update inference testing

Author: JordiCorbilla

download_market_data.py

@@ -14,7 +14,6 @@
 # ==============================================================================
 
 import pandas as pd
-import json
 import yfinance as yf
 import datetime
 

download_market_data_info.py

@@ -13,71 +13,70 @@
 # limitations under the License.
 # ==============================================================================
 
-import pandas as pd
 import json
 import yfinance as yf
 
-msft = yf.Ticker("GOOG")
+sec = yf.Ticker("GOOG")
 
 print('Info')
-print(json.dumps(msft.info, indent=4, sort_keys=True))
+print(json.dumps(sec.info, indent=4, sort_keys=True))
 print()
 print('ISIN')
-print(msft.isin)
+print(sec.isin)
 print()
 print('Major Holders')
-print(msft.major_holders)
+print(sec.major_holders)
 print()
 print('Institutional Holders')
-print(msft.institutional_holders)
+print(sec.institutional_holders)
 print()
 print('Dividents')
-print(msft.dividends)
+print(sec.dividends)
 print()
 print('Splits')
-print(msft.splits)
+print(sec.splits)
 print()
 print('Actions')
-print(msft.actions)
+print(sec.actions)
 print()
 print('Calendar')
-print(msft.calendar)
+print(sec.calendar)
 print()
 print('Recommendations')
-print(msft.recommendations)
+print(sec.recommendations)
 print()
 print('Earnings')
-print(msft.earnings)
+print(sec.earnings)
 print()
 print('Quarterly Earnings')
-print(msft.quarterly_earnings)
+print(sec.quarterly_earnings)
 print()
 print('Financials')
-print(msft.financials)
+print(sec.financials)
 print()
 print('Quarterly Financials')
-print(msft.quarterly_financials)
+print(sec.quarterly_financials)
 print()
 print('Balance Sheet')
-print(msft.balance_sheet)
+print(sec.balance_sheet)
 print()
 print('Quarterly Balance Sheet')
-print(msft.quarterly_balance_sheet)
+print(sec.quarterly_balance_sheet)
 print()
-print('Balancesheet')
-print(msft.balancesheet)
+print('BalanceSheet')
+print(sec.balancesheet)
 print()
-print('Quarterly Balancesheet')
-print(msft.quarterly_balancesheet)
+print('Quarterly BalanceSheet')
+print(sec.quarterly_balancesheet)
 print()
 print('Cashflow')
-print(msft.cashflow)
+print(sec.cashflow)
 print()
 print('Quarterly Cashflow')
-print(msft.quarterly_cashflow)
+print(sec.quarterly_cashflow)
 print()
 print('Sustainability')
-print(msft.sustainability)
+print(sec.sustainability)
 print()
 print('Options')
-print(msft.options)
+print(sec.options)

stock_prediction_deep_learning_inference.py

@@ -12,3 +12,44 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 # ==============================================================================
+import os
+from absl import app
+import tensorflow as tf
+import pandas as pd
+from sklearn.preprocessing import MinMaxScaler
+
+from stock_prediction_numpy import StockData
+from datetime import date
+os.environ["PATH"] += os.pathsep + 'C:/Program Files (x86)/Graphviz2.38/bin/'
+
+
+def main(argv):
+    print(tf.version.VERSION)
+    inference_folder = os.path.join(os.getcwd(), 'GOOG_20200704_b5f47746c83698528343678663ac3c96')
+
+    # load future data
+    data = StockData()
+    min_max = MinMaxScaler(feature_range=(0, 1))
+    x_test, y_test = data.generate_future_data(TIME_STEPS, min_max, date(2020, 7, 5), date(2021, 7, 5))
+
+    # load the weights from our best model
+    model = tf.keras.models.load_model(os.path.join(inference_folder, 'model_weights.h5'))
+    model.summary()
+
+    # display the content of the model
+    baseline_results = model.evaluate(x_test, y_test, verbose=2)
+    for name, value in zip(model.metrics_names, baseline_results):
+        print(name, ': ', value)
+    print()
+
+    # perform a prediction
+    test_predictions_baseline = model.predict(x_test)
+    test_predictions_baseline = min_max.inverse_transform(test_predictions_baseline)
+    test_predictions_baseline = pd.DataFrame(test_predictions_baseline)
+    test_predictions_baseline.to_csv(os.path.join(inference_folder, 'inference.csv'))
+    print(test_predictions_baseline)
+
+
+if __name__ == '__main__':
+    TIME_STEPS = 60
+    app.run(main)

stock_prediction_forecasting.py

@@ -15,6 +15,7 @@
 import os
 from absl import app
 import tensorflow as tf
+import pandas as pd
 from sklearn.preprocessing import MinMaxScaler
 
 from stock_prediction_numpy import StockData
@@ -35,8 +36,17 @@ def main(argv):
     model = tf.keras.models.load_model(os.path.join(inference_folder, 'model_weights.h5'))
     model.summary()
 
+    # display the content of the model
+    baseline_results = model.evaluate(x_test, y_test, verbose=2)
+    for name, value in zip(model.metrics_names, baseline_results):
+        print(name, ': ', value)
+    print()
+
     # perform a prediction
     test_predictions_baseline = model.predict(x_test)
+    test_predictions_baseline = min_max.inverse_transform(test_predictions_baseline)
+    test_predictions_baseline = pd.DataFrame(test_predictions_baseline)
+    test_predictions_baseline.to_csv(os.path.join(inference_folder, 'inference.csv'))
     print(test_predictions_baseline)
 
 

stock_prediction_numpy.py

@@ -15,6 +15,7 @@
 import numpy as np
 import pandas as pd
 from datetime import timedelta
+import random
 
 
 class StockData:
@@ -62,7 +63,7 @@ def generate_future_data(self, time_steps, min_max, start_date, end_date):
         y_future = []
         for single_date in self.__daterange(start_date, end_date):
             x_future.append(single_date)
-            y_future.append(0.0)
+            y_future.append(random.uniform(10, 100))
 
         test_data = pd.DataFrame({'Date': x_future, 'Close': y_future})
         test_data = test_data.set_index('Date')