def fit(self, X_train, y_train, X_test, y_test, batch_size=50, nb_epoch=3):
"""
:param X_train: each instance is a list of word index
:param y_train:
:return:
"""
print(len(X_train), 'train sequences')
print(len(X_test), 'test sequences')
print("Pad sequences (samples x time)")
X_train = sequence.pad_sequences(X_train, maxlen=self.maxlen)
X_test = sequence.pad_sequences(X_test, maxlen=self.maxlen)
print('X_train shape:', X_train.shape)
print('X_test shape:', X_test.shape)
y_train = expand_label(y_train)
y_test = expand_label(y_test)
#early stopping
early_stop = EarlyStopping(monitor='val_loss', patience=2)
self.model.fit({
'input': X_train,
'output': y_train
},
batch_size=batch_size,
nb_epoch=nb_epoch,
verbose=1,
validation_data=({
'input': X_test,
'output': y_test
}),
callbacks=[early_stop])
def fit(self, X_train, y_train, X_test, y_test,
batch_size=100, nb_epoch=3, show_accuracy=True):
"""
:param X_train: each instance is a list of word index
:param y_train:
:return:
"""
print(len(X_train), 'train sequences')
print(len(X_test), 'test sequences')
print("Pad sequences (samples x time)")
X_train = sequence.pad_sequences(X_train, maxlen=self.maxlen)
X_test = sequence.pad_sequences(X_test, maxlen=self.maxlen)
print('X_train shape:', X_train.shape)
print('X_test shape:', X_test.shape)
y_train = expand_label(y_train)
y_test = expand_label(y_test)
self.model.fit(X_train, y_train, batch_size=batch_size, nb_epoch=nb_epoch,
show_accuracy=True, validation_data=(X_test, y_test))
def fit(self, X_train, y_train, X_test, y_test,
batch_size=50, nb_epoch=3):
"""
:param X_train: each instance is a list of word index
:param y_train:
:return:
"""
print(len(X_train), 'train sequences')
print(len(X_test), 'test sequences')
print("Pad sequences (samples x time)")
X_train = sequence.pad_sequences(X_train, maxlen=self.maxlen)
X_test = sequence.pad_sequences(X_test, maxlen=self.maxlen)
print('X_train shape:', X_train.shape)
print('X_test shape:', X_test.shape)
y_train = expand_label(y_train)
y_test = expand_label(y_test)
#early stopping
early_stop = EarlyStopping(monitor='val_loss', patience=2)
self.model.fit({'input': X_train, 'output': y_train}, batch_size=batch_size, nb_epoch=nb_epoch,
verbose=1, validation_data=({'input': X_test, 'output': y_test}), callbacks=[early_stop])
def get_single_feature(raw_df, KPI_ID_name):
df = pd.DataFrame(raw_df.copy())
df = df.reset_index(drop=True)
# test_manual_feature = get_manual_feature(KPI_LIST_test[index])
# test_manual_feature['diff_shift']
# median_time = test_manual_feature['diff_timestamp_feature'].median()
print("KPI_LIST_test[index] length:", len(KPI_LIST_test[index]))
# print("test_manual_feature length:", len(test_manual_feature))
# test_processd_feature = get_process_feature(is_test=True, KPI_ID_name=KPI_ID_name, window=window)
# df['pro_diff'] = test_processd_feature
# df['pro_diff'] = df['pro_diff'] * 100
del df['timestamp']
del df['KPI ID']
# print(df)
if KPI_ID_name == '1c35dbf57f55f5e4':
y_df = df['value'].map(lg_1600)
elif KPI_ID_name == '7c189dd36f048a6c':
y_df = df['value'].map(lw_300)
elif KPI_ID_name == '8a20c229e9860d0c':
y_df = df['value'].map(a20)
elif KPI_ID_name == '8bef9af9a922e0b3':
y_df = df['value'].map(bef)
elif KPI_ID_name == '8c892e5525f3e491':
y_df = df['value'].map(c89)
elif KPI_ID_name == '9bd90500bfd11edb':
y_df = df['value'].map(bd9)
elif KPI_ID_name == '9ee5879409dccef9':
y_df = df['value'].map(ee5)
elif KPI_ID_name == '02e99bd4f6cfb33f':
y_df = df['value'].map(e99)
elif KPI_ID_name == '18fbb1d5a5dc099d':
y_df = df['value'].map(fbb)
elif KPI_ID_name == '40e25005ff8992bd':
y_df = df['value'].map(e25)
elif KPI_ID_name == '54e8a140f6237526':
y_df = df['value'].map(e8a)
elif KPI_ID_name == '76f4550c43334374':
y_df = df['value'].map(f45)
elif KPI_ID_name == '88cf3a776ba00e7c':
y_df = df['value'].map(cf3)
elif KPI_ID_name == '046ec29ddf80d62e':
y_df = df['value'].map(ec2)
elif KPI_ID_name == '07927a9a18fa19ae':
y_df = df['value'].map(a9a)
elif KPI_ID_name == '09513ae3e75778a3':
y_df = df['value'].map(ae3)
elif KPI_ID_name == '71595dd7171f4540':
y_df = df['value'].map(dd7)
elif KPI_ID_name == '769894baefea4e9e':
y_df = df['value'].map(bae)
elif KPI_ID_name == 'a5bf5d65261d859a':
y_df = df['value'].map(a5b)
elif KPI_ID_name == 'a40b1df87e3f1c87':
y_df = df['value'].map(a40)
elif KPI_ID_name == 'affb01ca2b4f0b45':
y_df = df['value'].map(aff)
elif KPI_ID_name == 'b3b2e6d1a791d63a':
y_df = df['value'].map(b3b)
elif KPI_ID_name == 'c58bfcbacb2822d1':
test_manual_feature = get_manual_feature(KPI_LIST_test[index])
y_df = test_manual_feature['is_diff_over_3sigma']
elif KPI_ID_name == 'cff6d3c01e6a6bfa':
y_df = df['value'].map(cff)
elif KPI_ID_name == 'da403e4e3f87c9e0':
y_df = df['value'].map(da4)
elif KPI_ID_name == 'e0770391decc44ce':
y_df = df['value'].map(e07)
# print("y_df before shift:")
# print(y_df)
df['y'] = y_df
if is_shift:
df['y'] = df['y'].shift(shift)
df['y'].bfill(inplace=True)
df['y'] = df['y'].map(to_int)
df['y'] = expand_label(df['y'].values, wide=wide)
return df['y']
