Skip to content
This repository was archived by the owner on Jan 3, 2023. It is now read-only.

Releases: NervanaSystems/ngraph

nGraph v0.19.0-rc.5

15 May 23:44

Choose a tag to compare

nGraph v0.19.0-rc.5 Pre-release
Pre-release

Release Candidate 5 of nGraph 0.19.0

  • More dynamic shape preparation
  • Distributed interface factored out
  • fp16 and bfloat16 types
  • codegen execution parameterized by context
  • NodeMap, NodeVector, ParameterVector, ResultVector now vectors
    • node_vector.hpp replaced by node.hpp
    • op/parameter_vector.hpp replaced by op/parameter.hpp
    • op/result_vector.hpp replaced by op/result.hpp
  • Additional ONNX ops

nGraph v0.19.0-rc.4

12 May 15:10

Choose a tag to compare

nGraph v0.19.0-rc.4 Pre-release
Pre-release

Release Candidate 4 of nGraph 0.19.0

  • More dynamic shape preparation
  • Distributed interface factored out
  • fp16 and bfloat16 types
  • codegen execution parameterized by context
  • NodeMap, NodeVector, ParameterVector, ResultVector now vectors
    • node_vector.hpp replaced by node.hpp
    • op/parameter_vector.hpp replaced by op/parameter.hpp
    • op/result_vector.hpp replaced by op/result.hpp
  • Additional ONNX ops

nGraph v0.19.0-rc.3

10 May 16:21

Choose a tag to compare

nGraph v0.19.0-rc.3 Pre-release
Pre-release

Release Candidate 3 of nGraph 0.19.0

  • More dynamic shape preparation
  • Distributed interface factored out
  • fp16 and bfloat16 types
  • codegen execution parameterized by context
  • NodeMap, NodeVector, ParameterVector, ResultVector now vectors
    • node_vector.hpp replaced by node.hpp
    • op/parameter_vector.hpp replaced by op/parameter.hpp
    • op/result_vector.hpp replaced by op/result.hpp
  • Additional ONNX ops

nGraph v0.19.0-rc.2

08 May 23:24
11a9f46

Choose a tag to compare

nGraph v0.19.0-rc.2 Pre-release
Pre-release

Release Candidate 2 of nGraph 0.19.0

  • More dynamic shape preparation
  • Distributed interface factored out
  • fp16 and bfloat16 types
  • codegen execution parameterized by context
  • NodeMap, NodeVector, ParameterVector, ResultVector now vectors
    • node_vector.hpp replaced by node.hpp
    • op/parameter_vector.hpp replaced by op/parameter.hpp
    • op/result_vector.hpp replaced by op/result.hpp
  • Additional ONNX ops

nGraph v0.19.0-rc.1

02 May 19:15

Choose a tag to compare

nGraph v0.19.0-rc.1 Pre-release
Pre-release

Release Candidate 1 of nGraph 0.19.0

  • More dynamic shape preparation
  • Distributed interface factored out
  • fp16 and bfloat16 types
  • codegen execution parameterized by context
  • NodeMap, NodeVector, ParameterVector, ResultVector now vectors
    • node_vector.hpp replaced by node.hpp
    • op/parameter_vector.hpp replaced by op/parameter.hpp
    • op/result_vector.hpp replaced by op/result.hpp
  • Additional ONNX ops

nGraph v0.19.0-rc.0

02 May 01:17

Choose a tag to compare

nGraph v0.19.0-rc.0 Pre-release
Pre-release

Release Candidate 0 of nGraph 0.19.0

  • More dynamic shape preparation
  • Distributed interface factored out
  • fp16 and bfloat16 types
  • codegen execution parameterized by context
  • NodeMap, NodeVector, ParameterVector, ResultVector now vectors
    • node_vector.hpp replaced by node.hpp
    • op/parameter_vector.hpp replaced by op/parameter.hpp
    • op/result_vector.hpp replaced by op/result.hpp
  • Additional ONNX ops

nGraph v0.18.1

19 Apr 15:19
950fbf1

Choose a tag to compare

Release 0.18.1 of nGraph

  • Python formatting issue
  • mkl-dnn work-around
  • Event tracing improvements
  • Gaussian error function
  • Begin tracking framework node names
  • ONNX quantization
  • More fusions
  • Allow negative padding in more places
  • Add code generation for some quantized ops
  • Preliminary dynamic shape support
  • initial distributed ops
  • Fixes reference convolution accumulation bug
  • NodeInput and NodeOutput classes prepares for simplifications of Node
  • Test improvements
  • Additional quantization ops
  • Build improvements
  • Performance improvements
  • Fix memory leak
  • Concat optimization
  • Doc updates

nGraph v0.18.1-rc.1

17 Apr 22:27
950fbf1

Choose a tag to compare

nGraph v0.18.1-rc.1 Pre-release
Pre-release

Release Candidate 1 of nGraph 0.18.1

  • mkl-dnn work-around
  • Event tracing improvements
  • Gaussian error function
  • Begin tracking framework node names
  • ONNX quantization
  • More fusions
  • Allow negative padding in more places
  • Add code generation for some quantized ops
  • Preliminary dynamic shape support
  • initial distributed ops
  • Fixes reference convolution accumulation bug
  • NodeInput and NodeOutput classes prepares for simplifications of Node
  • Test improvements
  • Additional quantization ops
  • Build improvements
  • Performance improvements
  • Fix memory leak
  • Concat optimization
  • Doc updates
  • Python formatting issue

nGraph v0.18.1-rc.0

16 Apr 00:23
85884f3

Choose a tag to compare

nGraph v0.18.1-rc.0 Pre-release
Pre-release

Release Candidate 0 of nGraph 0.18.1

  • mkl-dnn work-around
  • Event tracing improvements
  • Gaussian error function
  • Begin tracking framework node names
  • ONNX quantization
  • More fusions
  • Allow negative padding in more places
  • Add code generation for some quantized ops
  • Preliminary dynamic shape support
  • initial distributed ops
  • Fixes reference convolution accumulation bug
  • NodeInput and NodeOutput classes prepares for simplifications of Node
  • Test improvements
  • Additional quantization ops
  • Build improvements
  • Performance improvements
  • Fix memory leak
  • Concat optimization
  • Doc updates

nGraph v0.18.0

11 Apr 19:25
c5d52f1

Choose a tag to compare

Release 0.18.0 of nGraph

  • Event tracing improvements
  • Gaussian error function
  • Begin tracking framework node names
  • ONNX quantization
  • More fusions
  • Allow negative padding in more places
  • Add code generation for some quantized ops
  • Preliminary dynamic shape support
  • initial distributed ops
  • Fixes reference convolution accumulation bug
  • NodeInput and NodeOutput classes prepares for simplifications of Node
  • Test improvements
  • Additional quantization ops
  • Build improvements
  • Performance improvements
  • Fix memory leak
  • Concat optimization
  • Doc updates