TradingAI_LSTM_AppleDaily_v2.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"Copy of TradingAI_LSTM_AppleDaily.ipynb","provenance":[{"file_id":"https://github.com/whittlem/colabnotebooks/blob/main/TradingAI_LSTM_AppleDaily.ipynb","timestamp":1624542901675},{"file_id":"https://github.com/whittlem/colabnotebooks/blob/main/TradingAI_LSTM_AppleDaily.ipynb","timestamp":1624402052580},{"file_id":"https://github.com/whittlem/colabnotebooks/blob/main/TradingAI_LSTM_AppleDaily.ipynb","timestamp":1624381587865},{"file_id":"https://github.com/whittlem/colabnotebooks/blob/main/TradingAI_LSTM.ipynb","timestamp":1624273047252},{"file_id":"1-H0BWd-_4OydY8k7ZzL19TOSaJ1zrcBb","timestamp":1624269525430},{"file_id":"https://github.com/whittlem/colabnotebooks/blob/main/TradingAI_LSTM_2.ipynb","timestamp":1624057292632},{"file_id":"https://github.com/whittlem/colabnotebooks/blob/main/TradingAI_LSTM_1.ipynb","timestamp":1624029305263}],"collapsed_sections":[],"authorship_tag":"ABX9TyMyf2Y0V8O3e+68Dpjirg7a"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"accelerator":"GPU"},"cells":[{"cell_type":"code","metadata":{"id":"Tz-zgucvuT2m","executionInfo":{"status":"ok","timestamp":1624545402919,"user_tz":-60,"elapsed":2665,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}}},"source":["import io\n","import math\n","import keras\n","import pandas as pd\n","import numpy as np\n","import requests\n","import matplotlib.pyplot as plt\n","\n","from keras.models import Sequential, model_from_json\n","from keras.layers import Dense, LSTM, Dropout\n","from sklearn.preprocessing import MinMaxScaler\n","from sklearn.metrics import mean_squared_error\n","from sklearn.metrics import mean_absolute_error\n","from sklearn.metrics import classification_report\n","from sklearn.model_selection import train_test_split\n","\n","from google.colab import files\n","\n","plt.style.use('default')"],"execution_count":1,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":419},"id":"ues3ln6iOawo","executionInfo":{"status":"ok","timestamp":1624545403602,"user_tz":-60,"elapsed":686,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"75a62e97-20d9-4fd2-f8de-e403eee1c840"},"source":["# https://finance.yahoo.com/quote/AAPL/history?period1=345427200&period2=1624233600&interval=1d&filter=history&frequency=1d&includeAdjustedClose=true (Max Historical Apple Daily)\n","resp = requests.get(f'https://query1.finance.yahoo.com/v7/finance/download/AAPL?period1=345427200&period2=1624233600&interval=1d&events=history&includeAdjustedClose=true') # download link\n","csv_data = io.StringIO(resp.content.decode('utf8').replace(\"'\", '\"'))\n","df = pd.read_csv(csv_data, sep=',')\n","df"],"execution_count":2,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>Date</th>\n","      <th>Open</th>\n","      <th>High</th>\n","      <th>Low</th>\n","      <th>Close</th>\n","      <th>Adj Close</th>\n","      <th>Volume</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>1980-12-12</td>\n","      <td>0.128348</td>\n","      <td>0.128906</td>\n","      <td>0.128348</td>\n","      <td>0.128348</td>\n","      <td>0.100751</td>\n","      <td>469033600</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>1980-12-15</td>\n","      <td>0.122210</td>\n","      <td>0.122210</td>\n","      <td>0.121652</td>\n","      <td>0.121652</td>\n","      <td>0.095495</td>\n","      <td>175884800</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>1980-12-16</td>\n","      <td>0.113281</td>\n","      <td>0.113281</td>\n","      <td>0.112723</td>\n","      <td>0.112723</td>\n","      <td>0.088485</td>\n","      <td>105728000</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>1980-12-17</td>\n","      <td>0.115513</td>\n","      <td>0.116071</td>\n","      <td>0.115513</td>\n","      <td>0.115513</td>\n","      <td>0.090676</td>\n","      <td>86441600</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>1980-12-18</td>\n","      <td>0.118862</td>\n","      <td>0.119420</td>\n","      <td>0.118862</td>\n","      <td>0.118862</td>\n","      <td>0.093304</td>\n","      <td>73449600</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>10211</th>\n","      <td>2021-06-14</td>\n","      <td>127.820000</td>\n","      <td>130.539993</td>\n","      <td>127.070000</td>\n","      <td>130.479996</td>\n","      <td>130.479996</td>\n","      <td>96906500</td>\n","    </tr>\n","    <tr>\n","      <th>10212</th>\n","      <td>2021-06-15</td>\n","      <td>129.940002</td>\n","      <td>130.600006</td>\n","      <td>129.389999</td>\n","      <td>129.639999</td>\n","      <td>129.639999</td>\n","      <td>62746300</td>\n","    </tr>\n","    <tr>\n","      <th>10213</th>\n","      <td>2021-06-16</td>\n","      <td>130.369995</td>\n","      <td>130.889999</td>\n","      <td>128.460007</td>\n","      <td>130.149994</td>\n","      <td>130.149994</td>\n","      <td>91815000</td>\n","    </tr>\n","    <tr>\n","      <th>10214</th>\n","      <td>2021-06-17</td>\n","      <td>129.800003</td>\n","      <td>132.550003</td>\n","      <td>129.649994</td>\n","      <td>131.789993</td>\n","      <td>131.789993</td>\n","      <td>96721700</td>\n","    </tr>\n","    <tr>\n","      <th>10215</th>\n","      <td>2021-06-18</td>\n","      <td>130.710007</td>\n","      <td>131.509995</td>\n","      <td>130.240005</td>\n","      <td>130.460007</td>\n","      <td>130.460007</td>\n","      <td>108787300</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>10216 rows × 7 columns</p>\n","</div>"],"text/plain":["             Date        Open        High  ...       Close   Adj Close     Volume\n","0      1980-12-12    0.128348    0.128906  ...    0.128348    0.100751  469033600\n","1      1980-12-15    0.122210    0.122210  ...    0.121652    0.095495  175884800\n","2      1980-12-16    0.113281    0.113281  ...    0.112723    0.088485  105728000\n","3      1980-12-17    0.115513    0.116071  ...    0.115513    0.090676   86441600\n","4      1980-12-18    0.118862    0.119420  ...    0.118862    0.093304   73449600\n","...           ...         ...         ...  ...         ...         ...        ...\n","10211  2021-06-14  127.820000  130.539993  ...  130.479996  130.479996   96906500\n","10212  2021-06-15  129.940002  130.600006  ...  129.639999  129.639999   62746300\n","10213  2021-06-16  130.369995  130.889999  ...  130.149994  130.149994   91815000\n","10214  2021-06-17  129.800003  132.550003  ...  131.789993  131.789993   96721700\n","10215  2021-06-18  130.710007  131.509995  ...  