    print(file.name)\n",
        "def add_time_vars(data):\n",
        "  data['DOY_sin'] = np.sin((data['Day']-1)*(2*np.pi/12))\n",
        "  data['DOY_cos'] = np.cos((data['Day']-1)*(2*np.pi/12))\n",
        "  data['Month_sin'] = np.sin((data['Month']-1)*(2*np.pi/12))\n",
        "  data['Month_cos'] = np.cos((data['Month']-1)*(2*np.pi/12))\n",
        "  data['Hour_sin'] = np.sin(data['Hour']*(2*np.pi/24))\n",
        "  data['Hour_cos'] = np.cos(data['Hour']*(2*np.pi/24))\n",
        "  data.replace({-9999: np.nan}, inplace=True)\n",
        "  return data\n",
        "\n",
        "x_cols = ['TIMESTAMP_END','evi', 'gcc', 'SWC_1_f', 'SWC_4_f', 'Tair_f', 'Tsoil', 'VPD_f', 'PPFD_DIF', 'PPFD_DIR', 'PPFD_f', 'DOY_sin', 'DOY_cos', 'SW_IN_F']\n",
        "y_col = 'NEE_U50_orig'\n",
        "\n",
        "site_data = pd.read_csv(site_data_dir / \"site_data.csv\", index_col=0) # .dropna(subset=[y_col])\n",
        "site_data = site_data[site_data['Year'] > 2009]\n",
        "site_data = add_time_vars(site_data)\n",
        "site_data[\"SW_IN_F\"] = site_data['Rg']\n",
        "site_data[\"NEE\"] = site_data['NEE_U50_orig']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7wS1nZlTeLjV",
        "outputId": "bcc4a043-40b3-462f-a3fc-2737ec998cc1"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Tair_f     0\n",
            "Tsoil      0\n",
            "VPD_f      0\n",
            "SW_IN_F    0\n",
            "dtype: int64\n"
          ]
        }
      ],
      "source": [
        "# Define the x_cols\n",
        "x_cols = ['Tair_f', 'Tsoil', 'VPD_f', 'SW_IN_F']\n",
        "\n",
        "# Check for NaNs in the four variables\n",
        "nan_check = site_data[x_cols].isnull().sum()\n",
        "print(nan_check)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "xO6ZHb9BsWna"
      },
      "outputs": [],
      "source": [
        "# the codes are adapted fllowing Vekur et al. 2023\n",
        "random.seed(10)\n",
        "def generate_model(in_d):\n",
        "    \"\"\"\n",
        "    Model\n",
        "    \"\"\"\n",
        "    model = Sequential()\n",
        "    model.add(Dense(16, activation='linear',\n",
        "                    input_dim=in_d))\n",
        "    model.add(Dropout(0.2))\n",
        "    model.add(Dense(32, activation='tanh'))\n",
        "    model.add(Dropout(0.2))\n",
        "    model.add(Dense(32, activation='tanh'))\n",
        "    model.add(Dropout(0.2))\n",
        "    model.add(Dense(32, activation='relu'))\n",
        "    model.add(Dropout(0.2))\n",
        "    model.add(\n",
        "        Dense(1, activation=\"linear\"))\n",
        "    model.compile(optimizer='adam', loss='mean_squared_error')\n",
        "\n",
        "    return model\n",
        "\n",
        "def add_residuals(synth_data):\n",
        "    \"\"\"\n",
        "    Adds noise to synthetic NEE based on temperature\n",
        "    \"\"\"\n",
        "    synth_data['residual'] = (synth_data.NEE-synth_data.modelled_NEE).values\n",
        "    assignments, edges = pd.qcut(\n",
        "        synth_data['Tair_f'], 5, retbins=True, labels=False)\n",
        "    synth_data['label'] = assignments\n",
        "\n",
        "    for label in synth_data.label.unique():\n",
        "        cur = synth_data[(synth_data.label == label) & (\n",
        "            synth_data.residual.notnull())]\n",
        "        if len(cur) < 20:\n",
        "            cur = synth_data[(synth_data.label <= label+1) & (\n",
        "            synth_data.residual.notnull())]\n",
        "            if len(cur) < 20:\n",
        "                cur = synth_data[(synth_data.label <= label+2) & (\n",
        "                synth_data.residual.notnull())]\n",
        "                if len(cur) < 20:\n",
        "                    cur = synth_data[(synth_data.label <= label+3) & (\n",
        "                synth_data.residual.notnull())]\n",
        "\n",
        "        res = cur.residual.sample(\n",
        "            len(synth_data[(synth_data.label == label)]), replace=True).values\n",
        "\n",
        "        synth_data.loc[(synth_data.label == label),\n",
        "                           'synth_NEE'] = synth_data.loc[(synth_data.label == label), 'modelled_NEE'] + res\n",
        "\n",
        "    #new = synth_data[['synth_NEE', 'Tair_f', 'Tsoil', 'SW_IN_F', 'VPD_f']].copy()\n",
        "    new = synth_data[['synth_NEE', 'Tair_f', 'Tsoil', 'SW_IN_F', 'VPD_f', 'gcc']].copy()\n",
        "    # new.columns = ['NEE', 'Tair_f', 'Tsoil', 'SW_IN_F','VPD_f']\n",
        "    new.columns = ['NEE', 'Tair_f', 'Tsoil', 'SW_IN_F','VPD_f','gcc']\n",
        "    return new\n",
        "\n",
        "def make_synth_data(data, x_cols, y_col):\n",
        "    \"\"\"\n",
        "    Makes a synthetic data set using neural network\n",
        "    \"\"\"\n",
        "    all_train_data = data[(data.NEE.notnull())] # This is important\n",
        "    X = np.asarray(all_train_data[x_cols])\n",
        "    y = np.asarray(all_train_data[y_col])\n",
        "    X, y = shuffle(X, y)\n",
        "    scaler_x = StandardScaler().fit(X)\n",
        "    X = scaler_x.transform(X)\n",
        "    scaler_y = StandardScaler().fit(y.reshape(-1, 1))\n",
        "    y = scaler_y.transform(y.reshape(-1, 1))\n",
        "\n",
        "    X_all = scaler_x.transform(np.asarray(data[x_cols]))\n",
        "\n",
        "    model = generate_model(len(x_cols))\n",
        "    history = model.fit(X, y, epochs=200, batch_size=32)\n",
        "\n",
        "    pred_all_y = model.predict(X_all)\n",
        "    pred_all_y = scaler_y.inverse_transform(pred_all_y)\n",
        "\n",
        "    data['modelled_NEE'] = pred_all_y # then add noice to data\n",
        "\n",
        "    # Filter data for winter and summer months based on the 'Month' column\n",
        "    winter = data[data['Month'].isin([1, 2, 3, 4, 11, 12])].copy()\n",
        "    summer = data[data['Month'].isin([5, 6, 7, 8, 9, 10])].copy()\n",
        "\n",
        "    synth_data1 = add_residuals(winter[winter.SW_IN_F < 20].copy())\n",
        "    synth_data2 = add_residuals(winter[winter.SW_IN_F >= 20].copy())\n",
        "    synth_data3 = add_residuals(summer[summer.SW_IN_F < 20].copy())\n",
        "    synth_data4 = add_residuals(summer[summer.SW_IN_F >= 20].copy())\n",
        "    synth_data = pd.concat([synth_data1, synth_data2, synth_data3, synth_data4])\n",
