embedding_dim=64)(input)
item_emb_mlp = LatentFactorMapper(column_id=1,
num_of_entities=item_num,
embedding_dim=64)(input)
# Step 2.2: Setup interactors to handle models
innerproduct_output = InnerProductInteraction()([user_emb_gmf, item_emb_gmf])
mlp_output = MLPInteraction()([user_emb_mlp, item_emb_mlp])
# Step 2.3: Setup optimizer to handle the target task
output = RatingPredictionOptimizer()([innerproduct_output, mlp_output])
model = RPRecommender(inputs=input, outputs=output)
# Step 3: Build the searcher, which provides search algorithm
searcher = Search(model=model,
tuner='greedy', # random, greedy
tuner_params={"max_trials": 5, 'overwrite': True}
)
# Step 4: Use the searcher to search the recommender
searcher.search(x=[train_X_categorical],
y=train_y,
x_val=[val_X_categorical],
y_val=val_y,
objective='val_mse',
batch_size=1024,
epochs=1,
callbacks=[tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=1)])
logger.info('Validation Accuracy (mse): {}'.format(searcher.evaluate(x=val_X_categorical,
y_true=val_y)))
# Step 5: Evaluate the searched model
# Step 2.2: Setup interactors to handle models
attention_output = SelfAttentionInteraction()(
[dense_feat_emb, sparse_feat_emb])
bottom_mlp_output = MLPInteraction()([dense_feat_emb])
top_mlp_output = MLPInteraction()([attention_output, bottom_mlp_output])
# Step 2.3: Setup optimizer to handle the target task
output = CTRPredictionOptimizer()(top_mlp_output)
model = CTRRecommender(inputs=[dense_input_node, sparse_input_node],
outputs=output)
# Step 3: Build the searcher, which provides search algorithm
searcher = Search(
model=model,
tuner='random',
tuner_params={
'max_trials': 2,
'overwrite': True
},
)
# Step 4: Use the searcher to search the recommender
searcher.search(x=[train_X_numerical, train_X_categorical],
y=train_y,
x_val=[val_X_numerical, val_X_categorical],
y_val=val_y,
objective='val_BinaryCrossentropy',
batch_size=10000,
epochs=2,
callbacks=[
tf.keras.callbacks.EarlyStopping(monitor='val_loss',
patience=1)
if args.model == 'mf':
model = build_mf(user_num, item_num)
if args.model == 'mlp':
model = build_mlp(user_num, item_num)
if args.model == 'gmf':
model = build_gmf(user_num, item_num)
if args.model == 'neumf':
model = build_neumf(user_num, item_num)
if args.model == 'autorec':
model = build_autorec(user_num, item_num)
# search and predict.
searcher = Search(
model=model,
tuner=args.search, ## hyperband, bayesian
tuner_params={
'max_trials': args.trials,
'overwrite': True
})
start_time = time.time()
searcher.search(x=train_X,
y=train_y,
x_val=val_X,
y_val=val_y,
objective='val_mse',
batch_size=args.batch_size,
epochs=args.epochs,
callbacks=[
tf.keras.callbacks.EarlyStopping(
monitor='val_loss', patience=args.early_stop)
id_num=10000,
embedding_dim=64)(input)
item_emb_mlp = LatentFactorMapper(feat_column_id=1,
id_num=10000,
embedding_dim=64)(input)
innerproduct_output = ElementwiseInteraction(elementwise_type="innerporduct")(
[user_emb_gmf, item_emb_gmf])
mlp_output = MLPInteraction()([user_emb_mlp, item_emb_mlp])
output = PointWiseOptimizer()([innerproduct_output, mlp_output])
model = CTRRecommender(inputs=input, outputs=output)
# AutoML search and predict.
searcher = Search(
model=model,
tuner='random',
tuner_params={
'max_trials': 10,
'overwrite': True
},
)
searcher.search(x=train_X,
y=train_y,
x_val=val_X,
y_val=val_y,
objective='val_BinaryCrossentropy',
batch_size=256,
epochs=20,
callbacks=[
tf.keras.callbacks.EarlyStopping(monitor='val_loss',
patience=1)
])
logger.info('Predicted Ratings: {}'.format(searcher.predict(x=val_X)))
