pytorch/pytorch: Tensors and Dynamic neural networks in Python with strong GPU acceleration

52361
STARS
1594
WATCHERS
14346
FORKS
10584
ISSUES

pytorch's Language Statistics

pytorch's Other Repos

Star history of pytorch
Issue history of pytorch

pytorch Recent Issues

Issue Title State Comments Created Date Updated Date Closed Date
`torch.broadcast_shapes` should not handle shape with negative dimension open 0 2021-11-27 2021-11-20 -
Add the attr `_last_lr` for the schedular `SequentialLR`. open 0 2021-11-27 2021-11-20 -
Strange ouput of torch.histc open 0 2021-11-26 2021-11-20 -
Pytorch to ONNX model conversion open 0 2021-11-26 2021-11-20 -
Couldn't export Python operator _FunctionCorrelation open 0 2021-11-26 2021-11-20 -
Is the trick for hard gumbel_softmax is correct? closed 0 2021-11-26 2021-11-20 2021-11-26
Breakpoint training in a Tensorboard log open 0 2021-11-26 2021-11-20 -
repeat_interleave hangs in a forked subprocess open 0 2021-11-26 2021-11-20 -
The dim is not checked strictly for a model in GPU closed 1 2021-11-26 2021-11-20 2021-11-26
Dead lock happened ran pytorch 1.9.0 cuda11.2 on xeon gold 6326 cpu open 0 2021-11-26 2021-11-20 -
Could not run 'aten::empty_strided' with arguments from the 'HIP' open 0 2021-11-26 2021-11-20 -
[docs] {select, slice, diagonal}_scatter not available in documentation open 0 2021-11-26 2021-11-20 -
Thank you for the diagnostic script. open 1 2021-11-26 2021-11-20 -
How to implement `bucket_by_sequence_length` with IterableDataset and DataLoader open 0 2021-11-26 2021-11-20 -
[cpp extension] making it possible to parallelize the building process open 0 2021-11-26 2021-11-20 -
[Bug] forward function error occurs when the scriptmodule (made in Python) including the hook function is loaded from C++(Libtorch1.8.2) open 0 2021-11-26 2021-11-20 -
Breakes with -OO flag in script open 0 2021-11-25 2021-11-20 -
Support for movedim export to ONNX open 0 2021-11-25 2021-11-20 -
Bugs related to NCCL on RTX 6000, code freezes with no output when using DistributedDataParallel open 0 2021-11-25 2021-11-20 -
Crash loading fairseq based model open 0 2021-11-25 2021-11-20 -
Add support for deform_conv2d export to ONNX open 0 2021-11-25 2021-11-20 -
Bug about Dropout CUDA Kernel open 0 2021-11-25 2021-11-20 -
Shared `~/.cache/torch_extensions` needs to be pytorch version aware. open 0 2021-11-25 2021-11-20 -
`torch.jit.freeze` and `torch::jit::freeze` are inconsistent open 0 2021-11-25 2021-11-20 -
Python version is 3.7.11 in latest pytorch docker image open 0 2021-11-24 2021-11-20 -
Tensor.copy_ to a slice fails silently open 1 2021-11-24 2021-11-20 -
NCCL Network is unreachable / Connection refused when initializing DDP open 1 2021-11-24 2021-11-20 -
fusion in fx graph mode did not take of direct attribute access open 0 2021-11-24 2021-11-20 -
Implement the operator `_convolution_mode` in the ONNX exporter open 1 2021-11-24 2021-11-20 -
Stop gradient option for padding open 0 2021-11-24 2021-11-20 -
AtenLtcTsTensorTest.TestAmpForeachNonFiniteCheckAndUnscale, @desertfire open 0 2021-11-24 2021-11-20 -
Two missing header files in package list. open 1 2021-11-24 2021-11-20 -
`unfold` should not raise such error open 1 2021-11-24 2021-11-20 -
maxunpool2d gives incorrect partial inverse with `output_size`. closed 2 2021-11-24 2021-11-20 2021-11-24
reduction parameter in nn.NLLLoss() depricated? closed 1 2021-11-24 2021-11-20 2021-11-24
[LibTorch-Lite] Add a custom flag to build LibTorch-Lite with LAPACK included open 0 2021-11-24 2021-11-20 -
PyTorch's cpu group convolution exist bug!!! open 2 2021-11-24 2021-11-20 -
wait() does not block the default stream for NCCL's asynchronous P2P operations open 0 2021-11-24 2021-11-20 -
pytorch build from souce code third_party <ios-cmake> error closed 0 2021-11-24 2021-11-20 2021-11-24
Incorrect accuracy pytorch(1.0.1) and libtorch because of nn.LSTM open 3 2021-11-24 2021-11-20 -
find_unused_parameters of DDP open 0 2021-11-24 2021-11-20 -
[In Gunicorn & multiprocessing environment] Cannot re-initialize CUDA in forked subprocess open 1 2021-11-24 2021-11-20 -
DISABLED test_default_store_timeout_nccl (__main__.TimeoutTest) open 1 2021-11-21 2021-11-14 -
torch.utils.data.Sampler is not recognized as a collections.abc.Sized open 3 2021-11-21 2021-11-25 -
adding the alias `eye_like` for creating a batch of eye matrices open 1 2021-11-21 2021-11-14 -
MultiHeadAttention docs are unnecessarily confusing. open 1 2021-11-21 2021-11-14 -
Tensor precision manifests differently between CPU and GPU. open 0 2021-11-21 2021-11-14 -
Typos of documents open 0 2021-11-21 2021-11-14 -
Transpose of a sparse tensor is not a view operation open 0 2021-11-21 2021-11-14 -
Questions about the inconsistent (slightly different) results obtained from multiple runs of nn.NLLLoss closed 2 2021-11-21 2021-11-26 2021-11-22