        "    synth_data = synth_data.sort_index()\n",
        "\n",
        "    #return synth_data\n",
        "    return data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "kie7IzRh9slc",
        "outputId": "3ea232ef-e902-4166-e8b4-b2027f21dac8"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.3049\n",
            "Epoch 2/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2409\n",
            "Epoch 3/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2273\n",
            "Epoch 4/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2221\n",
            "Epoch 5/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2198\n",
            "Epoch 6/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2184\n",
            "Epoch 7/200\n",
            "2358/2358 [==============================] - 8s 3ms/step - loss: 0.2166\n",
            "Epoch 8/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2140\n",
            "Epoch 9/200\n",
            "2358/2358 [==============================] - 7s 3ms/step - loss: 0.2118\n",
            "Epoch 10/200\n",
            "2358/2358 [==============================] - 7s 3ms/step - loss: 0.2121\n",
            "Epoch 11/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2105\n",
            "Epoch 12/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2107\n",
            "Epoch 13/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2093\n",
            "Epoch 14/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2090\n",
            "Epoch 15/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2075\n",
            "Epoch 16/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2087\n",
            "Epoch 17/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2080\n",
            "Epoch 18/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2081\n",
            "Epoch 19/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2050\n",
            "Epoch 20/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2060\n",
            "Epoch 21/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2050\n",
            "Epoch 22/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2051\n",
            "Epoch 23/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2045\n",
            "Epoch 24/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2058\n",
            "Epoch 25/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2056\n",
            "Epoch 26/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2039\n",
            "Epoch 27/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2045\n",
            "Epoch 28/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2050\n",
            "Epoch 29/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2053\n",
            "Epoch 30/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2046\n",
            "Epoch 31/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2038\n",
            "Epoch 32/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2031\n",
            "Epoch 33/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2046\n",
            "Epoch 34/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2035\n",
            "Epoch 35/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2041\n",
            "Epoch 36/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2047\n",
            "Epoch 37/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2029\n",
            "Epoch 38/200\n",
            "2358/2358 [==============================] - 8s 3ms/step - loss: 0.2045\n",
            "Epoch 39/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2039\n",
            "Epoch 40/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2032\n",
            "Epoch 41/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2038\n",
            "Epoch 42/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2023\n",
            "Epoch 43/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2030\n",
            "Epoch 44/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2038\n",
            "Epoch 45/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2036\n",
            "Epoch 46/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2031\n",
            "Epoch 47/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2034\n",
            "Epoch 48/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2031\n",
            "Epoch 49/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2008\n",
            "Epoch 50/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2031\n",
            "Epoch 51/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2018\n",
            "Epoch 52/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2022\n",
            "Epoch 53/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2024\n",
            "Epoch 54/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2026\n",
            "Epoch 55/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2034\n",
            "Epoch 56/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2007\n",
            "Epoch 57/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2015\n",
            "Epoch 58/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2023\n",
            "Epoch 59/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2025\n",
            "Epoch 60/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2021\n",
            "Epoch 61/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2012\n",
            "Epoch 62/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2013\n",
            "Epoch 63/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2027\n",
            "Epoch 64/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2020\n",
            "Epoch 65/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2019\n",
            "Epoch 66/200\n",
            "2358/2358 [==============================] - 8s 3ms/step - loss: 0.2022\n",
            "Epoch 67/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2001\n",
            "Epoch 68/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2018\n",
            "Epoch 69/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2029\n",
            "Epoch 70/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2015\n",
            "Epoch 71/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2016\n",
            "Epoch 72/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2013\n",
