issues
search
microsoft
/
nnfusion
A flexible and efficient deep neural network (DNN) compiler that generates high-performance executable from a DNN model description.
MIT License
952
stars
158
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
enable function codegen
#383
xiayuqing0622
closed
2 years ago
0
support GRU op
#382
mzmssg
closed
2 years ago
1
Fix correctness problems in GNN training
#381
xysmlx
closed
2 years ago
1
No need to extract tarball after installation
#380
siahuat0727
closed
2 years ago
0
Add signatures to detect reusable kernel + nnf rt mem reservation + jit optional kwargs
#379
siahuat0727
closed
2 years ago
1
Implement jit demo version and add tests
#378
siahuat0727
closed
2 years ago
1
fix onnx export of dict input in python interface
#377
xysmlx
closed
2 years ago
0
[ENHANCEMENT] Let cmake find cudnn libraries from /usr/lib/x86_64-linux-gnu and /usr/include
#376
xysmlx
opened
2 years ago
0
fix shape check in infer_shape for scalar tensor
#375
jlxue
closed
2 years ago
0
execute install_dependency.sh in python setup
#374
xysmlx
closed
2 years ago
0
[Question]How to implement the barrier-rTask in generated code
#373
zqj2333
opened
2 years ago
2
Support PyTorch dict input
#372
mzmssg
closed
2 years ago
0
[BUG] Possible cuda context problem in mpirun/horovodrun
#371
xysmlx
opened
2 years ago
0
QUESTION: any document about how vDevice and vEU are implemented?
#370
puddingfjz
opened
2 years ago
7
[BUG] Support DCNv2 operator
#369
Fvoiretryzig
opened
2 years ago
1
[BUG] Kernel tuning list is not same as Antares kernel emitting list
#368
xysmlx
opened
2 years ago
0
[ENHANCEMENT] CLI throws error when the input flags do not match NNFusion's flags
#367
xysmlx
opened
2 years ago
0
Convert scalar to 1d tensor in python interface
#366
mzmssg
closed
2 years ago
0
add -fcodegen_pybind
#365
xiayuqing0622
closed
2 years ago
0
Support -ftuning_list in kernel tuning pass
#364
xysmlx
opened
2 years ago
1
Support python wheel distribution
#363
wenxcs
closed
2 years ago
6
[BUG] model compile error cuz shape disagree
#362
lixeon
closed
2 years ago
2
Use curl cmd line to replace libcurl
#361
yiyione
closed
2 years ago
0
[BUG] The build script of the generated code does not copy Constant/ folder to the build path
#360
xysmlx
opened
2 years ago
0
[BUG] -fcodegen_debug generates CUDA debug code in GENERIC_CPU backend
#359
xysmlx
opened
2 years ago
0
[BUG] performance issue of CUDA kernel `SUM`
#358
gbxu
opened
2 years ago
0
[BUG] compiling fails with too much arguments when multi CPU threads
#357
gbxu
opened
2 years ago
0
[BUG] wrong code generating for model I/O
#356
gbxu
opened
2 years ago
0
[FEATURE] the optimization of dot transpose and batchmatmul transpose
#355
gbxu
opened
2 years ago
0
[BUG] multi reshape folding fails in same cases
#354
gbxu
opened
2 years ago
0
[Hold for pipeline test] Support control-flow
#353
xysmlx
opened
2 years ago
0
[BUG] useless input of GatherGrad
#352
gbxu
closed
2 years ago
1
Fix AntaresCpuKernelEmitter and add ir_based_fusion in GENERIC_CPU backend
#351
xysmlx
closed
2 years ago
0
[BUG] GENERIC_CPU backend can not generate model code with Antares kernels
#350
xysmlx
closed
2 years ago
0
constant folding pass parallel
#349
yiyione
closed
2 years ago
0
Fix type conversion in ort_run_frozen script
#348
mzmssg
closed
2 years ago
0
Does NNfusoin support tensor core?
#347
donglinz
closed
2 years ago
3
add firfusion_blocklist
#346
xiayuqing0622
closed
2 years ago
0
Add roll backward
#345
yiyione
closed
2 years ago
1
A100 support
#344
lscat11
closed
2 years ago
1
Convert unsupported onnx op to generic op
#343
mzmssg
closed
2 years ago
0
Add Avg Pool Backwards;
#342
wenxcs
closed
2 years ago
1
Support statically build
#341
wenxcs
closed
2 years ago
0
Yuqxia/add ir gen with extension
#340
xiayuqing0622
closed
2 years ago
0
add conv1d bwd
#339
xiayuqing0622
closed
2 years ago
0
fix logical backward ops
#338
jlxue
closed
2 years ago
0
[BUG] Equal operator backward reports check error
#337
xysmlx
closed
2 years ago
0
Register backward for logical ops
#336
jlxue
closed
2 years ago
0
Add sigmoid autodiff
#335
siahuat0727
closed
2 years ago
6
Fix variable name typo
#334
siahuat0727
closed
2 years ago
2
Previous
Next