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
I feed the MultiInputNet with two custom network models, then I use Learner to do the training.
However, the output is not that good, and I think that this could be due to the difference in the numerical scale of the input data for the two models.
So I am trying to set different loss weights for the two network models, but I can not find how to do that.
I would be extremely grateful if someone could help me.
Hi, could someone help me with this?
I feed the
MultiInputNetwith two custom network models, then I useLearnerto do the training. However, the output is not that good, and I think that this could be due to the difference in the numerical scale of the input data for the two models. So I am trying to set different loss weights for the two network models, but I can not find how to do that.I would be extremely grateful if someone could help me.