Version 1.13.0 Release

release-note-bugfix

• Add test case for index of sorting equal values.
• Fix the CUDA transpose operation for large N-dimensional inputs.

release-note-build

• correct gnumakefile link library name

release-note-core

• fix forward_all to keep values of all given variables

release-note-doc

• Fix Equation Errors in Doc
• Update nnabla convert doc at 20200930
• Various Enhancements for Doc
• Add NNabla NAS to README

release-note-format-converter

• Add QuantizeLinear/DequantizeLinear support for onnx Importer.

release-note-op-layer

• Add RMSpropGraves
• Memory Optimization

release-note-utility

• Graph Converter Enhancement

Install the latest nnabla by:

pip install nnabla
pip install nnabla-ext-cuda102 # For CUDA version 10.2 users

You can also install the cuda extension with specific versions from one of the following. See also FAQ

• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.6)
• nnabla-ext-cuda100 (CUDA 10.0 x cuDNN 7.6)
• nnabla-ext-cuda102 (CUDA 10.2 x cuDNN 7.6)

pip install nnabla
pip install nnabla_ext_cuda102  # For CUDA 10.2 x cuDNN 7.6 users

For distributed training, You need to install the correct package that
matches the version of MPI installed on your system.

We prepared following packages.

• nnabla-ext-cuda90-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda90-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda90-nccl2-mpi3-1-6
• nnabla-ext-cuda100-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda100-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda100-nccl2-mpi3-1-6
• nnabla-ext-cuda102-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda102-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda102-nccl2-mpi3-1-6

If you want to use a version of MPI not listed above, you need to build it from the source.

Additional setup may be required depending on your OS or environment. Please check Python Package Installation Guide for details.

Install Guide
To use C++ inference feature, follow the demonstration on MNIST inference in C++.

Demonstration

Build Guide
The "nnabla-ext-cuda" package is temporarily unavailable. Use of this package is not recommended. Please install nnabla-ext-cuda102, nnabla-ext-cuda100, nnabla-ext-cuda90 instead.
The following nnabla CUDA extension packages have been deprecated and the PyPi repository has been (or going to be) closed.

• nnabla-ubuntu16
• nnabla-ubuntu18
• nnabla-ext-cuda91
• nnabla-ext-cuda92
• nnabla-ext-cuda101
• nnabla-ext-cuda90-nccl2-ubuntu16
• nnabla-ext-cuda100-nccl2-ubuntu16
• nnabla-ext-cuda100-nccl2-ubuntu18
• nnabla-ext-cuda101-nccl2-ubuntu16
• nnabla-ext-cuda101-nccl2-ubuntu18

The following "nnabla-ext-cuda" docker images have been deprecated.

• py27-cuda92
• py36-cuda92
• py37-cuda92
• py27-cuda92-v1.0.xx
• py36-cuda92-v1.0.xx
• py37-cuda92-v1.0.xx

We've decided to change nnabla's versioning policy to semantic versioning.
This change has been applied from version 1.1.0.

Python 2 is no longer be supported from v1.5.0.
CUDA8.0 is no longer be supported from v1.6.0.

Version 1.12.0 Release

release-note-doc

• Update nnabla convert doc at 20200807

release-note-format-converter

• Fix FusedBatchNormalization for ONNX Exporter.
• Rename variable for repeat node.
• Fill input roi of ONNX Resize.

release-note-utility

• fix several problems might cause crash or system resource use out.

Install the latest nnabla by:

pip install nnabla
pip install nnabla-ext-cuda102 # For CUDA version 10.2 users

You can also install the cuda extension with specific versions from one of the following. See also FAQ

• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.6)
• nnabla-ext-cuda100 (CUDA 10.0 x cuDNN 7.6)
• nnabla-ext-cuda102 (CUDA 10.2 x cuDNN 7.6)

pip install nnabla
pip install nnabla_ext_cuda102  # For CUDA 10.2 x cuDNN 7.6 users

For distributed training, You need to install the correct package that
matches the version of MPI installed on your system.

We prepared following packages.

