def main():
"""Trains a model locally to test get_model() and get_loss()."""
train_x, train_y, _, _ = load_data()
input_layer = tf.keras.layers.Input(shape=(train_x.shape[1], ))
params = argparse.Namespace(first_layer_size=50, num_layers=5)
predictions = get_model(input_layer, params)
model = tf.keras.models.Model(inputs=input_layer, outputs=predictions)
model.compile(optimizer="adam", loss=get_loss(), metrics=["accuracy"])
model.fit(train_x, train_y, epochs=1)
def main():
"""Trains a model locally to test get_model()."""
train_x, train_y, eval_x, eval_y = load_data()
train_y, eval_y = [np.ravel(x) for x in [train_y, eval_y]]
params = argparse.Namespace(C=1.0)
model = get_model(params)
model.fit(train_x, train_y)
score = model.score(eval_x, eval_y)
print(score)
def main():
"""Trains a model locally to test get_model()."""
train_x, train_y, eval_x, eval_y = load_data()
train_y, eval_y = [np.ravel(x) for x in [train_y, eval_y]]
params = argparse.Namespace(
n_estimators = 2,
max_depth = 3,
booster = "gbtree",
min_child_weight = 1,
learning_rate = 0.3,
gamma = 0,
subsample = 1,
colsample_bytree = 1,
reg_alpha = 0,
num_class = 1)
model = get_model(params)
model.fit(train_x, train_y)
y_pred = model.predict(eval_x)
score = metrics.roc_auc_score(eval_y, y_pred, average="macro")
print("ROC: {}".format(score))
def _upload_data_to_gcs(model):
load_data(model.data["train"], model.data["evaluation"])