130.460007  130.460007  108787300\n","\n","[10216 rows x 7 columns]"]},"metadata":{"tags":[]},"execution_count":2}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":419},"id":"8w3CQB0chlLJ","executionInfo":{"status":"ok","timestamp":1624545403603,"user_tz":-60,"elapsed":16,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"5361c020-ced1-4a41-b5ed-5d295d3a8532"},"source":["df['EMA12'] = df['Adj Close'].ewm(span=12, adjust=False).mean()\n","df['EMA26'] = df['Adj Close'].ewm(span=26, adjust=False).mean()\n","\n","df['MACD'] = df['EMA12'] - df['EMA26']\n","df['Signal'] = df['MACD'].ewm(span=9, adjust=False).mean() \n","\n","df['OBV'] = np.where(df['Adj Close'] == df['Adj Close'].shift(1), 0, np.where(df['Adj Close'] > df['Adj Close'].shift(1), df['Volume'], \n","        np.where(df['Adj Close'] < df['Adj Close'].shift(1), -df['Volume'], df.iloc[0]['Volume']))).cumsum() # on-balance volume\n","df['OBV Percent'] = df['OBV'].pct_change() * 100\n","df['OBV Percent'] = round(df['OBV Percent'].fillna(0), 2) # on-balance volume as a percentage\n","\n","df['Adj Close Delta'] = df['Adj Close'].diff().fillna(0) # delta between adj close\n","\n","df['Next Adj Close'] = df['Adj Close'].shift(-1)\n","df.dropna(inplace=True)\n","df"],"execution_count":3,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>Date</th>\n","      <th>Open</th>\n","      <th>High</th>\n","      <th>Low</th>\n","      <th>Close</th>\n","      <th>Adj Close</th>\n","      <th>Volume</th>\n","      <th>EMA12</th>\n","      <th>EMA26</th>\n","      <th>MACD</th>\n","      <th>Signal</th>\n","      <th>OBV</th>\n","      <th>OBV Percent</th>\n","      <th>Adj Close Delta</th>\n","      <th>Next Adj Close</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>1980-12-12</td>\n","      <td>0.128348</td>\n","      <td>0.128906</td>\n","      <td>0.128348</td>\n","      <td>0.128348</td>\n","      <td>0.100751</td>\n","      <td>469033600</td>\n","      <td>0.100751</td>\n","      <td>0.100751</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","      <td>469033600</td>\n","      <td>0.00</td>\n","      <td>0.000000</td>\n","      <td>0.095495</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>1980-12-15</td>\n","      <td>0.122210</td>\n","      <td>0.122210</td>\n","      <td>0.121652</td>\n","      <td>0.121652</td>\n","      <td>0.095495</td>\n","      <td>175884800</td>\n","      <td>0.099942</td>\n","      <td>0.100362</td>\n","      <td>-0.000419</td>\n","      <td>-0.000084</td>\n","      <td>293148800</td>\n","      <td>-37.50</td>\n","      <td>-0.005256</td>\n","      <td>0.088485</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>1980-12-16</td>\n","      <td>0.113281</td>\n","      <td>0.113281</td>\n","      <td>0.112723</td>\n","      <td>0.112723</td>\n","      <td>0.088485</td>\n","      <td>105728000</td>\n","      <td>0.098180</td>\n","      <td>0.099482</td>\n","      <td>-0.001302</td>\n","      <td>-0.000328</td>\n","      <td>187420800</td>\n","      <td>-36.07</td>\n","      <td>-0.007010</td>\n","      <td>0.090676</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>1980-12-17</td>\n","      <td>0.115513</td>\n","      <td>0.116071</td>\n","      <td>0.115513</td>\n","      <td>0.115513</td>\n","      <td>0.090676</td>\n","      <td>86441600</td>\n","      <td>0.097025</td>\n","      <td>0.098830</td>\n","      <td>-0.001804</td>\n","      <td>-0.000623</td>\n","      <td>273862400</td>\n","      <td>46.12</td>\n","      <td>0.002191</td>\n","      <td>0.093304</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>1980-12-18</td>\n","      <td>0.118862</td>\n","      <td>0.119420</td>\n","      <td>0.118862</td>\n","      <td>0.118862</td>\n","      <td>0.093304</td>\n","      <td>73449600</td>\n","      <td>0.096453</td>\n","      <td>0.098420</td>\n","      <td>-0.001968</td>\n","      <td>-0.000892</td>\n","      <td>347312000</td>\n","      <td>26.82</td>\n","      <td>0.002628</td>\n","      <td>0.098999</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>10210</th>\n","      <td>2021-06-11</td>\n","      <td>126.529999</td>\n","      <td>127.440002</td>\n","      <td>126.099998</td>\n","      <td>127.349998</td>\n","      <td>127.349998</td>\n","      <td>53522400</td>\n","      <td>126.201356</td>\n","      <td>126.598074</td>\n","      <td>-0.396718</td>\n","      <td>-0.720339</td>\n","      <td>161434764400</td>\n","      <td>0.03</td>\n","      <td>1.239997</td>\n","      <td>130.479996</td>\n","    </tr>\n","    <tr>\n","      <th>10211</th>\n","      <td>2021-06-14</td>\n","      <td>127.820000</td>\n","      <td>130.539993</td>\n","      <td>127.070000</td>\n","      <td>130.479996</td>\n","      <td>130.479996</td>\n","      <td>96906500</td>\n","      <td>126.859608</td>\n","      <td>126.885624</td>\n","      <td>-0.026015</td>\n","      <td>-0.581474</td>\n","      <td>161531670900</td>\n","      <td>0.06</td>\n","      <td>3.129998</td>\n","      <td>129.639999</td>\n","    </tr>\n","    <tr>\n","      <th>10212</th>\n","      <td>2021-06-15</td>\n","      <td>129.940002</td>\n","      <td>130.600006</td>\n","      <td>129.389999</td>\n","      <td>129.639999</td>\n","      <td>129.639999</td>\n","      <td>62746300</td>\n","      <td>127.287361</td>\n","      <td>127.089651</td>\n","      <td>0.197709</td>\n","      <td>-0.425637</td>\n","      <td>161468924600</td>\n","      <td>-0.04</td>\n","      <td>-0.839997</td>\n","      <td>130.149994</td>\n","    </tr>\n","    <tr>\n","      <th>10213</th>\n","      <td>2021-06-16</td>\n","      <td>130.369995</td>\n","      <td>130.889999</td>\n","      <td>128.460007</td>\n","      <td>130.149994</td>\n","      <td>130.149994</td>\n","      <td>91815000</td>\n","      <td>127.727766</td>\n","      <td>127.316344</td>\n","      <td>0.411422</td>\n","      <td>-0.258225</td>\n","      <td>161560739600</td>\n","      <td>0.06</td>\n","      <td>0.509995</td>\n","      <td>131.789993</td>\n","    </tr>\n","    <tr>\n","      <th>10214</th>\n","      <td>2021-06-17</td>\n","      <td>129.800003</td>\n","      <td>132.550003</td>\n","      <td>129.649994</td>\n","      <td>131.789993</td>\n","      <td>131.789993</td>\n","      <td>96721700</td>\n","      <td>128.352724</td>\n","      <td>127.647725</td>\n","      <td>0.704999</td>\n","      <td>-0.065581</td>\n","      <td>161657461300</td>\n","      <td>0.06</td>\n","      <td>1.639999</td>\n","      <td>130.460007</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>10215 rows × 15 columns</p>\n","</div>"],"text/plain":["             Date        Open  ...  Adj Close Delta  Next Adj Close\n","0      1980-12-12    0.128348  ...         0.000000        0.095495\n","1      1980-12-15    0.122210  ...        -0.005256        0.088485\n","2      1980-12-16    0.113281  ...        -0.007010        0.090676\n","3      1980-12-17    0.115513  ...         0.002191        0.093304\n","4      1980-12-18    0.118862  ...         0.002628        0.098999\n","...           ...         ...  ...              ...             ...\n","10210  2021-06-11  126.529999  ...         1.239997      130.479996\n","10211  2021-06-14  127.820000  ...         3.129998      129.639999\n","10212  2021-06-15  129.940002  ...        -0.839997      130.149994\n","10213  2021-06-16  130.369995  ...         0.509995      131.789993\n","10214  2021-06-17  129.800003  ...         