Hello everyone,
I don’t know if any of you have ever tried to process the examples in the PredRNN space-time model github directory (GitHub - thuml/predrnn-pytorch: Official implementation for NIPS'17 paper: PredRNN: Recurrent Neural Networks for Predictive Learning Using Spatiotemporal LSTMs.). In fact I’m new in the field and I tried to train the examples but I can’t get the loss curves during training and the different metrics. I don’t know if these are calculated separately after training the model and secondly the model is saved in a “mnist_model.ckpt” file and I would like to know how to read this type of file? Aren’t these directly the parameters of the model being saved?
Thanks for your feedback,
Best regards,
Glenn