NVIDIA apex issues - Githubissues

NVIDIA / apex

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

BSD 3-Clause "New" or "Revised" License

8.42k stars 1.4k forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Win11+Visual Studio 2022,install successfully.

#1809 aswordok opened 5 months ago
2
Fixed compute type for FP16 Tensor core wrapper around cublas GEMMEx

#1808 suachong closed 5 months ago
0
Error

#1807 silentghost1412 opened 5 months ago
0
Use torch.testing.all_close instead of get_max_diff in test_lamb.py

#1806 Fuzzkatt closed 5 months ago
1
"packaging" library exists but not found

#1805 mahmoodn closed 5 months ago
3
Cannot import name 'UnencryptedCookieSessionFactoryConfig'

#1804 mahmoodn closed 5 months ago
3
ImportError: cannot import name '_library_root_logger' from 'apex' (unknown location)

#1803 BBALU1660 opened 5 months ago
2
Contrib unit test failure in `openfold_triton/test_fused_adam_swa.py::FusedAdamSWATestCase::test_fused_update_on_random_data`

#1802 xwang233 opened 6 months ago
0
Avoid importing apex transformer automatically

#1801 nWEIdia opened 6 months ago
0
Not Able to install apex.

#1800 Avinash-py opened 6 months ago
0
[ INSTALLATION ] - Not able to install apex on a Linux machine

#1799 MBadriNarayanan opened 6 months ago
3
Fix reduce_blocks_into_lanes race condition

#1798 Fuzzkatt closed 7 months ago
0
NCCL userbuffer for DP RS in DistOpt

#1797 WanZzzzzz closed 7 months ago
0
Add nccl_allocator for zero-copy user buffer

#1796 Aidyn-A closed 7 months ago
0
Avoid unnecessary param write in distributed Adam kernel

#1795 timmoon10 closed 7 months ago
1
Enhance Distributed Fused Adam

#1794 alpha0422 closed 6 months ago
4
apex not installing

#1793 pradeepdev-1995 opened 7 months ago
0
Up to date patch for Windows compilation with Visual Studio 2022, CUDA 12.1 and PyTorch 2.2.2

#1792 doctorpangloss opened 7 months ago
2
fix building torch extension with glog

#1791 petronny opened 7 months ago
1
Add xentropy bf16 support

#1790 zyeric opened 7 months ago
0
Unclear licensing for contrib/sparsity

#1789 hyandell opened 8 months ago
0
install failure

#1788 52Hzaaa opened 8 months ago
0
No module named 'torch._six'

#1787 xujin1184104394 opened 8 months ago
1
64-bit indexing Adam

#1786 cdm114514 opened 8 months ago
0
cannot import name 'AutoencoderKLTemporalDecoder' from 'diffusers.models'

#1785 zj19941113 closed 8 months ago
0
Add 2D Fused RoPE

#1784 yaox12 closed 7 months ago
0
Move to the correct device for v1 state dict

#1783 acphile closed 7 months ago
2
Bump thresholds for `test_backward` in `test_fused_softmax.py`

#1782 eqy closed 8 months ago
0
On installing apex (+without sudo/docker)

#1781 stet-stet opened 8 months ago
1
[Questing] For apex sparsity model, when i export trt engine with flag sparsity=enable or force, only partial layer picked sparse implementation.

#1780 Bobo-y closed 8 months ago
4
Installation Problem: RuntimeError: Error compiling objects for extension

#1779 wwma opened 8 months ago
3
Cannot compile/build cuda_ext on H100

#1778 GuanhuaWang opened 8 months ago
0
[CUDNN][cudnn-frontend] Bump cuDNN to 1.0.3

#1777 eqy opened 9 months ago
0
memory format option is only supported by strided tensors

#1776 Cheny1m closed 9 months ago
1
Skip the p2p test on single GPU platforms

#1775 nWEIdia closed 9 months ago
0
Add GPUDirect Storage

#1774 Aidyn-A closed 9 months ago
1
Running apex with error: AttributeError: module 'torch.distributed' has no attribute '_reduce_scatter_base'

#1773 caseclose opened 9 months ago
3
[64bit indexing][Adam] Add annotation to large tensor test

#1772 eqy closed 10 months ago
0
Support scaled optimizer state in distributed Adam optimizer

#1771 timmoon10 closed 9 months ago
0
The build system requires torch. Currently, you cannot build the package using ordinary pip invocations.

#1770 doctorpangloss opened 10 months ago
1
Update test_bottleneck_module.py - Skip BottleNeck Peer Memory Test

#1769 nWEIdia closed 10 months ago
0
syncbn with "channel_last=True" produce wrong result when feature_num is not pow-of-two

#1768 Zehaos opened 10 months ago
1
Update test_transducer_joint.py

#1767 nWEIdia closed 10 months ago
1
Increase tolerance to workaround unit test failures on A100

#1766 nWEIdia closed 10 months ago
0
64-bit indexing Adam

#1765 eqy closed 10 months ago
0
apex installation failures

#1764 momo1986 opened 11 months ago
2
Installation instructions don't build/install the C modules

#1763 zxti opened 11 months ago
3
Apex installation fails

#1762 yang606 opened 11 months ago
1
Cannot install apex on the machine of CUDA 12.2

#1761 momo1986 opened 11 months ago
10
Make fused normalization functions backward-compatible

#1760 timmoon10 closed 10 months ago
2

Previous Next