1.639999      130.460007\n","\n","[10215 rows x 15 columns]"]},"metadata":{"tags":[]},"execution_count":3}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"vqYRlYZWdufJ","executionInfo":{"status":"ok","timestamp":1624545485248,"user_tz":-60,"elapsed":187,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"4a2a9133-2c27-4701-f2be-e4e480e967e3"},"source":["df_X = df[['Open','High','Low','Close','Volume','EMA12','EMA26','MACD','Signal','OBV Percent','Adj Close Delta']]\n","df_y = df[['Next Adj Close']]\n","\n","X_scaler = MinMaxScaler(feature_range=(0, 1))\n","X_scaled = X_scaler.fit_transform(df_X)\n","df_X = pd.DataFrame(X_scaled, columns=[df_X.columns])\n","\n","y_scaler = MinMaxScaler(feature_range=(0, 1))\n","y_scaled = y_scaler.fit_transform(df_y)\n","df_y = pd.DataFrame(y_scaled, columns=[df_y.columns])\n","\n","X_train, X_test, y_train, y_test = train_test_split(X_scaled, y_scaled, shuffle=False, stratify=None)\n","\n","X_train.shape, X_test.shape, y_train.shape, y_test.shape"],"execution_count":4,"outputs":[{"output_type":"execute_result","data":{"text/plain":["((7661, 11), (2554, 11), (7661, 1), (2554, 1))"]},"metadata":{"tags":[]},"execution_count":4}]},{"cell_type":"code","metadata":{"id":"GHGKVTvXaDKK","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1624545531642,"user_tz":-60,"elapsed":184,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"1edd856f-913a-4348-9638-34c7132a9b9b"},"source":["window = 1 # 1 day\n","\n","X_train_window=[]\n","y_train_window=[]\n","for i in range(window, len(X_train)):\n","  X_train_window.append(X_scaled[i-window:i, :])\n","  y_train_window.append(X_scaled[i, 0])\n","\n","X_train_window, y_train_window = np.array(X_train_window), np.array(y_train_window)\n","X_train_window = np.reshape(X_train_window, (X_train_window.shape[0], X_train_window.shape[1], X_train.shape[1]))\n","\n","print ('X_train_window.shape', X_train_window.shape)\n","print ('y_train_window.shape', y_train_window.shape)"],"execution_count":9,"outputs":[{"output_type":"stream","text":["X_train_window.shape (7660, 1, 11)\n","y_train_window.shape (7660,)\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"oML26KP1aEIh","colab":{"base_uri":"https://localhost:8080/","height":1000},"executionInfo":{"status":"ok","timestamp":1624545601987,"user_tz":-60,"elapsed":49933,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"b67e4063-dce5-4257-93f9-581f3dbab7cf"},"source":["def create_model():\n","  model = Sequential()\n","  model.add(LSTM(units=50, return_sequences=True, input_shape=(X_train_window.shape[1], X_train.shape[1]))) # layer 1 lstm\n","  model.add(Dropout(0.2)) # layer 1 dropout regularisation\n","  model.add(LSTM(units=50, return_sequences=True)) # layer 2 lstm\n","  model.add(Dropout(0.2)) # layer 2 dropout regularisation\n","  model.add(LSTM(units=50, return_sequences=True)) # layer 3 lstm\n","  model.add(Dropout(0.2)) # layer 3 dropout regularisation\n","  model.add(LSTM(units=50)) # layer 4 lstm\n","  model.add(Dropout(0.2)) # layer 4 dropout regularisation\n","  model.add(Dense(units=1)) # output layer\n","  model.compile(optimizer='adam', loss='mse', metrics=['mae']) # compile the rnn \n","  return model\n","\n","model = create_model()\n","history = model.fit(X_train_window, y_train_window, epochs=300, batch_size=1500, verbose=2)\n","model.summary()\n","\n","for metric in history.history:\n","  plt.title(metric)\n","  plt.plot(history.history[metric])\n","  plt.show()"],"execution_count":10,"outputs":[{"output_type":"stream","text":["Epoch 1/300\n","6/6 - 26s - loss: 3.0305e-04 - mae: 0.0104\n","Epoch 2/300\n","6/6 - 0s - loss: 2.8959e-04 - mae: 0.0125\n","Epoch 3/300\n","6/6 - 0s - loss: 2.8345e-04 - mae: 0.0097\n","Epoch 4/300\n","6/6 - 0s - loss: 2.7870e-04 - mae: 0.0110\n","Epoch 5/300\n","6/6 - 0s - loss: 2.7923e-04 - mae: 0.0116\n","Epoch 6/300\n","6/6 - 0s - loss: 2.7946e-04 - mae: 0.0104\n","Epoch 7/300\n","6/6 - 0s - loss: 2.7891e-04 - mae: 0.0106\n","Epoch 8/300\n","6/6 - 0s - loss: 2.7816e-04 - mae: 0.0112\n","Epoch 9/300\n","6/6 - 0s - loss: 2.7766e-04 - mae: 0.0108\n","Epoch 10/300\n","6/6 - 0s - loss: 2.7749e-04 - mae: 0.0106\n","Epoch 11/300\n","6/6 - 0s - loss: 2.7747e-04 - mae: 0.0113\n","Epoch 12/300\n","6/6 - 0s - loss: 2.7687e-04 - mae: 0.0111\n","Epoch 13/300\n","6/6 - 0s - loss: 2.7759e-04 - mae: 0.0103\n","Epoch 14/300\n","6/6 - 0s - loss: 2.7581e-04 - mae: 0.0109\n","Epoch 15/300\n","6/6 - 0s - loss: 2.7436e-04 - mae: 0.0107\n","Epoch 16/300\n","6/6 - 0s - loss: 2.7349e-04 - mae: 0.0110\n","Epoch 17/300\n","6/6 - 0s - loss: 2.7164e-04 - mae: 0.0106\n","Epoch 18/300\n","6/6 - 0s - loss: 2.6853e-04 - mae: 0.0107\n","Epoch 19/300\n","6/6 - 0s - loss: 2.6492e-04 - mae: 0.0106\n","Epoch 20/300\n","6/6 - 0s - loss: 2.5787e-04 - mae: 0.0106\n","Epoch 21/300\n","6/6 - 0s - loss: 2.4733e-04 - mae: 0.0102\n","Epoch 22/300\n","6/6 - 0s - loss: 2.2769e-04 - mae: 0.0094\n","Epoch 23/300\n","6/6 - 0s - loss: 1.9662e-04 - mae: 0.0087\n","Epoch 24/300\n","6/6 - 0s - loss: 1.4997e-04 - mae: 0.0075\n","Epoch 25/300\n","6/6 - 0s - loss: 9.2896e-05 - mae: 0.0057\n","Epoch 26/300\n","6/6 - 0s - loss: 6.5935e-05 - mae: 0.0053\n","Epoch 27/300\n","6/6 - 0s - loss: 5.6835e-05 - mae: 0.0051\n","Epoch 28/300\n","6/6 - 0s - loss: 4.0672e-05 - mae: 0.0042\n","Epoch 29/300\n","6/6 - 0s - loss: 3.3718e-05 - mae: 0.0038\n","Epoch 30/300\n","6/6 - 0s - loss: 2.7384e-05 - mae: 0.0033\n","Epoch 31/300\n","6/6 - 0s - loss: 2.0409e-05 - mae: 0.0029\n","Epoch 32/300\n","6/6 - 0s - loss: 1.6994e-05 - mae: 0.0027\n","Epoch 33/300\n","6/6 - 0s - loss: 1.4662e-05 - mae: 0.0024\n","Epoch 34/300\n","6/6 - 0s - loss: 1.3419e-05 - mae: 0.0023\n","Epoch 35/300\n","6/6 - 0s - loss: 1.2210e-05 - mae: 0.0022\n","Epoch 36/300\n","6/6 - 0s - loss: 1.1538e-05 - mae: 0.0021\n","Epoch 37/300\n","6/6 - 0s - loss: 1.1525e-05 - mae: 0.0019\n","Epoch 38/300\n","6/6 - 0s - loss: 1.0606e-05 - mae: 0.0019\n","Epoch 39/300\n","6/6 - 0s - loss: 1.0885e-05 - mae: 0.0019\n","Epoch 40/300\n","6/6 - 0s - loss: 9.6837e-06 - mae: 0.0018\n","Epoch 41/300\n","6/6 - 0s - loss: 9.3349e-06 - mae: 0.0018\n","Epoch 42/300\n","6/6 - 0s - loss: 1.0979e-05 - mae: 0.0019\n","Epoch 43/300\n","6/6 - 0s - loss: 1.0302e-05 - mae: 0.0018\n","Epoch 44/300\n","6/6 - 0s - loss: 1.0417e-05 - mae: 0.0018\n","Epoch 45/300\n","6/6 - 0s - loss: 1.0316e-05 - mae: 0.0018\n","Epoch 46/300\n","6/6 - 0s - loss: 1.0274e-05 - mae: 0.0018\n","Epoch 47/300\n","6/6 - 0s - loss: 1.1369e-05 - mae: 0.0020\n","Epoch 48/300\n","6/6 - 0s - loss: 1.0259e-05 - mae: 0.0019\n","Epoch 49/300\n","6/6 - 0s - loss: 1.0418e-05 - mae: 0.0018\n","Epoch 50/300\n","6/6 - 0s - loss: 1.0009e-05 - mae: 0.0018\n","Epoch 51/300\n","6/6 - 0s - loss: 9.3585e-06 - mae: 0.0017\n","Epoch 52/300\n","6/6 - 0s - loss: 9.9527e-06 - mae: 0.0018\n","Epoch 53/300\n","6/6 - 0s - loss: 9.2534e-06 - mae: 0.0018\n","Epoch 54/300\n","6/6 - 0s - loss: 9.3574e-06 - mae: 0.0017\n","Epoch 55/300\n","6/6 - 0s - loss: 9.8266e-06 - mae: 0.0018\n","Epoch 56/300\n","6/6 - 0s - loss: 8.9351e-06 - mae: 0.0016\n","Epoch 57/300\n","6/6 - 0s - loss: 9.3107e-06 - mae: 0.0017\n","Epoch 58/300\n","6/6 - 0s - loss: 8.4701e-06 - mae: 0.0016\n","Epoch 59/300\n","6/6 - 0s - loss: 8.6898e-06 - mae: 0.0016\n","Epoch 60/300\n","6/6 - 0s - loss: 9.2146e-06 - mae: 0.0017\n","Epoch 61/300\n","6/6 - 0s - loss: 8.7205e-06 - mae: 0.0016\n","Epoch 62/300\n","6/6 - 0s - loss: 9.2705e-06 - mae: 0.0016\n","Epoch 63/300\n","6/6 - 0s - loss: 9.5290e-06 - mae: 0.0017\n","Epoch 64/300\n","6/6 - 0s - loss: 8.6585e-06 - mae: 0.0016\n","Epoch 65/300\n","6/6 - 0s - loss: 8.7447e-06 - mae: 0.0016\n","Epoch 66/300\n","6/6 - 0s - loss: 9.4269e-06 - mae: 0.0017\n","Epoch 67/300\n","6/6 - 0s - loss: 9.3685e-06 - mae: 0.0018\n","Epoch 68/300\n","6/6 - 0s - loss: 9.4348e-06 - mae: 0.0017\n","Epoch 69/300\n","6/6 - 0s - loss: 9.3452e-06 - mae: 0.0017\n","Epoch 70/300\n","6/6 - 0s - loss: 8.9885e-06 - mae: 0.0017\n","Epoch 71/300\n","6/6 - 0s - loss: 8.3303e-06 - mae: 0.0016\n","Epoch 72/300\n","6/6 - 0s - loss: 8.6766e-06 - mae: 0.0016\n","Epoch 73/300\n","6/6 - 0s - loss: 8.1361e-06 - mae: 0.0016\n","Epoch 74/300\n","6/6 - 0s - loss: 9.2275e-06 - mae: 0.0017\n","Epoch 75/300\n","6/6 - 0s - loss: 8.6369e-06 - mae: 0.0016\n","Epoch 76/300\n","6/6 - 0s - loss: 7.7555e-06 - mae: 0.0015\n","Epoch 77/300\n","6/6 - 0s - loss: 7.8385e-06 - mae: 0.0015\n","Epoch 78/300\n","6/6 - 0s - loss: 8.1368e-06 - mae: 0.0016\n","Epoch 79/300\n","6/6 - 0s - loss: 8.1552e-06 - mae: 0.0015\n","Epoch 80/300\n","6/6 - 0s - loss: 7.9296e-06 - mae: 0.0015\n","Epoch 81/300\n","6/6 - 0s - loss: 8.3291e-06 - mae: 0.0016\n","Epoch 82/300\n","6/6 - 0s - loss: 8.7536e-06 - mae: 0.0016\n","Epoch 83/300\n","6/6 - 0s - loss: 7.8935e-06 - mae: 0.0016\n","Epoch 84/300\n","6/6 - 0s - loss: 7.9373e-06 - mae: 0.0016\n","Epoch 85/300\n","6/6 - 0s - loss: 7.8164e-06 - mae: 0.0016\n","Epoch 86/300\n","6/6 - 0s - loss: 9.0969e-06 - mae: 0.0017\n","Epoch 87/300\n","6/6 - 0s - loss: 8.7999e-06 - mae: 0.0016\n","Epoch 88/300\n","6/6 - 0s - loss: 8.5965e-06 - mae: 0.0015\n","Epoch 89/300\n","6/6 - 0s - loss: 8.3689e-06 - mae: 0.0016\n","Epoch 90/300\n","6/6 - 0s - loss: 8.3747e-06 - mae: 0.0015\n","Epoch 91/300\n","6/6 - 0s - loss: 7.7574e-06 - mae: 0.0015\n","Epoch 92/300\n","6/6 - 0s - loss: 7.7751e-06 - mae: 0.0015\n","Epoch 93/300\n","6/6 - 0s - loss: 7.4019e-06 - mae: 0.0015\n","Epoch 94/300\n","6/6 - 0s - loss: 7.7151e-06 - mae: 0.0015\n","Epoch 95/300\n","6/6 - 0s - loss: 8.7834e-06 - mae: 0.0015\n","Epoch 96/300\n","6/6 - 0s - loss: 8.2578e-06 - mae: 0.0016\n","Epoch 97/300\n","6/6 - 0s - loss: 7.8300e-06 - mae: 0.0017\n","Epoch 98/300\n","6/6 - 0s - loss: 8.4509e-06 - mae: 0.0016\n","Epoch 99/300\n","6/6 - 0s - loss: 8.6213e-06 - mae: 0.0016\n","Epoch 100/300\n","6/6 - 0s - loss: 8.3275e-06 - mae: 0.0016\n","Epoch 101/300\n","6/6 - 0s - loss: 7.4828e-06 - mae: 0.0015\n","Epoch 102/300\n","6/6 - 0s - loss: 7.6501e-06 - mae: 0.0015\n","Epoch 103/300\n","6/6 - 0s - loss: 7.5285e-06 - mae: 0.0015\n","Epoch 104/300\n","6/6 - 0s - loss: 7.2830e-06 - mae: 0.0015\n","Epoch 105/300\n","6/6 - 0s - loss: 7.0394e-06 - mae: 0.0015\n","Epoch 106/300\n","6/6 - 0s - loss: 7.6172e-06 - mae: 0.0015\n","Epoch 107/300\n","6/6 - 0s - loss: 7.1630e-06 - mae: 0.0014\n","Epoch 108/300\n","6/6 - 0s - loss: 7.4078e-06 - mae: 0.0015\n","Epoch 109/300\n","6/6 - 0s - loss: 7.5626e-06 - mae: 0.0015\n","Epoch 110/300\n","6/6 - 0s - loss: 7.6545e-06 - mae: 0.0015\n","Epoch 111/300\n","6/6 - 0s - loss: 8.1922e-06 - mae: 0.0016\n","Epoch 112/300\n","6/6 - 0s - loss: 7.5424e-06 - mae: 0.0015\n","Epoch 113/300\n","6/6 - 0s - loss: 7.3223e-06 - mae: 0.0015\n","Epoch 114/300\n","6/6 - 0s - loss: 7.1968e-06 - mae: 0.0015\n","Epoch 115/300\n","6/6 - 0s - loss: 7.1279e-06 - mae: 0.0014\n","Epoch 116/300\n","6/6 - 0s - loss: 7.6522e-06 - mae: 0.0015\n","Epoch 117/300\n","6/6 - 0s - loss: 7.4450e-06 - mae: 0.0016\n","Epoch 118/300\n","6/6 - 0s - loss: 7.2025e-06 - mae: 0.0015\n","Epoch 119/300\n","6/6 - 0s - loss: 6.8441e-06 - mae: 0.0014\n","Epoch 120/300\n","6/6 - 0s - loss: 8.0021e-06 - mae: 0.0016\n","Epoch 121/300\n","6/6 - 0s - loss: 7.2689e-06 - mae: 0.0016\n","Epoch 122/300\n","6/6 - 0s - loss: 7.5858e-06 - mae: 0.0015\n","Epoch 123/300\n","6/6 - 0s - loss: 7.7589e-06 - mae: 0.0016\n","Epoch 124/300\n","6/6 - 0s - loss: 8.2233e-06 - mae: 0.0017\n","Epoch 125/300\n","6/6 - 0s - loss: 7.4638e-06 - mae: 0.0016\n","Epoch 126/300\n","6/6 - 0s - loss: 7.2615e-06 - mae: 0.0015\n","Epoch 127/300\n","6/6 - 0s - loss: 6.7806e-06 - mae: 0.0015\n","Epoch 128/300\n","6/6 - 0s - loss: 7.2140e-06 - mae: 0.0014\n","Epoch 129/300\n","6/6 - 0s - loss: 7.3374e-06 - mae: 0.0015\n","Epoch 130/300\n","6/6 - 0s - loss: 7.5147e-06 - mae: 0.0015\n","Epoch 131/300\n","6/6 - 0s - loss: 7.6853e-06 - mae: 0.0015\n","Epoch 132/300\n","6/6 - 0s - loss: 6.8617e-06 - mae: 0.0015\n","Epoch 133/300\n","6/6 - 0s - loss: 7.0393e-06 - mae: 0.0014\n","Epoch 134/300\n","6/6 - 0s - loss: 7.0137e-06 - mae: 0.0015\n","Epoch 135/300\n","6/6 - 0s - loss: 7.1620e-06 - mae: 0.0015\n","Epoch 136/300\n","6/6 - 0s - loss: 6.4326e-06 - mae: 0.0014\n","Epoch 137/300\n","6/6 - 0s - loss: 6.7606e-06 - mae: 0.0014\n","Epoch 138/300\n","6/6 - 0s - loss: 6.4776e-06 - mae: 0.0014\n","Epoch 139/300\n","6/6 - 0s - loss: 6.9681e-06 - mae: 0.0014\n","Epoch 140/300\n","6/6 - 0s - loss: 7.0439e-06 - mae: 0.0014\n","Epoch 141/300\n","6/6 - 0s - loss: 7.3582e-06 - mae: 0.0015\n","Epoch 142/300\n","6/6 - 0s - loss: 7.2478e-06 - mae: 0.0015\n","Epoch 143/300\n","6/6 - 0s - loss: 7.1557e-06 - mae: 0.0014\n","Epoch 144/300\n","6/6 - 0s - loss: 7.2524e-06 - mae: 0.0015\n","Epoch 145/300\n","6/6 - 0s - loss: 7.4771e-06 - mae: 0.0014\n","Epoch 146/300\n","6/6 - 0s - loss: 6.5548e-06 - mae: 0.0014\n","Epoch 147/300\n","6/6 - 0s - loss: 6.7627e-06 - mae: 0.0014\n","Epoch 148/300\n","6/6 - 0s - loss: 6.6002e-06 - mae: 0.0014\n","Epoch 149/300\n","6/6 - 0s - loss: 7.2315e-06 - mae: 0.0014\n","Epoch 150/300\n","6/6 - 0s - loss: 7.3287e-06 - mae: 0.0015\n","Epoch 151/300\n","6/6 - 0s - loss: 6.6532e-06 - mae: 0.0014\n","Epoch 152/300\n","6/6 - 0s - loss: 7.5455e-06 - mae: 0.0015\n","Epoch 153/300\n","6/6 - 0s - loss: 6.5561e-06 - mae: 0.0014\n","Epoch 154/300\n","6/6 - 0s - loss: 6.6577e-06 - mae: 0.0014\n","Epoch 155/300\n","6/6 - 0s - loss: 6.8417e-06 - mae: 0.0014\n","Epoch 156/300\n","6/6 - 0s - loss: 7.5557e-06 - mae: 0.0016\n","Epoch 157/300\n","6/6 - 0s - loss: 7.4477e-06 - mae: 0.0015\n","Epoch 158/300\n","6/6 - 0s - loss: 6.9675e-06 - mae: 0.0015\n","Epoch 159/300\n","6/6 - 0s - loss: 6.9453e-06 - mae: 0.0015\n","Epoch 160/300\n","6/6 - 0s - loss: 6.7077e-06 - mae: 0.0015\n","Epoch 161/300\n","6/6 - 0s - loss: 7.1964e-06 - mae: 0.0016\n","Epoch 162/300\n","6/6 - 0s - loss: 7.2304e-06 - mae: 0.0015\n","Epoch 163/300\n","6/6 - 0s - loss: 7.2247e-06 - mae: 0.0015\n","Epoch 164/300\n","6/6 - 0s - loss: 6.0544e-06 - mae: 0.0014\n","Epoch 165/300\n","6/6 - 0s - loss: 7.4958e-06 - mae: 0.0015\n","Epoch 166/300\n","6/6 - 0s - loss: 6.5870e-06 - mae: 0.0015\n","Epoch 167/300\n","6/6 - 0s - loss: 6.2152e-06 - mae: 0.0014\n","Epoch 168/300\n","6/6 - 0s - loss: 6.6464e-06 - mae: 0.0015\n","Epoch 169/300\n","6/6 - 0s - loss: 6.9978e-06 - mae: 0.0015\n","Epoch 170/300\n","6/6 - 0s - loss: 6.4581e-06 - mae: 0.0014\n","Epoch 171/300\n","6/6 - 0s - loss: 6.7839e-06 - mae: 0.0014\n","Epoch 172/300\n","6/6 - 0s - loss: 