azure / deeplearningfortimeseriesforecasting Goto Github PK

Hi can this model be amended so that you can have a longer forecast horizon than the seasonality? I have tried extensively but failed. And if not, do you know of any python model online where that indeed can be done?

Hello

The ES_RNN model ran first epoch but at second epoch, it generated an error below. Can you help me to fix this issue ?

Thanks,

Train on 75912 samples, validate on 1488 samples
Epoch 1/10
75648/75912 [============================>.] - ETA: 0s - loss: 0.0078

InvalidArgumentError: Expected size[0] in [0, 23], but got 47
[[{{node es_1/Slice_25}} = Slice[Index=DT_INT32, T=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_input_1_0_0, es_1/Slice_25/begin, es_1/Slice_25/size)]]