def main(input_filepath, output_filepath):
# make_data_set(input_filepath, output_filepath)
# visualize(output_filepath)
# create_trained_model(output_filepath)
print(
predict(
"my cat doest like the milk that I bought from the store yesterday"
))
def main():
train, test, words, users = make_dataset.main()
X, y, final_test = build_features.main(train, test, words, users)
pipeline = train_model.fit(X, y)
submissions = predict_model.predict(pipeline, final_test)
save_submission.save(submissions)
def full_workflow(historical=True):
# Create dataset
df_canonical = build_features.make()
build_features.persist(df_canonical)
# Normalize dataset
df_normalized = normalize_features.make()
normalize_features.persist(df_normalized)
if historical is True:
(train_dataset, train_labels), (
test_dataset, test_labels
) = split_dataset.split_training_data_randomly_with_seed(df_normalized)
# Train model
model = train_model.make("dev")
train_model.persist(model)
# Create predictions
predictions = predict_model.predict()
# Evaluate predictions
predict_model.evaluate_predictions(predictions, test_labels)
else:
train_dataset, train_labels = split_dataset.pop_label(df_normalized)
test_dataset = build_features.make(build_test_game_data.make())
normalized_test_dataset = normalize_features.make(test_dataset)
normalized_test_dataset = add_dummies(normalized_test_dataset,
train_dataset)
# Train model
model = train_model.make("test", train_dataset, train_labels)
train_model.persist(model)
# Create predictions
predictions = predict_model.predict(normalized_test_dataset)
df_predictions = build_test_output(test_dataset, predictions)
fr = ProcessedFilePersistence('2019Predictions.csv')
fr.write_to_csv(df_predictions)
build_bracket_output()
build_data()
create_freq_chart()
create_wordclouds()
transform_data()
train_model(algorithm="svm", method="bag_of_words")
test_model(algorithm="svm", method="bag_of_words")
train_model(algorithm="knn", method="bag_of_words")
test_model(algorithm="knn", method="bag_of_words")
train_model(algorithm="naive_bayes", method="bag_of_words")
test_model(algorithm="naive_bayes", method="bag_of_words")
train_model(algorithm="linear_regression", method="bag_of_words")
test_model(algorithm="linear_regression", method="bag_of_words")
train_model(algorithm="random_forest", method="bag_of_words")
test_model(algorithm="random_forest", method="bag_of_words")
train_model(algorithm="svm", method="tf_idf")
test_model(algorithm="svm", method="tf_idf")
train_model(algorithm="knn", method="tf_idf")
test_model(algorithm="knn", method="tf_idf")
train_model(algorithm="naive_bayes", method="tf_idf")
test_model(algorithm="naive_bayes", method="tf_idf")
train_model(algorithm="linear_regression", method="tf_idf")
test_model(algorithm="linear_regression", method="tf_idf")
train_model(algorithm="random_forest", method="tf_idf")
test_model(algorithm="random_forest", method="tf_idf")
create_results_chart()
txt = "Fahrettin Koca yeni tedbirlere yönelik açıklamalarda bulundu."
result = predict(txt, algorithm="svm", method="bag_of_words")
print(f"Tahmin sonucu: {result}")
def predict(customer_id: int, month: int):
predict_model.predict(customer_id, month)