issues
search
artyom-beilis
/
pytorch_dlprim
DLPrimitives/OpenCL out of tree backend for pytorch
http://blog.dlprimitives.org/
MIT License
224
stars
16
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Error when trying to use dlprimitives in the diffusion model
#78
YuHaozhe0703
opened
1 week ago
0
Updating documents and PyTorch compatibility for all systems?
#77
tangjinchuan
opened
2 weeks ago
12
Undefined symbols
#76
skn123
opened
1 month ago
3
Prebuilt binaries for Windows
#75
SA-j00u
opened
2 months ago
4
Support of time exceeded error.
#74
sukamenev
opened
3 months ago
0
Need rename benchark_all.sh to benchmark_all.sh
#73
sukamenev
opened
3 months ago
1
Strange error in test: Diff too big
#72
sukamenev
opened
3 months ago
2
Error in OpenCL on CPU: python tests/validate_network.py --device privateuseone:0
#71
sukamenev
closed
3 months ago
2
Error in python tests/validate_network.py --device privateuseone:1
#70
sukamenev
opened
3 months ago
5
Error by python tests/test_op.py --device ocl:1
#69
sukamenev
opened
3 months ago
4
Benchmark script do not run. 'tools' dir is absent.
#68
sukamenev
closed
3 months ago
2
I fixed all of bugs with contiguous()
#67
sukamenev
opened
3 months ago
8
I fixed all of bugs with contiguous() and added speed improvements: convert to contiguous only if needed
#66
sukamenev
closed
3 months ago
1
Many warnings with PyTorch 2.2.1: aten::_foreach_lerp_.Scalar, aten::_foreach_mul_.Scalar, aten::_foreach_sqrt and others
#65
sukamenev
opened
3 months ago
1
Small bugs and improvement for self, self_c, contiguous()
#64
sukamenev
closed
3 months ago
2
Model 5MB and batch of data ~50MB takes up 1 GB gpu memory
#63
sukamenev
opened
3 months ago
5
clamp and clamp_min need to use self_c for X instead of self
#62
artyom-beilis
opened
3 months ago
4
Error with torch.save(model.state_dict(), sPath): don't know how to determine data location of torch.storage.UntypedStorage
#61
sukamenev
opened
3 months ago
3
Adding support clamp_min_out for nn.BCEWithLogits()
#60
sukamenev
closed
3 months ago
0
By using nn.BCEWithLogitsLoss() i got: 'aten::clamp_min.out' is not currently supported on the ocl backend.
#59
sukamenev
closed
3 months ago
2
When I test mnist.py after completing make, I get an error
#58
ParadosBoy
opened
3 months ago
9
A question about Benchmark
#57
ggangliu
closed
4 months ago
2
Error when Running Pyannote Audio on OpenCL Backend - RuntimeError: Running sums are expected to be present
#55
HideyoshiNakazone
opened
4 months ago
4
Request to support more functions!
#54
CCMKCCMK
opened
5 months ago
1
LLM support
#53
huizhanyi
opened
5 months ago
2
without backend specific native_batch_norm operator, the test goes failure and trainning goes well
#52
smilealvin92
opened
5 months ago
3
Issue with OpenCL 3.0
#51
stevef1uk
closed
5 months ago
3
RuntimeError: Only last dimension is supported for softmax
#50
filipposkat
opened
5 months ago
8
SEGFAULT, when loaded from C++ frontend
#49
tada123
opened
7 months ago
6
Why register the relu operator in AutogradPrivateUse1 ?
#47
JingliangGao
closed
7 months ago
4
Problem about opencl backend in libtorch
#46
johndoanee
opened
7 months ago
5
The operator 'aten::index_select' is not currently supported on the ocl backend.
#44
leviathanch
opened
7 months ago
1
The operator 'aten::arange.start_out' is not currently supported on the ocl backend.
#43
leviathanch
opened
7 months ago
2
variable of "or" in pooling.cl
#42
smilealvin92
closed
7 months ago
1
The operator 'aten::linspace.out' is not currently supported on the ocl backend.
#41
karolherbst
closed
7 months ago
3
Compile Error: cl_mem p=static_cast<cl_mem>(tt.getIntrusivePtr()->storage().data());
#40
madtunebk
closed
10 months ago
2
RuntimeError: Failed to build program source pooling on mnist.py
#39
DWhettam
opened
12 months ago
3
Operators not supported
#38
VirxEC
closed
7 months ago
8
Rusticl & `OCL_PATH-NOTFOUND`
#37
VirxEC
closed
1 year ago
4
tensor comparison work error under some testdata!!!
#36
Xrigok
opened
1 year ago
3
Pytorch 2.0 and greater
#35
apcameron
opened
1 year ago
7
python mnist.py --device ocl:0 runs on UHD Graphics 630, remains heavy load without progress after first epoch
#34
rmast
opened
1 year ago
5
Pytorch 2.0 reduced ops specs
#32
nlgranger
closed
7 months ago
1
[Donations] LiberaPay or OpenCollective?
#31
EchedelleLR
opened
1 year ago
0
Implementation of aten::eq.Scalar_out
#30
yamaura
closed
1 year ago
1
implementation for aten::max.dim_max
#29
eazy-f
closed
7 months ago
2
Occasional Memory Access Fault (GPU)
#28
arch-user-france1
closed
4 months ago
3
GPT2 model - Running sums are expected to be present
#27
arch-user-france1
opened
1 year ago
6
Runtime Error
#26
GreatestTrain
closed
1 year ago
2
Support for MacOS Ventura
#25
dbl001
opened
1 year ago
21
Next