6.5464e-06 - mae: 0.0015\n","Epoch 173/300\n","6/6 - 0s - loss: 6.6728e-06 - mae: 0.0015\n","Epoch 174/300\n","6/6 - 0s - loss: 7.1207e-06 - mae: 0.0016\n","Epoch 175/300\n","6/6 - 0s - loss: 6.2870e-06 - mae: 0.0015\n","Epoch 176/300\n","6/6 - 0s - loss: 6.8580e-06 - mae: 0.0015\n","Epoch 177/300\n","6/6 - 0s - loss: 6.8139e-06 - mae: 0.0014\n","Epoch 178/300\n","6/6 - 0s - loss: 6.4927e-06 - mae: 0.0015\n","Epoch 179/300\n","6/6 - 0s - loss: 6.5294e-06 - mae: 0.0014\n","Epoch 180/300\n","6/6 - 0s - loss: 6.6232e-06 - mae: 0.0014\n","Epoch 181/300\n","6/6 - 0s - loss: 6.2687e-06 - mae: 0.0014\n","Epoch 182/300\n","6/6 - 0s - loss: 6.6189e-06 - mae: 0.0014\n","Epoch 183/300\n","6/6 - 0s - loss: 6.5063e-06 - mae: 0.0014\n","Epoch 184/300\n","6/6 - 0s - loss: 6.2263e-06 - mae: 0.0014\n","Epoch 185/300\n","6/6 - 0s - loss: 5.9794e-06 - mae: 0.0014\n","Epoch 186/300\n","6/6 - 0s - loss: 7.0659e-06 - mae: 0.0014\n","Epoch 187/300\n","6/6 - 0s - loss: 5.9196e-06 - mae: 0.0014\n","Epoch 188/300\n","6/6 - 0s - loss: 6.2798e-06 - mae: 0.0014\n","Epoch 189/300\n","6/6 - 0s - loss: 5.8652e-06 - mae: 0.0014\n","Epoch 190/300\n","6/6 - 0s - loss: 6.0479e-06 - mae: 0.0014\n","Epoch 191/300\n","6/6 - 0s - loss: 6.8389e-06 - mae: 0.0014\n","Epoch 192/300\n","6/6 - 0s - loss: 6.2753e-06 - mae: 0.0013\n","Epoch 193/300\n","6/6 - 0s - loss: 6.1119e-06 - mae: 0.0013\n","Epoch 194/300\n","6/6 - 0s - loss: 6.5182e-06 - mae: 0.0013\n","Epoch 195/300\n","6/6 - 0s - loss: 6.5211e-06 - mae: 0.0014\n","Epoch 196/300\n","6/6 - 0s - loss: 6.8959e-06 - mae: 0.0014\n","Epoch 197/300\n","6/6 - 0s - loss: 6.5801e-06 - mae: 0.0016\n","Epoch 198/300\n","6/6 - 0s - loss: 6.8804e-06 - mae: 0.0016\n","Epoch 199/300\n","6/6 - 0s - loss: 7.0229e-06 - mae: 0.0016\n","Epoch 200/300\n","6/6 - 0s - loss: 6.9015e-06 - mae: 0.0015\n","Epoch 201/300\n","6/6 - 0s - loss: 7.1317e-06 - mae: 0.0015\n","Epoch 202/300\n","6/6 - 0s - loss: 6.4427e-06 - mae: 0.0015\n","Epoch 203/300\n","6/6 - 0s - loss: 7.1583e-06 - mae: 0.0015\n","Epoch 204/300\n","6/6 - 0s - loss: 6.8346e-06 - mae: 0.0015\n","Epoch 205/300\n","6/6 - 0s - loss: 7.3815e-06 - mae: 0.0015\n","Epoch 206/300\n","6/6 - 0s - loss: 6.7796e-06 - mae: 0.0014\n","Epoch 207/300\n","6/6 - 0s - loss: 6.6494e-06 - mae: 0.0015\n","Epoch 208/300\n","6/6 - 0s - loss: 6.4854e-06 - mae: 0.0014\n","Epoch 209/300\n","6/6 - 0s - loss: 6.5953e-06 - mae: 0.0014\n","Epoch 210/300\n","6/6 - 0s - loss: 6.5122e-06 - mae: 0.0015\n","Epoch 211/300\n","6/6 - 0s - loss: 5.9361e-06 - mae: 0.0014\n","Epoch 212/300\n","6/6 - 0s - loss: 6.5706e-06 - mae: 0.0015\n","Epoch 213/300\n","6/6 - 0s - loss: 6.1177e-06 - mae: 0.0014\n","Epoch 214/300\n","6/6 - 0s - loss: 6.1755e-06 - mae: 0.0014\n","Epoch 215/300\n","6/6 - 0s - loss: 6.6078e-06 - mae: 0.0014\n","Epoch 216/300\n","6/6 - 0s - loss: 6.0559e-06 - mae: 0.0013\n","Epoch 217/300\n","6/6 - 0s - loss: 6.9266e-06 - mae: 0.0016\n","Epoch 218/300\n","6/6 - 0s - loss: 6.1126e-06 - mae: 0.0015\n","Epoch 219/300\n","6/6 - 0s - loss: 6.6698e-06 - mae: 0.0014\n","Epoch 220/300\n","6/6 - 0s - loss: 6.5734e-06 - mae: 0.0014\n","Epoch 221/300\n","6/6 - 0s - loss: 6.7927e-06 - mae: 0.0014\n","Epoch 222/300\n","6/6 - 0s - loss: 6.4725e-06 - mae: 0.0014\n","Epoch 223/300\n","6/6 - 0s - loss: 5.8455e-06 - mae: 0.0013\n","Epoch 224/300\n","6/6 - 0s - loss: 6.3823e-06 - mae: 0.0014\n","Epoch 225/300\n","6/6 - 0s - loss: 6.7381e-06 - mae: 0.0014\n","Epoch 226/300\n","6/6 - 0s - loss: 6.5431e-06 - mae: 0.0014\n","Epoch 227/300\n","6/6 - 0s - loss: 6.6092e-06 - mae: 0.0014\n","Epoch 228/300\n","6/6 - 0s - loss: 5.9491e-06 - mae: 0.0013\n","Epoch 229/300\n","6/6 - 0s - loss: 5.8552e-06 - mae: 0.0013\n","Epoch 230/300\n","6/6 - 0s - loss: 6.0436e-06 - mae: 0.0013\n","Epoch 231/300\n","6/6 - 0s - loss: 5.7563e-06 - mae: 0.0013\n","Epoch 232/300\n","6/6 - 0s - loss: 5.3911e-06 - mae: 0.0013\n","Epoch 233/300\n","6/6 - 0s - loss: 6.2420e-06 - mae: 0.0013\n","Epoch 234/300\n","6/6 - 0s - loss: 5.5357e-06 - mae: 0.0013\n","Epoch 235/300\n","6/6 - 0s - loss: 6.3219e-06 - mae: 0.0013\n","Epoch 236/300\n","6/6 - 0s - loss: 6.7757e-06 - mae: 0.0014\n","Epoch 237/300\n","6/6 - 0s - loss: 5.9376e-06 - mae: 0.0013\n","Epoch 238/300\n","6/6 - 0s - loss: 6.1114e-06 - mae: 0.0013\n","Epoch 239/300\n","6/6 - 0s - loss: 6.6291e-06 - mae: 0.0014\n","Epoch 240/300\n","6/6 - 0s - loss: 6.6250e-06 - mae: 0.0013\n","Epoch 241/300\n","6/6 - 0s - loss: 6.9545e-06 - mae: 0.0014\n","Epoch 242/300\n","6/6 - 0s - loss: 6.1233e-06 - mae: 0.0014\n","Epoch 243/300\n","6/6 - 0s - loss: 5.9657e-06 - mae: 0.0013\n","Epoch 244/300\n","6/6 - 0s - loss: 6.2823e-06 - mae: 0.0013\n","Epoch 245/300\n","6/6 - 0s - loss: 5.9108e-06 - mae: 0.0013\n","Epoch 246/300\n","6/6 - 0s - loss: 6.4838e-06 - mae: 0.0014\n","Epoch 247/300\n","6/6 - 0s - loss: 6.0475e-06 - mae: 0.0014\n","Epoch 248/300\n","6/6 - 0s - loss: 6.7643e-06 - mae: 0.0014\n","Epoch 249/300\n","6/6 - 0s - loss: 5.8291e-06 - mae: 0.0013\n","Epoch 250/300\n","6/6 - 0s - loss: 5.6803e-06 - mae: 0.0013\n","Epoch 251/300\n","6/6 - 0s - loss: 6.2865e-06 - mae: 0.0014\n","Epoch 252/300\n","6/6 - 0s - loss: 6.2537e-06 - mae: 0.0014\n","Epoch 253/300\n","6/6 - 0s - loss: 5.7215e-06 - mae: 0.0014\n","Epoch 254/300\n","6/6 - 0s - loss: 6.0614e-06 - mae: 0.0014\n","Epoch 255/300\n","6/6 - 0s - loss: 6.0263e-06 - mae: 0.0014\n","Epoch 256/300\n","6/6 - 0s - loss: 6.2082e-06 - mae: 0.0014\n","Epoch 257/300\n","6/6 - 0s - loss: 6.0904e-06 - mae: 0.0014\n","Epoch 258/300\n","6/6 - 0s - loss: 5.7514e-06 - mae: 0.0013\n","Epoch 259/300\n","6/6 - 0s - loss: 5.4714e-06 - mae: 0.0013\n","Epoch 260/300\n","6/6 - 0s - loss: 6.1786e-06 - mae: 0.0013\n","Epoch 261/300\n","6/6 - 0s - loss: 6.5507e-06 - mae: 0.0014\n","Epoch 262/300\n","6/6 - 0s - loss: 6.3122e-06 - mae: 0.0015\n","Epoch 263/300\n","6/6 - 0s - loss: 6.1735e-06 - mae: 0.0014\n","Epoch 264/300\n","6/6 - 0s - loss: 6.0872e-06 - mae: 0.0013\n","Epoch 265/300\n","6/6 - 0s - loss: 5.7613e-06 - mae: 0.0013\n","Epoch 266/300\n","6/6 - 0s - loss: 6.3465e-06 - mae: 0.0013\n","Epoch 267/300\n","6/6 - 0s - loss: 6.3090e-06 - mae: 0.0015\n","Epoch 268/300\n","6/6 - 0s - loss: 6.8904e-06 - mae: 0.0016\n","Epoch 269/300\n","6/6 - 0s - loss: 6.6083e-06 - mae: 0.0015\n","Epoch 270/300\n","6/6 - 0s - loss: 7.3264e-06 - mae: 0.0017\n","Epoch 271/300\n","6/6 - 0s - loss: 6.4673e-06 - mae: 0.0015\n","Epoch 272/300\n","6/6 - 0s - loss: 7.3889e-06 - mae: 0.0016\n","Epoch 273/300\n","6/6 - 0s - loss: 6.4298e-06 - mae: 0.0014\n","Epoch 274/300\n","6/6 - 0s - loss: 6.6729e-06 - mae: 0.0014\n","Epoch 275/300\n","6/6 - 0s - loss: 5.9366e-06 - mae: 0.0013\n","Epoch 276/300\n","6/6 - 0s - loss: 6.1976e-06 - mae: 0.0014\n","Epoch 277/300\n","6/6 - 0s - loss: 5.8419e-06 - mae: 0.0013\n","Epoch 278/300\n","6/6 - 0s - loss: 5.7110e-06 - mae: 0.0013\n","Epoch 279/300\n","6/6 - 0s - loss: 5.6597e-06 - mae: 0.0013\n","Epoch 280/300\n","6/6 - 0s - loss: 5.7958e-06 - mae: 0.0013\n","Epoch 281/300\n","6/6 - 0s - loss: 5.8176e-06 - mae: 0.0013\n","Epoch 282/300\n","6/6 - 0s - loss: 6.3461e-06 - mae: 0.0014\n","Epoch 