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
933
stars
155
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
[BUG] Failed to compile yolov5 onnx format model
#532
fomiuna
opened
1 week ago
0
[Security] Fix dangling s3 bucket
#531
mzmssg
closed
3 months ago
0
Can nnfusion train VHR images whoes feature map in a CNN model exceeds the gpu memory?
#530
AugustineYang
opened
6 months ago
3
[BUG] welder docker build failed
#529
avatarstones
opened
7 months ago
1
support folding onnx>2GB with onnxruntime
#528
mzmssg
closed
8 months ago
0
Some issue when i reproduce.can someone help me
#527
Muuuchen
closed
7 months ago
0
Can't download these models in model zoo
#526
Lhr-study
opened
11 months ago
1
Does nnfusion support the partition of a convolution layer?
#525
baichen318
opened
1 year ago
0
Some initial implementations for DSL frontend
#524
donglinb
closed
1 year ago
0
Update Cocktailer Artifact
#523
heheda12345
closed
1 year ago
0
Fix antares ir for gather
#522
mzmssg
closed
1 year ago
0
[BUG] Integrate the latest cudnn with nnfusion
#521
wxthu
opened
1 year ago
1
artifacts problems
#520
YixinSong-e
opened
1 year ago
1
whether supported on other platforms
#519
wxthu
opened
1 year ago
4
Cocktailer Artifact
#518
heheda12345
closed
1 year ago
1
QUESTION: how to add a new type of gpu for roller?
#517
cavalry-yu
opened
1 year ago
1
Merge welder
#516
xiayuqing0622
closed
1 year ago
0
fix bug
#515
xiayuqing0622
closed
1 year ago
0
[Bug fix] Register Fusion Pass fuse policy assign wrong output edges
#514
LeiWang1999
opened
1 year ago
0
Disable external result memory
#513
mzmssg
closed
1 year ago
0
meet int16 request in fp16 bloom model
#512
LeiWang1999
closed
1 year ago
1
Fix FileNotFoundError when tuning_steps is 0
#511
DeanAlkene
closed
1 year ago
0
Release loaded DLL libraries in destructor of class Executor
#510
donglinb
closed
11 months ago
0
Flags to disable emitting cudaSetDevice call
#509
mzmssg
closed
1 year ago
0
QUESTION: What are the meanings of the parameters of different Arch in the roller
#508
GitHubstart0916
opened
1 year ago
0
QUESTION: does roller support fused operator?
#507
SunflowerAries
opened
1 year ago
0
Yuqxia/diffuser
#506
xiayuqing0622
closed
1 year ago
1
Can NNfusion fuse cudnn's kernel?
#505
five12
opened
1 year ago
0
Some fixes
#504
mzmssg
closed
1 year ago
0
Support onnx einsum op
#503
mzmssg
closed
1 year ago
0
Invoke all ONNX tests
#502
wenxcs
closed
1 year ago
0
fix argmax/argmin missing arguments in tensorflow import unitest
#501
LeiWang1999
closed
1 year ago
0
Sync antares wrapper
#500
mzmssg
closed
1 year ago
0
Fix symbolic codegen and sync kernel tuning
#499
jlxue
closed
1 year ago
0
convert int64 to long long for built-in cuda kernels
#498
mzmssg
closed
1 year ago
0
support xbox branch op check: Abs Acos Add And ArgMax ArgMin
#497
LeiWang1999
closed
1 year ago
3
revert fp32 cast for tan, normcdf
#496
xiayuqing0622
closed
1 year ago
0
Revert "fix ceil,floor,tan,tanh"
#495
xiayuqing0622
closed
1 year ago
0
added version 15 support for Shape operator
#494
donglinb
closed
1 year ago
1
fix ceil,floor,tan,tanh
#493
xiayuqing0622
closed
1 year ago
0
Fix op with fp16 type
#492
jlxue
closed
1 year ago
0
Make test flag as bool
#491
mzmssg
closed
1 year ago
0
Copy python entrypoint to hlsl project
#490
mzmssg
closed
1 year ago
0
Fix merge bug
#489
jlxue
closed
1 year ago
0
onnx backend test script for fp16/64
#488
wenxcs
closed
1 year ago
0
Xbox fix
#487
wenxcs
closed
1 year ago
0
Check and fix operators: AvgPool, GlobalAvgPool, MaxPool, GlobalMaxPool, Greater, GreaterOrEqual, Less, LessOrEqual, Identity, InstanceNorm, LSTM, LayerNorm, LeakyRelu, Log, LogSoftmax, MatMul, Size, Reciprocal
#486
xysmlx
closed
1 year ago
0
Convert int64 to long long for antares cuda backend
#485
mzmssg
closed
1 year ago
2
fix bug
#484
xiayuqing0622
closed
1 year ago
0
Yuxia/fix result
#483
xiayuqing0622
closed
1 year ago
0
Next