GitHub / pytorch/pytorch issues and pull requests
Labelled with: module: nestedtensor
#123698 - [RFC] TensorImpl::has_symbolic_sizes_strides can return false when sizes() actually contains symbolic values
Issue -
State: closed - Opened by davidberard98 over 1 year ago
- 7 comments
Labels: triaged, module: nestedtensor, oncall: pt2, module: dynamic shapes
#120510 - DISABLED test_sdpa_compile_cuda_float16 (__main__.TestNestedTensorSubclassCUDA)
Issue -
State: closed - Opened by huydhn over 1 year ago
- 2 comments
Labels: triaged, module: nestedtensor, skipped, oncall: transformer/mha
#118446 - [Nested Tensor] Support NT construction inside PT2 graph
Issue -
State: closed - Opened by davidberard98 almost 2 years ago
- 3 comments
Labels: feature, triaged, module: nestedtensor, oncall: pt2
#115978 - Implement KV cached attention with NestedTensor
Issue -
State: open - Opened by aploium almost 2 years ago
Labels: triaged, module: nestedtensor
#115749 - Conversions between strided and jagged layouts for Nested Tensors
Pull Request -
State: open - Opened by ani300 almost 2 years ago
- 47 comments
Labels: open source, module: nestedtensor, Merged, Reverted, Stale, ciflow/trunk, release notes: nested tensor, keep-going
#115749 - Conversions between strided and jagged layouts for Nested Tensors
Pull Request -
State: open - Opened by ani300 almost 2 years ago
- 45 comments
Labels: open source, module: nestedtensor, Merged, Reverted, ciflow/trunk, release notes: nested tensor, keep-going
#112909 - DISABLED test_transpose_inference_mode_interaction_cpu_float16 (__main__.TestNestedTensorDeviceTypeCPU)
Issue -
State: open - Opened by pytorch-bot[bot] about 2 years ago
- 4 comments
Labels: triaged, module: flaky-tests, module: nestedtensor, skipped
#112908 - DISABLED test_transpose_inference_mode_interaction_cpu_float64 (__main__.TestNestedTensorDeviceTypeCPU)
Issue -
State: open - Opened by pytorch-bot[bot] about 2 years ago
- 2 comments
Labels: triaged, module: flaky-tests, module: nestedtensor, skipped
#112846 - DISABLED test_transpose_inference_mode_interaction_cpu_float32 (__main__.TestNestedTensorDeviceTypeCPU)
Issue -
State: open - Opened by pytorch-bot[bot] about 2 years ago
- 6 comments
Labels: triaged, module: flaky-tests, module: nestedtensor, skipped
#112845 - DISABLED test_transpose_cpu_float32 (__main__.TestNestedTensorDeviceTypeCPU)
Issue -
State: open - Opened by pytorch-bot[bot] about 2 years ago
- 7 comments
Labels: triaged, module: flaky-tests, module: nestedtensor, skipped
#112509 - Add a function to torch.nested to create nested tensors from a buffer and sizes
Issue -
State: open - Opened by fleonce about 2 years ago
- 2 comments
Labels: module: serialization, triaged, module: nestedtensor
#112464 - DISABLED test_layer_norm_backward_size_1024_cpu (__main__.TestNestedTensorAutogradCPU)
Issue -
State: open - Opened by pytorch-bot[bot] about 2 years ago
- 2 comments
Labels: triaged, module: flaky-tests, module: nestedtensor, skipped, oncall: pt2
#112356 - DISABLED test_layer_norm_backward_5d_size_4_cpu (__main__.TestNestedTensorAutogradCPU)
Issue -
State: closed - Opened by pytorch-bot[bot] about 2 years ago
- 2 comments
Labels: triaged, module: flaky-tests, module: nestedtensor, skipped, oncall: pt2
#112237 - DISABLED test_layer_norm_backward_5d_size_128_cpu (__main__.TestNestedTensorAutogradCPU)
Issue -
State: closed - Opened by pytorch-bot[bot] about 2 years ago
- 3 comments
Labels: triaged, module: flaky-tests, module: nestedtensor, skipped, oncall: pt2
#112105 - DISABLED test_backward_for_sub_op_cpu (__main__.TestNestedTensorAutogradCPU)
Issue -
State: closed - Opened by pytorch-bot[bot] about 2 years ago
- 3 comments
Labels: triaged, module: flaky-tests, module: nestedtensor, skipped, oncall: pt2
#112049 - DISABLED test_backward_for_add_op_cpu (__main__.TestNestedTensorAutogradCPU)
Issue -
State: open - Opened by pytorch-bot[bot] about 2 years ago
- 2 comments
Labels: triaged, module: flaky-tests, module: nestedtensor, skipped, oncall: pt2
#110669 - Backward pass for Nested Tensors using flash attention in sdpa fails
Issue -
State: closed - Opened by mrpep about 2 years ago
- 2 comments
Labels: triaged, module: nestedtensor, oncall: transformer/mha, module: multi-headed-attention
#110211 - DISABLED test_activations_abs_cpu (__main__.TestNestedTensorDeviceTypeCPU)
Issue -
State: open - Opened by pytorch-bot[bot] about 2 years ago
- 3 comments
Labels: triaged, module: flaky-tests, module: nestedtensor, skipped, oncall: pt2
#109304 - Created nested utils.cpp
Pull Request -
State: open - Opened by drisspg about 2 years ago
- 9 comments
Labels: module: nestedtensor, Merged, Reverted, ciflow/trunk, topic: not user facing
#108567 - Allow slicing of Nested Tensors along constant dimensions
Issue -
State: open - Opened by EntropicEffect about 2 years ago
- 3 comments
Labels: triaged, module: nestedtensor
#108213 - reduce overhead in split and chunk for NestedTensor
Pull Request -
State: open - Opened by cpuhrsch about 2 years ago
- 3 comments
Labels: triaged, module: nestedtensor, topic: performance
#105325 - Padded tensor subclass
Issue -
State: open - Opened by ezyang over 2 years ago
- 12 comments
Labels: feature, triaged, module: nestedtensor, tensor subclass
#102167 - Calling pin_memory() fails for nested tensor
Issue -
State: closed - Opened by EgorLakomkin over 2 years ago
- 2 comments
Labels: triaged, module: nestedtensor, topic: new features
#97485 - Adding Backward Support for NestedTensors and FlashAttention
Pull Request -
State: open - Opened by drisspg over 2 years ago
- 13 comments
Labels: module: nestedtensor, Merged, Reverted, ciflow/trunk, ciflow/periodic, release notes: nested tensor, module: multi-headed-attention
#93843 - EmbeddingBag to support mini-batches with offsets
Issue -
State: closed - Opened by alexshtf almost 3 years ago
- 2 comments
Labels: triaged, enhancement, module: nestedtensor
#83994 - General NestedTensor op coverage tracking issue
Issue -
State: open - Opened by drisspg about 3 years ago
- 2 comments
Labels: feature, triaged, module: nestedtensor
#79477 - Implement NestedTensor size function
Issue -
State: closed - Opened by erichan1 over 3 years ago
- 2 comments
Labels: module: nestedtensor
#76191 - [Feature request] Exclusive prefix sum, `torch.cumsum(input, dim=0, exclusive=True)`
Issue -
State: open - Opened by yueyericardo over 3 years ago
- 8 comments
Labels: high priority, module: cuda, triaged, enhancement, module: nestedtensor, module: reductions