283/300\n","6/6 - 0s - loss: 5.9082e-06 - mae: 0.0013\n","Epoch 284/300\n","6/6 - 0s - loss: 5.7443e-06 - mae: 0.0013\n","Epoch 285/300\n","6/6 - 0s - loss: 5.7805e-06 - mae: 0.0013\n","Epoch 286/300\n","6/6 - 0s - loss: 5.8853e-06 - mae: 0.0014\n","Epoch 287/300\n","6/6 - 0s - loss: 5.9652e-06 - mae: 0.0014\n","Epoch 288/300\n","6/6 - 0s - loss: 6.9054e-06 - mae: 0.0015\n","Epoch 289/300\n","6/6 - 0s - loss: 5.7163e-06 - mae: 0.0013\n","Epoch 290/300\n","6/6 - 0s - loss: 6.4535e-06 - mae: 0.0013\n","Epoch 291/300\n","6/6 - 0s - loss: 5.7018e-06 - mae: 0.0013\n","Epoch 292/300\n","6/6 - 0s - loss: 6.3671e-06 - mae: 0.0014\n","Epoch 293/300\n","6/6 - 0s - loss: 6.0793e-06 - mae: 0.0013\n","Epoch 294/300\n","6/6 - 0s - loss: 5.4270e-06 - mae: 0.0013\n","Epoch 295/300\n","6/6 - 0s - loss: 5.9496e-06 - mae: 0.0013\n","Epoch 296/300\n","6/6 - 0s - loss: 6.1314e-06 - mae: 0.0014\n","Epoch 297/300\n","6/6 - 0s - loss: 5.8395e-06 - mae: 0.0014\n","Epoch 298/300\n","6/6 - 0s - loss: 6.4321e-06 - mae: 0.0014\n","Epoch 299/300\n","6/6 - 0s - loss: 5.9247e-06 - mae: 0.0013\n","Epoch 300/300\n","6/6 - 0s - loss: 5.9081e-06 - mae: 0.0013\n","Model: \"sequential\"\n","_________________________________________________________________\n","Layer (type)                 Output Shape              Param #   \n","=================================================================\n","lstm (LSTM)                  (None, 1, 50)             12400     \n","_________________________________________________________________\n","dropout (Dropout)            (None, 1, 50)             0         \n","_________________________________________________________________\n","lstm_1 (LSTM)                (None, 1, 50)             20200     \n","_________________________________________________________________\n","dropout_1 (Dropout)          (None, 1, 50)             0         \n","_________________________________________________________________\n","lstm_2 (LSTM)                (None, 1, 50)             20200     \n","_________________________________________________________________\n","dropout_2 (Dropout)          (None, 1, 50)             0         \n","_________________________________________________________________\n","lstm_3 (LSTM)                (None, 50)                20200     \n","_________________________________________________________________\n","dropout_3 (Dropout)          (None, 50)                0         \n","_________________________________________________________________\n","dense (Dense)                (None, 1)                 51        \n","=================================================================\n","Total params: 73,051\n","Trainable params: 73,051\n","Non-trainable params: 0\n","_________________________________________________________________\n"],"name":"stdout"},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[],"needs_background":"light"}},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[],"needs_background":"light"}}]},{"cell_type":"code","metadata":{"id":"_Pwg5momQELY","colab":{"base_uri":"https://localhost:8080/","height":85},"executionInfo":{"status":"ok","timestamp":1624281332400,"user_tz":-60,"elapsed":263,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"9fe06eb2-15bd-4267-d2fb-469616970539"},"source":["'''Optionally save model, model.json and weights.h5'''\n","\n","model_filename = f'model_apple_daily_{len(X_train.columns)}-inputs.json'\n","weights_filename = f'weights_apple_daily_{len(X_train.columns)}-inputs.h5'\n","\n","!ls /content\n","\n","# save structure to json\n","model_json = model.to_json()\n","with open(model_filename, 'w') as json_file:\n","  json_file.write(model_json)\n","\n","# save weights to hdf5\n","model.save_weights(weights_filename)\n","\n","files.download(f'/content/{model_filename}')\n","files.download(f'/content/{weights_filename}')"],"execution_count":null,"outputs":[{"output_type":"stream","text":[" model_apple_daily_12-inputs.json       weights_apple_daily_12-inputs.h5\n","'model_apple_daily_5-inputs (1).json'  'weights_apple_daily_5-inputs (1).h5'\n"," model_apple_daily_5-inputs.json        weights_apple_daily_5-inputs.h5\n"," sample_data\n"],"name":"stdout"},{"output_type":"display_data","data":{"application/javascript":["\n","    async function download(id, filename, size) {\n","      if (!google.colab.kernel.accessAllowed) {\n","        return;\n","      }\n","      const div = document.createElement('div');\n","      const label = document.createElement('label');\n","      label.textContent = `Downloading \"${filename}\": `;\n","      div.appendChild(label);\n","      const progress = document.createElement('progress');\n","      progress.max = size;\n","      div.appendChild(progress);\n","      document.body.appendChild(div);\n","\n","      const buffers = [];\n","      let downloaded = 0;\n","\n","      const channel = await google.colab.kernel.comms.open(id);\n","      // Send a message to notify the kernel that we're ready.\n","      channel.send({})\n","\n","      for await (const message of channel.messages) {\n","        // Send a message to notify the kernel that we're ready.\n","        channel.send({})\n","        if (message.buffers) {\n","          for (const buffer of message.buffers) {\n","            buffers.push(buffer);\n","            downloaded += buffer.byteLength;\n","            progress.value = downloaded;\n","          }\n","        }\n","      }\n","      const blob = new Blob(buffers, {type: 'application/binary'});\n","      const a = document.createElement('a');\n","      a.href = window.URL.createObjectURL(blob);\n","      a.download = filename;\n","      div.appendChild(a);\n","      a.click();\n","      div.remove();\n","    }\n","  "],"text/plain":["<IPython.core.display.Javascript object>"]},"metadata":{"tags":[]}},{"output_type":"display_data","data":{"application/javascript":["download(\"download_dc6e2362-f456-438f-847f-bf1d1554e50b\", \"model_apple_daily_12-inputs.json\", 4655)"],"text/plain":["<IPython.core.display.Javascript object>"]},"metadata":{"tags":[]}},{"output_type":"display_data","data":{"application/javascript":["\n","    async function download(id, filename, size) {\n","      if (!google.colab.kernel.accessAllowed) {\n","        return;\n","      }\n","      const div = document.createElement('div');\n","      const label = document.createElement('label');\n","      label.textContent = `Downloading \"${filename}\": `;\n","      div.appendChild(label);\n","      const progress = document.createElement('progress');\n","      progress.max = size;\n","      div.appendChild(progress);\n","      document.body.appendChild(div);\n","\n","      const buffers = [];\n","      let downloaded = 0;\n","\n","      const channel = await google.colab.kernel.comms.open(id);\n","      // Send a message to notify the kernel that we're ready.\n","      channel.send({})\n","\n","      for await (const message of channel.messages) {\n","        // Send a message to notify the kernel that we're ready.