Pinned issues
Issues
- 0
ImportError: cannot import name '_library_root_logger' from 'apex' (unknown location)
#1803 opened by BBALU1660 - 1
On installing apex (+without sudo/docker)
#1781 opened by stet-stet - 1
Up to date patch for Windows compilation with Visual Studio 2022, CUDA 12.1 and PyTorch 2.2.2
#1792 opened by doctorpangloss - 1
- 6
Cannot install apex on the machine of CUDA 12.2
#1761 opened by momo1986 - 3
Running apex with error: AttributeError: module 'torch.distributed' has no attribute '_reduce_scatter_base'
#1773 opened by cs-wangfeng - 0
Contrib unit test failure in `openfold_triton/test_fused_adam_swa.py::FusedAdamSWATestCase::test_fused_update_on_random_data`
#1802 opened by xwang233 - 0
Not Able to install apex.
#1800 opened by Avinash-py - 12
Installation fails (due to recent change?)
#1735 opened by hector-gr - 1
No module named 'torch._six'
#1787 opened by xujin1184104394 - 0
apex not installing
#1793 opened by pradeepdev-1995 - 3
- 3
Whether to support Cuda 12.1
#1748 opened by yangzhipeng1108 - 0
Unclear licensing for contrib/sparsity
#1789 opened by hyandell - 2
Failed to install.
#1708 opened by pengsl-lab - 0
install failure
#1788 opened by 52Hzaaa - 1
apex installation failures
#1764 opened by momo1986 - 0
- 2
ModuleNotFoundError: No module named 'fused_layer_norm_cuda', ubuntu 22.04, Successfully installed apex-0.1
#1755 opened by dhamaraiselvi - 8
When doing pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./, shows ModuleNotFoundError: No module named 'packaging'
#1737 opened by lainmn - 1
[Questing] For apex sparsity model, when i export trt engine with flag sparsity=enable or force, only partial layer picked sparse implementation.
#1780 opened by Bobo-y - 1
- 0
Cannot compile/build cuda_ext on H100
#1778 opened by GuanhuaWang - 11
torch2.0.1 No module named 'torch._six
#1724 opened by darrenwang00 - 1
The build system requires torch. Currently, you cannot build the package using ordinary pip invocations.
#1770 opened by doctorpangloss - 1
memory format option is only supported by strided tensors
#1776 opened by Cheny1m - 2
Installation instructions don't build/install the C modules
#1763 opened by zxti - 1
syncbn with "channel_last=True" produce wrong result when feature_num is not pow-of-two
#1768 opened by Zehaos - 1
Apex installation fails
#1762 opened by yang606 - 1
raise child_exception_type(errno_num, err_msg, err_filename) FileNotFoundError: [Errno 2] No such file or directory: ':/usr/local/cuda/bin/nvcc'
#1700 opened by HSC472 - 4
- 2
Apex is not correctly built for pytorch 2.1.0
#1722 opened by acphile - 1
- 0
- 5
Fail to install apex
#1702 opened by DIY-Z - 2
torch1.13.1 cuda11.6 python3.8 TypeError: unsupported operand type(s) for +: 'NoneType' and 'str'
#1729 opened by yuhuai4554 - 0
GPU memory leak with Flair and APEX
#1744 opened by astropic - 2
FusedAdam doesn't allocate master weights for bfloat16
#1728 opened by cbcase - 0
ninja: error: '/app/csrc/amp_C_frontend.cpp', needed by '/app/build/temp.linux-x86_64-cpython-310/csrc/amp_C_frontend.o', missing and no known rule to make it
#1741 opened by Tolga-Karahan - 1
- 0
fused_layer_norm_cuda.rms_forward_affine gives runtime error when run on cuda:1
#1736 opened by Kushdesh - 2
- 0
Option to only build `amp_C` module
#1725 opened by ezhang887 - 4
Apex Tensor Parallelism and LoRA
#1712 opened by conceptofmind - 0
- 2
- 0
Is there any doc about fmha
#1709 opened by wukong1992 - 0
- 1
`apex.contrib.group_norm` would better have an import guard of `group_norm_cuda`
#1701 opened by crcrpar - 2
Fail to install apex: TypeError: unsupported operand type(s) for +: 'NoneType' and 'str'
#1697 opened by yu1679959321