NVIDIA/apex

Issues

• ImportError: cannot import name '_library_root_logger' from 'apex' (unknown location)

  #1803 opened 6 days ago by BBALU1660
  0

• On installing apex (+without sudo/docker)

  #1781 opened 3 months ago 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 2 months ago by doctorpangloss
  1

• [ INSTALLATION ] - Not able to install apex on a Linux machine

  #1799 opened a month ago by MBadriNarayanan
  1

• Cannot install apex on the machine of CUDA 12.2

  #1761 opened 6 months ago by momo1986
  6

• Running apex with error: AttributeError: module 'torch.distributed' has no attribute '_reduce_scatter_base'

  #1773 opened 4 months ago by cs-wangfeng
  3

• Contrib unit test failure in `openfold_triton/test_fused_adam_swa.py::FusedAdamSWATestCase::test_fused_update_on_random_data`

  #1802 opened 24 days ago by xwang233
  0

• Not Able to install apex.

  #1800 opened a month ago by Avinash-py
  0

• Installation fails (due to recent change?)

  #1735 opened 8 months ago by hector-gr
  12

• No module named 'torch._six'

  #1787 opened 3 months ago by xujin1184104394
  1

• apex not installing

  #1793 opened 2 months ago by pradeepdev-1995
  0

• Installation Problem: RuntimeError: Error compiling objects for extension

  #1779 opened 3 months ago by wwma
  3

• Whether to support Cuda 12.1

  #1748 opened 7 months ago by yangzhipeng1108
  3

• Unclear licensing for contrib/sparsity

  #1789 opened 2 months ago by hyandell
  0

• Failed to install.

  #1708 opened 10 months ago by pengsl-lab
  2

• install failure

  #1788 opened 3 months ago by 52Hzaaa
  0

• apex installation failures

  #1764 opened 5 months ago by momo1986
  1

• cannot import name 'AutoencoderKLTemporalDecoder' from 'diffusers.models'

  #1785 opened 3 months ago by zj19941113
  0

• ModuleNotFoundError: No module named 'fused_layer_norm_cuda', ubuntu 22.04, Successfully installed apex-0.1

  #1755 opened 6 months ago by dhamaraiselvi
  2

• When doing pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./, shows ModuleNotFoundError: No module named 'packaging'

  #1737 opened 8 months ago by lainmn
  8

• [Questing] For apex sparsity model, when i export trt engine with flag sparsity=enable or force, only partial layer picked sparse implementation.

  #1780 opened 3 months ago by Bobo-y
  1

• The storage format of the compressed matrix in module 'apex.contrib.sparsity'

  #1706 opened 10 months ago by Shan2L
  1

• Cannot compile/build cuda_ext on H100

  #1778 opened 3 months ago by GuanhuaWang
  0

• torch2.0.1 No module named 'torch._six

  #1724 opened 9 months ago by darrenwang00
  11

• The build system requires torch. Currently, you cannot build the package using ordinary pip invocations.

  #1770 opened 5 months ago by doctorpangloss
  1

• memory format option is only supported by strided tensors

  #1776 opened 4 months ago by Cheny1m
  1

• Installation instructions don't build/install the C modules

  #1763 opened 5 months ago by zxti
  2

• syncbn with "channel_last=True" produce wrong result when feature_num is not pow-of-two

  #1768 opened 5 months ago by Zehaos
  1

• Apex installation fails

  #1762 opened 6 months ago 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 10 months ago by HSC472
  1

• ModuleNotFoundError: No module named 'fast_multihead_attn'

  #1717 opened 9 months ago by ICENacl
  4

• Apex is not correctly built for pytorch 2.1.0

  #1722 opened 9 months ago by acphile
  2

• No module named 'amp_C' error for py3.9 pytorch2.1.0 cuda12.1

  #1757 opened 6 months ago by rocke2020
  1

• why a kernel like CUDAFunctor_add appears when testing MixedFusedRMSNorm?

  #1753 opened 6 months ago by HangJie720
  0

• Fail to install apex

  #1702 opened 10 months ago by DIY-Z
  5

• torch1.13.1 cuda11.6 python3.8 TypeError: unsupported operand type(s) for +: 'NoneType' and 'str'

  #1729 opened 9 months ago by yuhuai4554
  2

• GPU memory leak with Flair and APEX

  #1744 opened 8 months ago by astropic
  0

• FusedAdam doesn't allocate master weights for bfloat16

  #1728 opened 9 months ago by cbcase
  2

• 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 8 months ago by Tolga-Karahan
  0

• error: command '/usr/local/cuda-11.3/bin/nvcc' failed with exit code 1

  #1738 opened 8 months ago by Brion112233
  1

• fused_layer_norm_cuda.rms_forward_affine gives runtime error when run on cuda:1

  #1736 opened 8 months ago by Kushdesh
  0

• is it possible to update `conda-forge/nvidia-apex` to a recent tag?

  #1730 opened 9 months ago by stas00
  2

• Option to only build `amp_C` module

  #1725 opened 9 months ago by ezhang887
  0

• Apex Tensor Parallelism and LoRA

  #1712 opened 10 months ago by conceptofmind
  4

• Scale optimizer state with updated distributed size

  #1707 opened 10 months ago by jayakrishnaanvesh
  0

• AttributeError: module 'apex.amp' has no attribute 'state_dict'

  #1710 opened 10 months ago by caoren-shuai
  2

• Is there any doc about fmha

  #1709 opened 10 months ago by wukong1992
  0

• Apex installation is stucked in infinite loop with printing warnings

  #1703 opened 10 months ago by GalJakob
  0

• `apex.contrib.group_norm` would better have an import guard of `group_norm_cuda`

  #1701 opened 10 months ago by crcrpar
  1

• Fail to install apex: TypeError: unsupported operand type(s) for +: 'NoneType' and 'str'

  #1697 opened a year ago by yu1679959321
  2