\n","        channel.send({})\n","        if (message.buffers) {\n","          for (const buffer of message.buffers) {\n","            buffers.push(buffer);\n","            downloaded += buffer.byteLength;\n","            progress.value = downloaded;\n","          }\n","        }\n","      }\n","      const blob = new Blob(buffers, {type: 'application/binary'});\n","      const a = document.createElement('a');\n","      a.href = window.URL.createObjectURL(blob);\n","      a.download = filename;\n","      div.appendChild(a);\n","      a.click();\n","      div.remove();\n","    }\n","  "],"text/plain":["<IPython.core.display.Javascript object>"]},"metadata":{"tags":[]}},{"output_type":"display_data","data":{"application/javascript":["download(\"download_22b81e38-86b2-44c7-82d1-c73c81fcabe1\", \"weights_apple_daily_12-inputs.h5\", 315032)"],"text/plain":["<IPython.core.display.Javascript object>"]},"metadata":{"tags":[]}}]},{"cell_type":"code","metadata":{"colab":{"resources":{"http://localhost:8080/nbextensions/google.colab/files.js":{"data":"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","ok":true,"headers":[["content-type","application/javascript"]],"status":200,"status_text":""}},"base_uri":"https://localhost:8080/","height":157},"id":"Zs50-wzE1fZE","executionInfo":{"status":"ok","timestamp":1624278452191,"user_tz":-60,"elapsed":21020,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"6dec0ae3-43ca-40d1-c4e8-3d851ee935bb"},"source":["'''Optionally load model, model.json and weights.h5'''\n","\n","model_filename = f'model_apple_daily_{len(X_train.columns)}-inputs.json'\n","weights_filename = f'weights_apple_daily_{len(X_train.columns)}-inputs.h5'\n","\n","try:\n","  files.upload()\n","  !ls /content\n","\n","  # read structure from json\n","  model = open(model_filename, 'r')\n","  json = model.read()\n","  model.close()\n","  model = model_from_json(json)\n","\n","  # read weights from hdf5\n","  model.load_weights(f'/content/{weights_filename}')\n","except Exception as e:\n","  print (e)"],"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/html":["\n","     <input type=\"file\" id=\"files-14974d46-9cca-4c6c-afe7-b3b5723d3266\" name=\"files[]\" multiple disabled\n","        style=\"border:none\" />\n","     <output id=\"result-14974d46-9cca-4c6c-afe7-b3b5723d3266\">\n","      Upload widget is only available when the cell has been executed in the\n","      current browser session. Please rerun this cell to enable.\n","      </output>\n","      <script src=\"/nbextensions/google.colab/files.js\"></script> "],"text/plain":["<IPython.core.display.HTML object>"]},"metadata":{"tags":[]}},{"output_type":"stream","text":["Saving model_apple_daily_5-inputs.json to model_apple_daily_5-inputs (1).json\n","Saving weights_apple_daily_5-inputs.h5 to weights_apple_daily_5-inputs (1).h5\n","'model_apple_daily_5-inputs (1).json'  'weights_apple_daily_5-inputs (1).h5'\n"," model_apple_daily_5-inputs.json        weights_apple_daily_5-inputs.h5\n"," sample_data\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"LQP2ZA0txrhZ","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1624545930185,"user_tz":-60,"elapsed":2671,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"c8c0accf-5ef1-45aa-ab33-b236b8880723"},"source":["X_test_window=[]\n","y_test_window=[]\n","for i in range(window, len(X_test)):\n","  X_test_window.append(X_scaled[i-window:i, :])\n","  y_test_window.append(X_scaled[i, 0])\n","\n","X_test_window, y_test_window = np.array(X_test_window), np.array(y_test_window)\n","X_test_window = np.reshape(X_test_window, (X_test_window.shape[0], X_test_window.shape[1], X_test.shape[1]))\n","\n","print ('X_test_window.shape:', X_test_window.shape)\n","print ('y_test_window.shape:', y_test_window.shape)\n","\n","y_pred = model.predict(X_test_window)\n","y_pred = y_scaler.inverse_transform(y_pred)\n","\n","print ('y_pred.shape:', y_pred.shape)"],"execution_count":11,"outputs":[{"output_type":"stream","text":["X_test_window.shape: (2553, 1, 11)\n","y_test_window.shape: (2553,)\n","y_pred.shape: (2553, 1)\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"lA6Dqqllf3Cv","executionInfo":{"status":"ok","timestamp":1624546188026,"user_tz":-60,"elapsed":368,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"5d6b8010-24e9-402a-af67-95f93131b10e"},"source":["y_pred.min(), y_pred.max()"],"execution_count":16,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(0.18338653, 0.6345592)"]},"metadata":{"tags":[]},"execution_count":16}]},{"cell_type":"code","metadata":{"id":"xgkK5E9K5Dn5","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1624545930186,"user_tz":-60,"elapsed":5,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"408b2815-0f8a-443a-82b9-93f8a42694ce"},"source":["print ('ae:', mean_absolute_error(df['Adj Close'][-len(y_pred):].values, y_pred))\n","print ('mse:', mean_squared_error(df['Adj Close'][-len(y_pred):].values, y_pred, squared=False))\n","print ('rmse:', mean_squared_error(df['Adj Close'][-len(y_pred):].values, y_pred))"],"execution_count":12,"outputs":[{"output_type":"stream","text":["ae: 39.052390141324715\n","mse: 49.899118142615926\n","rmse: 2489.9219914107416\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"0ci9sG1zcLle","colab":{"base_uri":"https://localhost:8080/","height":670},"executionInfo":{"status":"ok","timestamp":1624545996593,"user_tz":-60,"elapsed":6153,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"c322fb9c-b9af-4f9c-d2a4-1743d193a869"},"source":["plt.figure(figsize=(30,10))\n","plt.plot(df['Date'].tail(len(y_pred)), df['Adj Close'][-len(y_pred):].values, color='red', label=f'Actual Apple Daily {df_y.columns.values[0][0].title()}')\n","plt.plot(df['Date'].tail(len(y_pred)), y_pred, color='blue', label=f'Predicted Apple Daily {df_y.columns.values[0][0].title()}')\n","plt.title(f'Apple Daily {df_y.columns.values[0][0].title()} Prediction')\n","plt.xlabel('Time')\n","plt.ylabel(f'{df_y.columns.values[0][0].title()}')\n","plt.xticks(rotation=90)\n","\n","ax = plt.gca()\n","for index, label in enumerate(ax.xaxis.get_ticklabels()):\n","  if index % 92 != 0:\n","    label.set_visible(False)\n","\n","plt.legend()\n","plt.show()"],"execution_count":15,"outputs":[{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 2160x720 with 1 Axes>"]},"metadata":{"tags":[],"needs_background":"light"}}]},{"cell_type":"code","metadata":{"id":"XZlO5cxit3oR","colab":{"base_uri":"https://localhost:8080/","height":419},"executionInfo":{"status":"ok","timestamp":1624402166883,"user_tz":-60,"elapsed":14,"user":{"displayName":"Michael Whittle","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gih6jGZ_ZDRChRiZZo2DwHoBerm3MdEG4NWtpSGYg=s64","userId":"16245925105622995030"}},"outputId":"df0a6b31-7d12-40e8-da1b-b651545c6cad"},"source":["pd.set_option('mode.chained_assignment', None)\n","\n","df_test = df.tail(len(y_pred))\n","df_test.loc[:, 'Predicted Next Adj Close'] = y_pred\n","df_test[['Open','High','Low','Close','Volume','EMA12','EMA26','MACD','Signal','OBV Percent','Adj Close Delta','Next Adj Close','Next Adj Close']]\n","df_test.rename(columns={'Next Adj Close': 'Actual Next Adj