search

GheodeAI / va_am

VA-AM (Various Advanced - Analogue Methods) is a Python package based on the deep learning enhancement of the classical statistical Analogue Method. It provides several tools to analyse climatological extreme events, particularly heat waves.
https://va-am.readthedocs.io/en/latest/
GNU General Public License v3.0
2 stars 0 forks source link

Timestamp error when training analogue autoencoder model #37

Open eugenioLR opened 1 week ago

eugenioLR commented 1 week ago

At line 664 data_of_interest_init is already a timestamp, it doesn't need to be converted to a date.

eugenioLR commented 1 week ago

Also, at line 881 there's a variable with the name x_train_pre_pred_pred that shouldn't be named like that.

eugenioLR commented 1 week ago

Also also, at line 1155 the variable data doesn't exist, it should be called data_target

eugenioLR commented 1 week ago

In line 537 interest_isnot_target is incorrect, the comparision should be !=

eugenioLR commented 1 week ago

New param pred_interest_dataset, as example in simul6