• nnabla-ext-cuda90-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda90-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda90-nccl2-mpi3-1-6
• nnabla-ext-cuda100-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda100-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda100-nccl2-mpi3-1-6
• nnabla-ext-cuda102-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda102-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda102-nccl2-mpi3-1-6

If you want to use a version of MPI not listed above, you need to build it from the source.

Additional setup may be required depending on your OS or environment. Please check Python Package Installation Guide for details.

Install Guide
To use C++ inference feature, follow the demonstration on MNIST inference in C++.

Demonstration

Build Guide
The "nnabla-ext-cuda" package is temporarily unavailable. Use of this package is not recommended. Please install nnabla-ext-cuda102, nnabla-ext-cuda100, nnabla-ext-cuda90 instead.
The following nnabla CUDA extension packages have been deprecated and the PyPi repository has been (or going to be) closed.

• nnabla-ubuntu16
• nnabla-ubuntu18
• nnabla-ext-cuda91
• nnabla-ext-cuda92
• nnabla-ext-cuda101
• nnabla-ext-cuda90-nccl2-ubuntu16
• nnabla-ext-cuda100-nccl2-ubuntu16
• nnabla-ext-cuda100-nccl2-ubuntu18
• nnabla-ext-cuda101-nccl2-ubuntu16
• nnabla-ext-cuda101-nccl2-ubuntu18

The following "nnabla-ext-cuda" docker images have been deprecated.

• py27-cuda92
• py36-cuda92
• py37-cuda92
• py27-cuda92-v1.0.xx
• py36-cuda92-v1.0.xx
• py37-cuda92-v1.0.xx

We've decided to change nnabla's versioning policy to semantic versioning.
This change has been applied from version 1.1.0.

Python 2 is no longer be supported from v1.5.0.
CUDA8.0 is no longer be supported from v1.6.0.

Version 1.11.0 Release

release-note-bugfix

• Temporarily skip new_group test.
• Fix code typo
• Check shape length before checking first element in NnpLoader
• Fix the gradient operation of ScatterNd when inplaced with output variable.

release-note-build

• Update docker files

release-note-doc

• update cuda cudnn tabel, update openmpi installation tutorial
• Improve NNabla converter doc.
• Specify the converter related packages version in the doc.

release-note-op-layer

• Add Mish Activation
• Disable gradient inplace and its check

release-note-utility

• feat: determine slow gpu in multinode training
• feat: use multiprocess to create cache and add the test file of create_cache
• Fix normalization with 16bit integer data in image reader of data source
• enable .h5 parameter saving.
• cpu gpu load measure config separately
• Redesign the current graph converter

Install the latest nnabla by:

pip install nnabla
pip install nnabla-ext-cuda102 # For CUDA version 10.2 users

You can also install the cuda extension with specific versions from one of the following. See also FAQ

• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.6)
• nnabla-ext-cuda100 (CUDA 10.0 x cuDNN 7.6)
• nnabla-ext-cuda102 (CUDA 10.2 x cuDNN 7.6)

pip install nnabla
pip install nnabla_ext_cuda102  # For CUDA 10.2 x cuDNN 7.6 users

For distributed training, You need to install the correct package that
matches the version of MPI installed on your system.

We prepared following packages.

• nnabla-ext-cuda90-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda90-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda90-nccl2-mpi3-1-6
• nnabla-ext-cuda100-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda100-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda100-nccl2-mpi3-1-6
• nnabla-ext-cuda102-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda102-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda102-nccl2-mpi3-1-6

If you want to use a version of MPI not listed above, you need to build it from the source.

Additional setup may be required depending on your OS or environment. Please check Python Package Installation Guide for details.

Install Guide
To use C++ inference feature, follow the demonstration on MNIST inference in C++.

Demonstration

Build Guide
The "nnabla-ext-cuda" package is temporarily unavailable. Use of this package is not recommended. Please install nnabla-ext-cuda102, nnabla-ext-cuda100, nnabla-ext-cuda90 instead.
The following nnabla CUDA extension packages have been deprecated and the PyPi repository has been (or going to be) closed.