Close'}, errors='raise', inplace=True)\n","df_test"],"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>Date</th>\n","      <th>Open</th>\n","      <th>High</th>\n","      <th>Low</th>\n","      <th>Close</th>\n","      <th>Adj Close</th>\n","      <th>Volume</th>\n","      <th>EMA12</th>\n","      <th>EMA26</th>\n","      <th>MACD</th>\n","      <th>Signal</th>\n","      <th>OBV</th>\n","      <th>OBV Percent</th>\n","      <th>Adj Close Delta</th>\n","      <th>Actual Next Adj Close</th>\n","      <th>Predicted Next Adj Close</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>7662</th>\n","      <td>2011-04-27</td>\n","      <td>12.580000</td>\n","      <td>12.583929</td>\n","      <td>12.396429</td>\n","      <td>12.505357</td>\n","      <td>10.753392</td>\n","      <td>356213200</td>\n","      <td>10.553949</td>\n","      <td>10.518959</td>\n","      <td>0.034989</td>\n","      <td>-0.033985</td>\n","      <td>158941624800</td>\n","      <td>-0.22</td>\n","      <td>-0.008291</td>\n","      <td>10.648976</td>\n","      <td>0.178104</td>\n","    </tr>\n","    <tr>\n","      <th>7663</th>\n","      <td>2011-04-28</td>\n","      <td>12.363929</td>\n","      <td>12.491071</td>\n","      <td>12.340000</td>\n","      <td>12.383929</td>\n","      <td>10.648976</td>\n","      <td>360959200</td>\n","      <td>10.568568</td>\n","      <td>10.528590</td>\n","      <td>0.039978</td>\n","      <td>-0.019192</td>\n","      <td>158580665600</td>\n","      <td>-0.23</td>\n","      <td>-0.104416</td>\n","      <td>10.752778</td>\n","      <td>0.187542</td>\n","    </tr>\n","    <tr>\n","      <th>7664</th>\n","      <td>2011-04-29</td>\n","      <td>12.385000</td>\n","      <td>12.641071</td>\n","      <td>12.381071</td>\n","      <td>12.504643</td>\n","      <td>10.752778</td>\n","      <td>1006345200</td>\n","      <td>10.596908</td>\n","      <td>10.545197</td>\n","      <td>0.051712</td>\n","      <td>-0.005011</td>\n","      <td>159587010800</td>\n","      <td>0.63</td>\n","      <td>0.103802</td>\n","      <td>10.634542</td>\n","      <td>0.077263</td>\n","    </tr>\n","    <tr>\n","      <th>7665</th>\n","      <td>2011-05-02</td>\n","      <td>12.490714</td>\n","      <td>12.516786</td>\n","      <td>12.339286</td>\n","      <td>12.367143</td>\n","      <td>10.634542</td>\n","      <td>442713600</td>\n","      <td>10.602698</td>\n","      <td>10.551815</td>\n","      <td>0.050883</td>\n","      <td>0.006168</td>\n","      <td>159144297200</td>\n","      <td>-0.28</td>\n","      <td>-0.118236</td>\n","      <td>10.693504</td>\n","      <td>0.691720</td>\n","    </tr>\n","    <tr>\n","      <th>7666</th>\n","      <td>2011-05-03</td>\n","      <td>12.428214</td>\n","      <td>12.496071</td>\n","      <td>12.343571</td>\n","      <td>12.435714</td>\n","      <td>10.693504</td>\n","      <td>313348000</td>\n","      <td>10.616668</td>\n","      <td>10.562310</td>\n","      <td>0.054358</td>\n","      <td>0.015806</td>\n","      <td>159457645200</td>\n","      <td>0.20</td>\n","      <td>0.058962</td>\n","      <td>10.735583</td>\n","      <td>0.166991</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>10210</th>\n","      <td>2021-06-11</td>\n","      <td>126.529999</td>\n","      <td>127.440002</td>\n","      <td>126.099998</td>\n","      <td>127.349998</td>\n","      <td>127.349998</td>\n","      <td>53522400</td>\n","      <td>126.201356</td>\n","      <td>126.598074</td>\n","      <td>-0.396718</td>\n","      <td>-0.720339</td>\n","      <td>161434764400</td>\n","      <td>0.03</td>\n","      <td>1.239997</td>\n","      <td>130.479996</td>\n","      <td>126.268509</td>\n","    </tr>\n","    <tr>\n","      <th>10211</th>\n","      <td>2021-06-14</td>\n","      <td>127.820000</td>\n","      <td>130.539993</td>\n","      <td>127.070000</td>\n","      <td>130.479996</td>\n","      <td>130.479996</td>\n","      <td>96906500</td>\n","      <td>126.859608</td>\n","      <td>126.885624</td>\n","      <td>-0.026015</td>\n","      <td>-0.581474</td>\n","      <td>161531670900</td>\n","      <td>0.06</td>\n","      <td>3.129998</td>\n","      <td>129.639999</td>\n","      <td>126.529991</td>\n","    </tr>\n","    <tr>\n","      <th>10212</th>\n","      <td>2021-06-15</td>\n","      <td>129.940002</td>\n","      <td>130.600006</td>\n","      <td>129.389999</td>\n","      <td>129.639999</td>\n","      <td>129.639999</td>\n","      <td>62746300</td>\n","      <td>127.287361</td>\n","      <td>127.089651</td>\n","      <td>0.197709</td>\n","      <td>-0.425637</td>\n","      <td>161468924600</td>\n","      <td>-0.04</td>\n","      <td>-0.839997</td>\n","      <td>130.149994</td>\n","      <td>127.682014</td>\n","    </tr>\n","    <tr>\n","      <th>10213</th>\n","      <td>2021-06-16</td>\n","      <td>130.369995</td>\n","      <td>130.889999</td>\n","      <td>128.460007</td>\n","      <td>130.149994</td>\n","      <td>130.149994</td>\n","      <td>91815000</td>\n","      <td>127.727766</td>\n","      <td>127.316344</td>\n","      <td>0.411422</td>\n","      <td>-0.258225</td>\n","      <td>161560739600</td>\n","      <td>0.06</td>\n","      <td>0.509995</td>\n","      <td>131.789993</td>\n","      <td>128.403000</td>\n","    </tr>\n","    <tr>\n","      <th>10214</th>\n","      <td>2021-06-17</td>\n","      <td>129.800003</td>\n","      <td>132.550003</td>\n","      <td>129.649994</td>\n","      <td>131.789993</td>\n","      <td>131.789993</td>\n","      <td>96721700</td>\n","      <td>128.352724</td>\n","      <td>127.647725</td>\n","      <td>0.704999</td>\n","      <td>-0.065581</td>\n","      <td>161657461300</td>\n","      <td>0.06</td>\n","      <td>1.639999</td>\n","      <td>130.460007</td>\n","      <td>128.171310</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>2553 rows × 16 columns</p>\n","</div>"],"text/plain":["             Date        Open  ...  Actual Next Adj Close  Predicted Next Adj Close\n","7662   2011-04-27   12.580000  ...              10.648976                  0.178104\n","7663   2011-04-28   12.363929  ...              10.752778                  0.187542\n","7664   2011-04-29   12.385000  ...              10.634542                  0.077263\n","7665   2011-05-02   12.490714  ...              10.693504                  0.691720\n","7666   2011-05-03   12.428214  ...              10.735583                  0.166991\n","...           ...         ...  ...                    ...                       ...\n","10210  2021-06-11  126.529999  ...             130.479996                126.268509\n","10211  2021-06-14  127.820000  ...             129.639999                126.529991\n","10212  2021-06-15  129.940002  ...             130.149994                127.682014\n","10213  2021-06-16  130.369995  ...             131.789993                128.403000\n","10214  2021-06-17  129.800003  ...             130.460007                128.171310\n","\n","[2553 rows x 16 columns]"]},"metadata":{"tags":[]},"execution_count":10}]},{"cell_type":"code","metadata":{"id":"UstG4kwolrrU"},"source":[""],"execution_count":null,"outputs":[]}]}