def main(sample_num,
embed_dim,
learning_rate,
epochs,
batch_size,
mode='max',
attention_hidden_unit=None):
"""
:param sample_num: the num of training sample
:param embed_dim: the dimension of all embedding layer
:param learning_rate:
:param epochs:
:param batch_size:
:param mode
:param attention_hidden_unit:
:return:
"""
feature_columns, train_X, test_X, train_y, test_y = create_dataset(
sample_num, embed_dim)
# ============================Build Model==========================
model = AFM(feature_columns,
mode,
attention_hidden_unit=attention_hidden_unit)
model.summary()
# ============================model checkpoint======================
# check_path = 'save/afm_weights.epoch_{epoch:04d}.val_loss_{val_loss:.4f}.ckpt'
# checkpoint = tf.keras.callbacks.ModelCheckpoint(check_path, save_weights_only=True,
# verbose=1, period=5)
# =========================Compile============================
model.compile(loss=binary_crossentropy,
optimizer=Adam(learning_rate=learning_rate),
metrics=[AUC()])
# ===========================Fit==============================
model.fit(
train_X,
train_y,
epochs=epochs,
# callbacks=[checkpoint],
batch_size=batch_size,
validation_split=0.1)
# ===========================Test==============================
print('test AUC: %f' % model.evaluate(test_X, test_y)[1])
def run_AFM():
afm = AFM.AFM(field_size, feature_sizes, batch_size=32 * 8, is_shallow_dropout=False, verbose=True, use_cuda=True,
weight_decay=0.00002, use_fm=True, use_ffm=False, n_epochs= num_epoch)
if online = False:
afm.fit(Xi_train, Xv_train, y_train, Xi_test, Xv_test, y_test, ealry_stopping=True,refit=True)
for feat in tqdm(sparse_features):
lbe = LabelEncoder()
data[feat] = lbe.fit_transform(data[feat])
# 将连续值归一化
for feat in tqdm(dense_features):
mean = data[feat].mean()
std = data[feat].std()
data[feat] = (data[feat] - mean) / (std + 1e-12)
# print(data.shape)
# print(data.head())
train, valid = train_test_split(data, test_size=0.1, random_state=42)
# print(train.shape) # (540000, 40)
# print(valid.shape) # (60000, 40)
train_dataset = TensorDataset(torch.LongTensor(train[sparse_features].values),
torch.FloatTensor(train[dense_features].values),
torch.FloatTensor(train['label'].values))
train_loader = DataLoader(dataset=train_dataset, batch_size=args.train_batch_size, shuffle=True)
valid_dataset = TensorDataset(torch.LongTensor(valid[sparse_features].values),
torch.FloatTensor(valid[dense_features].values),
torch.FloatTensor(valid['label'].values))
valid_loader = DataLoader(dataset=valid_dataset, batch_size=args.eval_batch_size, shuffle=False)
cat_fea_unique = [data[f].nunique() for f in sparse_features]
model = AFM(cat_fea_unique, num_fea_size=len(dense_features))
train_model(model)
learning_rate = 0.001
batch_size = 4096
epochs = 10
# ========================== Create dataset =======================
feature_columns, train, test = create_criteo_dataset(file=file,
embed_dim=embed_dim,
read_part=read_part,
sample_num=sample_num,
test_size=test_size)
train_X, train_y = train
test_X, test_y = test
# ============================Build Model==========================
mirrored_strategy = tf.distribute.MirroredStrategy()
with mirrored_strategy.scope():
model = AFM(feature_columns, mode, att_vector, activation, dropout,
embed_reg)
model.summary()
# =========================Compile============================
model.compile(loss=binary_crossentropy,
optimizer=Adam(learning_rate=learning_rate),
metrics=[AUC()])
# ============================model checkpoint======================
# check_path = 'save/afm_weights.epoch_{epoch:04d}.val_loss_{val_loss:.4f}.ckpt'
# checkpoint = tf.keras.callbacks.ModelCheckpoint(check_path, save_weights_only=True,
# verbose=1, period=5)
# ===========================Fit==============================
model.fit(
train_X,
train_y,
epochs=epochs,
callbacks=[
'''
from model import AFM
from utils import create_criteo_dataset
import tensorflow as tf
from tensorflow.keras import optimizers, losses, metrics
from sklearn.metrics import accuracy_score
if __name__ == '__main__':
file = 'E:\\PycharmProjects\\推荐算法\\data\\criteo_sample.txt'
test_size = 0.2
feature_columns, (X_train, y_train), (X_test, y_test) = \
create_criteo_dataset(file, test_size=test_size)
model = AFM(feature_columns, 'att')
optimizer = optimizers.SGD(0.01)
# dataset = tf.data.Dataset.from_tensor_slices((X_train, y_train))
# dataset = dataset.batch(32).prefetch(tf.data.experimental.AUTOTUNE)
#
# model.compile(optimizer=optimizer, loss='binary_crossentropy', metrics=['accuracy'])
# model.fit(dataset, epochs=100)
# pre = model.predict(X_test)
summary = tf.summary.create_file_writer('E:\\PycharmProjects\\tensorboard')
for i in range(100):
with tf.GradientTape() as tape:
pre = model(X_train)
loss = tf.reduce_mean(losses.binary_crossentropy(y_train, pre))
print(loss.numpy())
metric_train_auc = []
metric_test_auc = []
metric_train_log_loss = []
metric_test_log_loss = []
model_list = ['FM', 'DeepFM']
xi_train, xv_train, y_train = get_deep_fm_data_format(
train_data_list[i] + '.libsvm', config['field_info'])
xi_test, xv_test, y_test = get_deep_fm_data_format(
test_data_list[i] + '.libsvm', config['field_info'])
# afm
afm = AFM.AFM(config['field_size'],
config['feature_size'],
verbose=True,
use_cuda=False,
weight_decay=0.0001,
use_fm=True,
use_ffm=False)
train_auc, train_loss, valid_auc, valid_loss = \
afm.fit(xi_train, xv_train, y_train, xi_test, xv_test, y_test, early_stopping=True, refit=False)
logging.info('validating')
y_pred_afm = afm.predict_proba(xi_test, xv_test)
# dump deep_fm result
with open('afm_result', 'wb') as f:
pickle.dump(y_pred_afm, f)
learning_rate = 0.001
batch_size = 512
epochs = 5
# ========================== Create dataset =======================
feature_columns, train, test = create_criteo_dataset(file=file,
embed_dim=embed_dim,
read_part=read_part,
sample_num=sample_num,
test_size=test_size)
train_X, train_y = train
test_X, test_y = test
# ============================Build Model==========================
model = AFM(feature_columns,
mode,
attention_hidden_unit=attention_hidden_unit)
model.summary()
# ============================model checkpoint======================
# check_path = 'save/afm_weights.epoch_{epoch:04d}.val_loss_{val_loss:.4f}.ckpt'
# checkpoint = tf.keras.callbacks.ModelCheckpoint(check_path, save_weights_only=True,
# verbose=1, period=5)
# =========================Compile============================
model.compile(loss=binary_crossentropy,
optimizer=Adam(learning_rate=learning_rate),
metrics=[AUC()])
# ===========================Fit==============================
model.fit(
train_X,
train_y,
epochs=epochs,