            "Epoch 73/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2009\n",
            "Epoch 74/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2005\n",
            "Epoch 75/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2013\n",
            "Epoch 76/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2011\n",
            "Epoch 77/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.1999\n",
            "Epoch 78/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2029\n",
            "Epoch 79/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2021\n",
            "Epoch 80/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2021\n",
            "Epoch 81/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2020\n",
            "Epoch 82/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2013\n",
            "Epoch 83/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2006\n",
            "Epoch 84/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2007\n",
            "Epoch 85/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2014\n",
            "Epoch 86/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2020\n",
            "Epoch 87/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2018\n",
            "Epoch 88/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2024\n",
            "Epoch 89/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2005\n",
            "Epoch 90/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2023\n",
            "Epoch 91/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2013\n",
            "Epoch 92/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2010\n",
            "Epoch 93/200\n",
            "2358/2358 [==============================] - 7s 3ms/step - loss: 0.2009\n",
            "Epoch 94/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2010\n",
            "Epoch 95/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2026\n",
            "Epoch 96/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2017\n",
            "Epoch 97/200\n",
            "2358/2358 [==============================] - 7s 3ms/step - loss: 0.1998\n",
            "Epoch 98/200\n",
            "2358/2358 [==============================] - 8s 3ms/step - loss: 0.2016\n",
            "Epoch 99/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2018\n",
            "Epoch 100/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2015\n",
            "Epoch 101/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2028\n",
            "Epoch 102/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2009\n",
            "Epoch 103/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2007\n",
            "Epoch 104/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2030\n",
            "Epoch 105/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2015\n",
            "Epoch 106/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2020\n",
            "Epoch 107/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.1994\n",
            "Epoch 108/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2013\n",
            "Epoch 109/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2026\n",
            "Epoch 110/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2010\n",
            "Epoch 111/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2007\n",
            "Epoch 112/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2004\n",
            "Epoch 113/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2009\n",
            "Epoch 114/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.1999\n",
            "Epoch 115/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2008\n",
            "Epoch 116/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2005\n",
            "Epoch 117/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2008\n",
            "Epoch 118/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2008\n",
            "Epoch 119/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2004\n",
            "Epoch 120/200\n",
            "2358/2358 [==============================] - 8s 3ms/step - loss: 0.2007\n",
            "Epoch 121/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2016\n",
            "Epoch 122/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2011\n",
            "Epoch 123/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2011\n",
            "Epoch 124/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2002\n",
            "Epoch 125/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.1999\n",
            "Epoch 126/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2020\n",
            "Epoch 127/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2018\n",
            "Epoch 128/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.1993\n",
            "Epoch 129/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2003\n",
            "Epoch 130/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2011\n",
            "Epoch 131/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2019\n",
            "Epoch 132/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2018\n",
            "Epoch 133/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2009\n",
            "Epoch 134/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2007\n",
            "Epoch 135/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2008\n",
            "Epoch 136/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2005\n",
            "Epoch 137/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2016\n",
            "Epoch 138/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2004\n",
            "Epoch 139/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2012\n",
            "Epoch 140/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2021\n",
            "Epoch 141/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2014\n",
            "Epoch 142/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2001\n",
            "Epoch 143/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2002\n",
            "Epoch 144/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2005\n",
            "Epoch 145/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2003\n",
            "Epoch 146/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2001\n",
            "Epoch 147/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2012\n",
