issues
search
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
[NCCL] Prevent premature destroy of PGs following PyTorch upstream change
#1859
eqy
closed
6 hours ago
0
[GroupNorm] Skip GroupNorm tests on A16, A2 etc.,
#1858
eqy
closed
2 days ago
0
AttributeError: module 'torch.compiler' has no attribute 'is_compiling'
#1857
LukeLIN-web
opened
1 week ago
5
Failed the last time, succeeded the next time?上一次还失败,下一次就成功了?
#1856
zhangs-a-n
opened
3 weeks ago
2
`Tensor.type()` -> `Tensor.scalar_type()`
#1855
crcrpar
closed
3 weeks ago
0
[PT2] Normalisation: use manual impl when compiling
#1854
alexdremov
closed
3 weeks ago
1
FusedRMSNormAffineMixedDtypesFunction is not importable in the PyTorch build without distributed support
#1853
IvanYashchuk
opened
3 weeks ago
0
关于解决ModuleNotFoundError: No module named 'torch'导致安装失败
#1852
Eikwang
opened
1 month ago
23
How to install apex
#1851
Zerycii
opened
1 month ago
8
到底要怎样才能安装apex
#1850
poyingshihuang
opened
1 month ago
1
AdamW implementation does not truly decouple learning rate and weight decay
#1849
leenachennuru
opened
1 month ago
2
Failed to build installable wheels for some pyproject.toml based projects (apex)
#1848
RukaiaAfsana
opened
1 month ago
4
Avoid unnecessary NCCL collective coalescing in distributed optimizer
#1847
timmoon10
closed
1 month ago
0
Literature associated with fused_dense
#1846
prmudgal
opened
1 month ago
1
fix groupnorm int32 index overflow
#1845
tlogn
opened
1 month ago
2
Main
#1844
63days
closed
2 months ago
0
ASP Automatic Sparsity forward function For Loop Error
#1843
maro-jeon
opened
2 months ago
0
Discrepancy with Optimizer States and Model State Dict when using store_param_remainders==True
#1842
alxzhang-amazon
opened
2 months ago
8
Gradient Overflow with Specific GPU Combinations in Multi-GPU Setup (NVIDIA RTX 3090)
#1841
SylU0
opened
2 months ago
0
No module named 'amp_C'
#1840
KanyuBao
opened
2 months ago
0
loss scale
#1839
yjy-10
opened
2 months ago
0
How to improve training performance with Apex package
#1838
tjk9501
opened
2 months ago
0
Reformat Grad Output If It's Not Channels Last
#1837
alpha0422
closed
2 months ago
0
Add Unittest For Distributed Adam With CUDA Graph
#1836
alpha0422
closed
2 months ago
0
Traceable GroupNorm
#1835
alpha0422
closed
2 months ago
0
install bug with pytorch2.0.1
#1834
Duanjinyi1
opened
2 months ago
1
Unsupported NVHPC compiler found. nvc++ is the only NVHPC compiler that is supported.
#1833
mz687
closed
1 month ago
0
Enhance Distributed Fused Adam
#1832
alpha0422
closed
2 months ago
0
Installation with Cuda extentions is failling
#1831
SaiedaJN
opened
3 months ago
2
remove `run_transformer` from default lists
#1830
crcrpar
closed
1 month ago
0
Fix DistributedTestBase for transformer distributed tests
#1829
xwang233
closed
3 months ago
1
No CUDA runtime is found, using CUDA_HOME='/home/shengjieyi/cuda1108' .
#1828
vvsherryvv
opened
3 months ago
0
Unsuccessful installation of apex library. (Preparing metadata (pyproject.toml) did not run successfully.)
#1827
ssaral
opened
3 months ago
4
Slow Performance with "Exhaustive Search" Permutation Strategy for Channel Pruning in CNN
#1826
Ulorewien
opened
3 months ago
0
Fix illegal memory access with multi_tensor_apply size above INT_MAX
#1825
gdb
closed
3 months ago
3
Unable to install Apex
#1824
JoongunPark
opened
3 months ago
1
Setting up Apex and get this error: ModuleNotFoundError: No module named 'torch'
#1823
Mayolov
closed
3 months ago
10
Install set.up
#1822
Maritime-Moon
opened
4 months ago
0
Allow Configurable Cache Directory
#1821
leimao
closed
4 months ago
0
[Distributed optimizer] Do not monkey-patch class methods
#1820
timmoon10
closed
4 months ago
0
Unknown CUDA arch (compute) or GPU not supported error while installing on docker ubuntu with cuda 12.1
#1819
AvisP
opened
4 months ago
1
NCCLAllocator: Fix build failure
#1818
Aidyn-A
opened
4 months ago
3
Unable to install Apex on Linux(debian) with CUDA 12.1 and torch 2.2.2
#1817
SamitM1
opened
4 months ago
2
Release GIL
#1816
crcrpar
closed
4 months ago
1
unable to install
#1815
lxy51
opened
4 months ago
5
Release GIL when calling C extensions
#1814
szmigacz
closed
4 months ago
0
deprecate uses of torch.cuda.amp
#1813
Fuzzkatt
closed
4 months ago
2
Only print the warning message about `TORCH_CUDA_ARCH_LIST` if not set
#1812
aurianer
opened
4 months ago
0
fixup concats for grouped convolution
#1811
techshoww
opened
5 months ago
0
Unable to install Apex
#1810
Anupam-5
opened
5 months ago
2
Next