how to reshape array to predict with LSTM
Question:
I made an LSTM model based on this tutorial where the model input batch shape is:
print(config["layers"][0]["config"]["batch_input_shape"])
returns:
(None, 1, 96)
Can someone give me a tip on how change my testing data to this array shape to match the model input batch size?
testday = read_csv('./data.csv', index_col=[0], parse_dates=True)
testday_scaled = scaler.fit_transform(testday.values)
print(testday_scaled.shape)
returns
(96, 1)
Answers:
IIUC, You need to use numpy.swapaxes and then add None to the first dimension.
import numpy as np
testday_scaled = np.swapaxes(testday_scaled, 0, 1)
testday_scaled = testday_scaled[None, ...]
I made an LSTM model based on this tutorial where the model input batch shape is:
print(config["layers"][0]["config"]["batch_input_shape"])
returns:
(None, 1, 96)
Can someone give me a tip on how change my testing data to this array shape to match the model input batch size?
testday = read_csv('./data.csv', index_col=[0], parse_dates=True)
testday_scaled = scaler.fit_transform(testday.values)
print(testday_scaled.shape)
returns
(96, 1)
IIUC, You need to use numpy.swapaxes and then add None to the first dimension.
import numpy as np
testday_scaled = np.swapaxes(testday_scaled, 0, 1)
testday_scaled = testday_scaled[None, ...]