            "Epoch 148/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2015\n",
            "Epoch 149/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2007\n",
            "Epoch 150/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2016\n",
            "Epoch 151/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2009\n",
            "Epoch 152/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.1996\n",
            "Epoch 153/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2016\n",
            "Epoch 154/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.1999\n",
            "Epoch 155/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2016\n",
            "Epoch 156/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2001\n",
            "Epoch 157/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2011\n",
            "Epoch 158/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2009\n",
            "Epoch 159/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2013\n",
            "Epoch 160/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2015\n",
            "Epoch 161/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2003\n",
            "Epoch 162/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.1996\n",
            "Epoch 163/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.1993\n",
            "Epoch 164/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2005\n",
            "Epoch 165/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2002\n",
            "Epoch 166/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2007\n",
            "Epoch 167/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2017\n",
            "Epoch 168/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2005\n",
            "Epoch 169/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2008\n",
            "Epoch 170/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2006\n",
            "Epoch 171/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2001\n",
            "Epoch 172/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2002\n",
            "Epoch 173/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2010\n",
            "Epoch 174/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2004\n",
            "Epoch 175/200\n",
            "2358/2358 [==============================] - 7s 3ms/step - loss: 0.2005\n",
            "Epoch 176/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.2003\n",
            "Epoch 177/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2011\n",
            "Epoch 178/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.1996\n",
            "Epoch 179/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2009\n",
            "Epoch 180/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2010\n",
            "Epoch 181/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2002\n",
            "Epoch 182/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.1993\n",
            "Epoch 183/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.1993\n",
            "Epoch 184/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2001\n",
            "Epoch 185/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2016\n",
            "Epoch 186/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.1990\n",
            "Epoch 187/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2003\n",
            "Epoch 188/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2003\n",
            "Epoch 189/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2020\n",
            "Epoch 190/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.1997\n",
            "Epoch 191/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.1991\n",
            "Epoch 192/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2022\n",
            "Epoch 193/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.1993\n",
            "Epoch 194/200\n",
            "2358/2358 [==============================] - 6s 2ms/step - loss: 0.1995\n",
            "Epoch 195/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2007\n",
            "Epoch 196/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.2011\n",
            "Epoch 197/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2009\n",
            "Epoch 198/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.2012\n",
            "Epoch 199/200\n",
            "2358/2358 [==============================] - 6s 3ms/step - loss: 0.1998\n",
            "Epoch 200/200\n",
            "2358/2358 [==============================] - 5s 2ms/step - loss: 0.1999\n",
            "7122/7122 [==============================] - 10s 1ms/step\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "x_cols = ['Tair_f', 'Tsoil', 'VPD_f', 'SW_IN_F']\n",
        "x_cols = ['Tair_f', 'Tsoil', 'VPD_f', 'SW_IN_F', 'gcc', 'SWC_1_f', 'SWC_4_f']\n",
        "make_synth_data(site_data, x_cols, y_col)\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "# Create a figure and axis object\n",
        "fig, ax1 = plt.subplots()\n",
        "\n",
        "# Plot site_data[\"modelled_NEE\"] on the primary y-axis with transparent blue color\n",
        "ax1.scatter(site_data[\"Date\"], site_data[\"modelled_NEE\"], color='b', alpha=0.9)\n",
        "ax1.set_xlabel(\"Date\")\n",
        "ax1.set_ylabel(\"modelled_NEE\", color='b')\n",
        "\n",
        "# Create a secondary y-axis and plot site_data[\"NEE\"] on it with transparent red color\n",
        "ax2 = ax1.twinx()\n",
        "ax2.scatter(site_data[\"Date\"], site_data[\"NEE\"], color='r', alpha=0.9)\n",
        "ax2.set_ylabel(\"NEE\", color='r')\n",
        "\n",
        "# Add title\n",
        "plt.title(\"Scatter Plot of Date vs. modelled_NEE and NEE\")\n",
        "\n",
        "# Show the plot\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cxMFtYQCDImm",
        "outputId": "f55fd054-e3b7-433d-e775-a5681ca2828f"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "/content/drive/MyDrive/synthetic_data\n"
          ]
        }
      ],
      "source": [
        "import os\n",
        "print(site_data_dir)\n",
        "# Define the file path for saving\n",
        "csv_file_path = os.path.join(site_data_dir, \"ANN_site_data.csv\")\n",
        "# Save site_data to a CSV file\n",
        "site_data.to_csv(csv_file_path, index=False)"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}