pyvarinf
pyvarinf copied to clipboard
ctallec
Python package facilitating the use of Bayesian Deep Learning methods with Variational Inference for PyTorch
"loss_prior = var_model.prior_loss() / 60000" Does "60000" mean batchsize? Thank you very much!
Hi @ctallec, Great package! Thanks for writing it. I recently found a bug that I wanted to seek your advice on. When I Variationalize an LSTM model, I see an...
I used python2.7 to test the example: import pyvarinf import torch import torch.nn as nn import torch.nn.functional as F import torch.optim as optim class Net(nn.Module): def __init__(self): super(Net, self).__init__() self.conv1...
