Jacob Schreiber
Jacob Schreiber
There appear to be issues related to building the scipy wheel for py3.10 and also for deployment.
## 🚀 Feature Currently, it seems like all baseline examples are used for each input example. Would it be possible, potentially by adding support for passing in a tensor with...
The code currently looks like this: ``` def get_one_layer_mlp(hidden, k): # input user = mx.symbol.Variable('user') item = mx.symbol.Variable('item') score = mx.symbol.Variable('score') # user latent features user = mx.symbol.Embedding(data = user,...
Howdy! Thanks for putting all of your data online. I can see that the regions file contains all of the regions in the data file, but does the first region...
## 🐛 Bug When `torch.exp` is present in the model in any form, including softmax, logsoftmax, and logsumexp operations, the deltas seem to get pretty big for my model. I've...
## 🐛 Bug I am observing a large divergence when using DeepLiftShap on a model with ReLU activations (or any type of activation) but not when using `torch.nn.Identity` instead. This...
Hello all. I have a model with the following style: ```python class Test1(torch.nn.Module): def __init__(self....): self.conv1 = torch.nn.Conv1d(...) self.conv2 = torch.nn.Conv2d(...) def forward(self, X): X = torch.nn.ReLU()(self.conv1(X)) X = torch.nn.ReLU()(self.conv2(X))...
Hello! Maybe this isn't the right place, since you're just hosting the data, but I tried downloading the Buenrostro 2018 raw data from dropbox and each of the bed files...
## 🐛 Bug Between captum 0.5.0 and 0.6.0 the arguments `grad_input` and `grad_output` in the DeepLIFT implementation switched from being a tuple of length 1 to just being a PyTorch...
I had added state.identity as a way for people to use repeated names for their states, but I think this overcomplicates the issue. We should remove this, and force people...