timeseriesAI / tsai

Time series Timeseries Deep Learning Machine Learning Python Pytorch fastai | State-of-the-art Deep Learning library for Time Series and Sequences in Pytorch / fastai
https://timeseriesai.github.io/tsai/
Apache License 2.0
4.95k stars 625 forks source link

get_X_pred() : get different predicting probablity when change the input to tensor, shouldn't the result be the same? #754

Closed yifeiSunny closed 1 year ago

yifeiSunny commented 1 year ago

when I input array type input, I get a different predicting probability from when I change the type of input to tensor. issue

I wonder why the results are different, shouldn't the results be the same? Thanks!

oguiza commented 1 year ago

Thanks for raising this issue @yifeiSunny. Yes, they should be the same. It's a bug. I'll fix it as soon as I can. The correct predictions are the ones generated from the numpy array.

oguiza commented 1 year ago

Hi @yifeiSunny , this issue is fixed now. You should get the correct prediction even if you pass a tensor.