parallel_model.fit(Xtrain,
Ytrain,
validation_data=(Xvalid, Yvalid),
epochs=1000,
batch_size=64,
verbose=2,
callbacks=[early_stopping])
# generate parallel data for test
Xtrain, Ytrain, Xtest, Ytest = parallel_data_generator_for_test(
X_train, Y_train, X_test, Y_test, len(models))
# model evaluation
print("Model Name %s" % name)
print("Training...")
run_test(models, parallel_model, Xtrain, Ytrain, y_train)
print("Testing...")
run_test(models, parallel_model, Xtest, Ytest, y_test)
# parallel model saving
print("Model Data and CSV Saving...")
save_model(models, parallel_model, model_dir, dataset)
# intermediate feature extraction
print('Intermediate Feature Extracting...')
intermediate_output_tarin = feature_predict(models, model_schema, Xtrain)
intermediate_output_test = feature_predict(models, model_schema, Xtest)
# save feature as csv
csv_save(dataset, y_train, intermediate_output_tarin, y_test,
intermediate_output_test, models, feature_dir)
Xtrain, Ytrain = X_train[train_idx], Y_train[train_idx]
Xvalid, Yvalid = X_train[valid_idx], Y_train[valid_idx]
# fitting parallel model
print('Parallel DNN Training...')
model.fit(Xtrain,
Ytrain,
validation_data=(Xvalid, Yvalid),
epochs=1000,
batch_size=128,
verbose=2,
callbacks=[early_stopping])
# model evaluation
print("Model Name %s" % name)
print("Training...")
run_test(name, model, X_train, Y_train, y_train)
print("Testing...")
run_test(name, model, X_test, Y_test, y_test)
# parallel model saving
print("Model Data and CSV Saving...")
save_model(name, model, model_dir, dataset)
# intermediate feature extraction
print('Intermediate Feature Extracting...')
intermediate_output_tarin = feature_predict(
name, model, model_schema, X_train)
intermediate_output_test = feature_predict(
name, model, model_schema, X_test)
# save feature as csv