6.3 jena_climate target leakage problem

[Dolores2333]

The author foget deleting the temperature column in the float_data which is used as a pool for sampling. Thus the outcome looks perfect due to the target leakage.

Delete the target response when try the code yourself, though it will let u see how bad the Network performs. 😄

temp = float_data[:, 1] a = float_data[:, 0] a = np.reshape(a, newshape=(len(a), 1)) b = float_data[:, 2:] print(a.shape, b.shape) float_data = np.concatenate([a, b], axis=1)

[zaquest]

The objective is to predict future temperature values, while doing so it is perfectly fine to use past values of the target variable. In fact there are methods that rely solely on the target variable history to predict future values (e.g. the naive method author used as a baseline).

[Dolores2333]

I see. Thanks for your rely!