Skip to content
Snippets Groups Projects
Select Git revision
  • OptimizeSub
  • WinCI
  • add_abs
  • allowNoInputProducer
  • batchnorm_flag
  • betterpip
  • bitshift_op
  • bn_flag
  • dev protected
  • feat_formatting
  • fix
  • main default protected
  • old_scaling
  • scaling
  • stable
  • v0.6.1
  • v0.6.0
  • v0.5.1
  • v0.5.0
  • v0.4.0
  • v0.3.2
  • v0.3.1
  • v0.3.0
  • v0.2.2
  • v0.2.1
  • v0.2.0
  • v0.1.0
  • v0.0.1
28 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.019Jun1018May1576529Apr262524232216151410928Mar16320Feb31Jan28231710213Dec12111065229Nov2726252019141287631Oct2928201587127Sep2625181210654327Aug2221161514131265131Jul2485432130Jun282725242120191817141312111054331May3029151413230Apr2221181611943229Mar28272622211914828Feb19127652124Jan155414Dec1110987547Nov23Oct20171354321Sep2018141311Fixed FC backward with biasdevdevrefactor resize's interpolation functionsadd Resize operatorVersion 0.6.0 -> 0.6.1v0.6.1 mainv0.6.1 mainFix roundingRemoved wrong assertationsshowCudaVersion now use fmt print instead of Log::info.Fix documentation link.Added missing BitShift includeMissing changesAdded BitShiftRemoved empty backward()Replaced cudaMalloc with thrust::device_vectorSpecify which manylinux 2.28 docker to use, avoiding differences between CI and local.v0.6.0v0.6.0Remove need for sudo in the CI by setting ownership inside of CIBW docker.Add back sudo in gitlabci.Update cuda 12.4 -> 12.8Remove sudo from gitlabci.Add installation of gcc 13.Update search path.Use CUDNN image to test backend cuda.Update path for new runner.Update CIBW.Add option -DPYBIND11_FINDPYTHON=ON.Update minimum version to python 3.10Update 0.5.1 -> 0.6.0Add Alpha and Beta in Backward methodadd: function to get inference time and output of some GraphViewedit: TensorImpl: gave up original approach to use class partialedit: TensorImpl: try to fix template ?edit: ILayerNorm: fixed testedit: TensorImpl.{hpp,cu}: Fixed template in the general caseUpdate TensorImpl.hpp add include for type_traitsedit: ILayerNormImpl: fixed type possible overflow warningfix: TensorImpl: Enable if on return type instead of parameterAdded missing guards for multi-GPUMake device selection effectiveCorrected Pad fixFixed wrong paddingUpdate Pybind 2.10.4 -> 2.13.6

Copyright © Eclipse Foundation, Inc. All Rights Reserved.     Privacy Policy | Terms of Use | Copyright Agent