issues
search
microsoft
/
antares
Antares: an automatic engine for multi-platform kernel generation and optimization. Supporting CPU, CUDA, ROCm, DirectX12, GraphCore, SYCL for CPU/GPU, OpenCL for AMD/NVIDIA, Android CPU/GPU backends.
Other
449
stars
46
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Is there any document for performance benchmark result vs pytorch2.1 compile mode?
#377
fsword73
opened
6 months ago
1
update samples
#376
shenzhiy21
closed
8 months ago
1
Lack operator implementation for DirectX: torch.abs()
#375
PedroMagar
closed
4 months ago
2
Is ROCm no longer supported by 0.9.x?
#374
Lookforworld
opened
11 months ago
16
will this project replace torch-directml?
#373
pty819
opened
1 year ago
2
Is this project based on AI? What is the goal of this project?
#372
yhyu13
closed
1 year ago
3
Synchronize Lastest
#371
ghostplant
closed
1 year ago
0
Assertion error: SDK for `c-rocm_win64` is not configured correctly,
#370
harish0201
opened
1 year ago
3
Support msvc build
#369
mzmssg
closed
1 year ago
0
[Error] error: ‘CHECK_EQ’ was not declared in this scope; did you mean ‘CHECK_OK’?
#367
Looong01
opened
1 year ago
17
The residue of the last issue (#365)
#366
Looong01
closed
1 year ago
0
Fail to compile, when I use "AMDGFX=gfx1031 BACKEND=c-rocm_win64 antares"
#365
Looong01
closed
1 year ago
9
Not an issue but a question due to lack of docs.
#364
mudapanda2
opened
1 year ago
1
is it possible c-ocl_*_win64
#363
kh-abd-kh
opened
1 year ago
15
append __habs for float16 abs in cuda backend
#362
LeiWang1999
closed
2 years ago
1
register hostmem
#361
ghostplant
closed
2 years ago
0
Benchmarks
#360
sebastienwood
opened
2 years ago
3
how can antares surport loop which index doesn't start with 0
#359
lethean1
opened
2 years ago
5
Can antares assign specified gpus for evaluation?
#358
LeiWang1999
closed
2 years ago
1
Antares dll export flag
#357
mzmssg
closed
2 years ago
1
add dxModuleSetCompat for HLSL lib
#356
ghostplant
closed
2 years ago
0
Merge space of tiling & w_reduce
#355
ghostplant
closed
2 years ago
0
[Help Request] How can Antares IR support stride size > 1 's Slice operation?
#354
LeiWang1999
closed
2 years ago
3
[BUG] Tune a bert-base-fp16 failed
#353
LeiWang1999
closed
2 years ago
1
add base2 option
#352
ghostplant
closed
2 years ago
0
improve gpu gemm tuning space
#351
ghostplant
closed
2 years ago
0
Change the cache directory
#350
LeiWang1999
closed
2 years ago
4
This repo is missing important files
#349
microsoft-github-policy-service[bot]
closed
2 years ago
0
Adding Microsoft SECURITY.MD
#348
microsoft-github-policy-service[bot]
closed
2 years ago
0
Extend normcdf & remove signed char in hlsl backend
#347
mzmssg
closed
2 years ago
0
Support infinite/nan from xbox tuner
#346
mzmssg
closed
2 years ago
0
-
#345
ghost
closed
2 days ago
6
-
#344
ghost
closed
2 years ago
1
access AMDGFX for custom Windows ROCm spec
#343
ghostplant
closed
2 years ago
0
-
#342
ghost
closed
2 years ago
20
update README.md with backend listing
#341
ghostplant
closed
2 years ago
0
Workaround xbox tanh numeric error, fix wrapper
#340
mzmssg
closed
2 years ago
0
remove error for c-hlsl_win64 running on Linux
#339
ghostplant
closed
2 years ago
0
Fix thread size calculation
#338
mzmssg
closed
2 years ago
0
Prohibit generating improper shaders for xbox, fix compiler error check
#337
mzmssg
closed
2 years ago
0
fix undefined intermediate_output name
#336
ghostplant
closed
2 years ago
0
update hmax, hmin defines
#335
ghostplant
closed
2 years ago
0
update plugin apis for torch
#334
ghostplant
closed
2 years ago
0
enhance torch-setup backend detection
#333
ghostplant
closed
2 years ago
0
Refine error message for xbox
#332
mzmssg
closed
2 years ago
0
Refine error message for xbox
#331
mzmssg
closed
2 years ago
1
hide timeout msg when PROGRESS=1
#330
ghostplant
closed
2 years ago
0
update torch plugin
#329
ghostplant
closed
2 years ago
0
periodically updates
#328
ghostplant
closed
2 years ago
0
add options for plugin setup
#327
ghostplant
closed
2 years ago
0
Next