# build the pipeline.
input = Input(shape=[2])
user_emb = LatentFactorMapper(feat_column_id=0,
id_num=user_num,
embedding_dim=64)(input)
item_emb = LatentFactorMapper(feat_column_id=1,
id_num=item_num,
embedding_dim=64)(input)
output1 = HyperInteraction()([user_emb, item_emb])
output2 = HyperInteraction()([output1, user_emb, item_emb])
output3 = HyperInteraction()([output1, output2, user_emb, item_emb])
output4 = HyperInteraction()([output1, output2, output3, user_emb, item_emb])
output = RatingPredictionOptimizer()(output4)
model = RPRecommender(inputs=input, outputs=output)
# AutoML search and predict.
searcher = Search(model=model,
tuner='random', ## hyperband, bayesian
tuner_params={'max_trials': 100, 'overwrite': True},)
searcher.search(x=train_X,
y=train_y,
x_val=val_X,
y_val=val_y,
objective='val_mse',
batch_size=1024,
epochs=10,
callbacks=[tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=1)])
logger.info('Predicting Val Dataset Accuracy (mse): {}'.format(searcher.evaluate(x=val_X, y_true=val_y)))
logger.info('Predicting Test Dataset Accuracy (mse): {}'.format(searcher.evaluate(x=test_X, y_true=test_y)))
user_emb_mlp = LatentFactorMapper(column_id=0,
num_of_entities=10000,
embedding_dim=64)(input)
item_emb_mlp = LatentFactorMapper(column_id=1,
num_of_entities=10000,
embedding_dim=64)(input)
innerproduct_output = InnerProductInteraction()([user_emb_gmf, item_emb_gmf])
mlp_output = MLPInteraction()([user_emb_mlp, item_emb_mlp])
output = CTRPredictionOptimizer()([innerproduct_output, mlp_output])
model = CTRRecommender(inputs=input, outputs=output)
# AutoML search and predict.
searcher = Search(
model=model,
tuner='random',
tuner_params={
'max_trials': 10,
'overwrite': True
},
)
searcher.search(x=[criteo.get_x_categorical(train_X)],
y=train_y,
x_val=[criteo.get_x_categorical(val_X)],
y_val=val_y,
objective='val_BinaryCrossentropy',
batch_size=256,
epochs=20,
callbacks=[
tf.keras.callbacks.EarlyStopping(monitor='val_loss',
patience=1)
])
logger.info('Predicted Ratings: {}'.format(
# build the pipeline.
input = Input(shape=[2])
user_emb = LatentFactorMapper(feat_column_id=0,
id_num=user_num,
embedding_dim=64)(input)
item_emb = LatentFactorMapper(feat_column_id=1,
id_num=item_num,
embedding_dim=64)(input)
output = ElementwiseInteraction(elementwise_type="innerporduct")(
[user_emb, item_emb])
output = RatingPredictionOptimizer()(output)
model = RPRecommender(inputs=input, outputs=output)
# AutoML search and predict
searcher = Search(
model=model,
tuner='greedy', # hyperband, greedy, bayesian
tuner_params={"max_trials": 5})
searcher.search(x=train_X,
y=train_y,
x_val=val_X,
y_val=val_y,
objective='val_mse',
batch_size=1024,
epochs=10,
callbacks=[
tf.keras.callbacks.EarlyStopping(monitor='val_loss',
patience=1)
])
logger.info('Predicting Val Dataset Accuracy (mse): {}'.format(
searcher.evaluate(x=val_X, y_true=val_y)))
if args.model == 'deepfm':
output = build_deepfm(emb_dict)
if args.model == 'crossnet':
output = build_neumf(emb_dict)
if args.model == 'autoint':
output = build_autorec(emb_dict)
if args.model == 'autorec':
output = build_autorec(emb_dict)
# Step 2.3: Setup optimizer to handle the target task
output = CTRPredictionOptimizer()(output)
model = CTRRecommender(inputs=input, outputs=output)
# Step 3: Build the searcher, which provides search algorithm
searcher = Search(model=model,
tuner=args.search,
tuner_params={'max_trials': args.trials, 'overwrite': True}
)
# Step 4: Use the searcher to search the recommender
start_time = time.time()
searcher.search(x=train_X,
y=train_y,
x_val=val_X,
y_val=val_y,
objective='val_BinaryCrossentropy',
batch_size=args.batch_size,
epochs=1,
callbacks=[tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=1)]
)
end_time = time.time()
print("running time:", end_time - start_time)