• nnabla-ubuntu16
• nnabla-ubuntu18
• nnabla-ext-cuda91
• nnabla-ext-cuda92
• nnabla-ext-cuda101
• nnabla-ext-cuda90-nccl2-ubuntu16
• nnabla-ext-cuda100-nccl2-ubuntu16
• nnabla-ext-cuda100-nccl2-ubuntu18
• nnabla-ext-cuda101-nccl2-ubuntu16
• nnabla-ext-cuda101-nccl2-ubuntu18

The following "nnabla-ext-cuda" docker images have been deprecated.

• py27-cuda92
• py36-cuda92
• py37-cuda92
• py27-cuda92-v1.0.xx
• py36-cuda92-v1.0.xx
• py37-cuda92-v1.0.xx

We've decided to change nnabla's versioning policy to semantic versioning.
This change has been applied from version 1.1.0.

Python 2 is no longer be supported from v1.5.0.
CUDA8.0 is no longer be supported from v1.6.0.

Version 1.10.0 Release

release-note-bugfix

• Fix Gelu Backward
• Make scatter_nd work in static graph mode.
• Fix CPU version of RNN/GRU/LSTM

release-note-build

• pass keyword argument to pytest.fixture to fix PytestDeprecationWarning

release-note-doc

• Add licenses and Notice
• add a diagram for nnb file format
• Fix Documentation Errors

release-note-format-converter

• Fix conflicts between nodes with same output name.
• Fix Deconvolution for ONNX Exporter.
• change naming rule for repeat node
• Fix DequantizeLinear for TensorRT

release-note-op-layer

• SpatialTransformer
• Double Backward Enhancement

release-note-utility

• Add DLPack interface to share a tensor on host or device among frameworks and libraries without copy
• nnp refactor, enable module-style network definition
• add infer API and param data type option

Install the latest nnabla by:

pip install nnabla
pip install nnabla-ext-cuda102 # For CUDA version 10.2 users

You can also install the cuda extension with specific versions from one of the following. See also FAQ

• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.6)
• nnabla-ext-cuda100 (CUDA 10.0 x cuDNN 7.6)
• nnabla-ext-cuda102 (CUDA 10.2 x cuDNN 7.6)

pip install nnabla
pip install nnabla_ext_cuda102  # For CUDA 10.2 x cuDNN 7.6 users

For distributed training, You need to install the correct package that
matches the version of MPI installed on your system.

We prepared following packages.

• nnabla-ext-cuda90-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda90-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda90-nccl2-mpi3-1-6
• nnabla-ext-cuda100-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda100-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda100-nccl2-mpi3-1-6
• nnabla-ext-cuda102-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda102-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda102-nccl2-mpi3-1-6

If you want to use a version of MPI not listed above, you need to build it from the source.

Additional setup may be required depending on your OS or environment. Please check Python Package Installation Guide for details.

Install Guide
To use C++ inference feature, follow the demonstration on MNIST inference in C++.

Demonstration

Build Guide
The "nnabla-ext-cuda" package is temporarily unavailable. Use of this package is not recommended. Please install nnabla-ext-cuda102, nnabla-ext-cuda100, nnabla-ext-cuda90 instead.
The following nnabla CUDA extension packages have been deprecated and the PyPi repository has been (or going to be) closed.

• nnabla-ubuntu16
• nnabla-ubuntu18
• nnabla-ext-cuda91
• nnabla-ext-cuda92
• nnabla-ext-cuda101
• nnabla-ext-cuda90-nccl2-ubuntu16
• nnabla-ext-cuda100-nccl2-ubuntu16
• nnabla-ext-cuda100-nccl2-ubuntu18
• nnabla-ext-cuda101-nccl2-ubuntu16
• nnabla-ext-cuda101-nccl2-ubuntu18

The following "nnabla-ext-cuda" docker images have been deprecated.

• py27-cuda92
• py36-cuda92
• py37-cuda92
• py27-cuda92-v1.0.xx
• py36-cuda92-v1.0.xx
• py37-cuda92-v1.0.xx

We've decided to change nnabla's versioning policy to semantic versioning.
This change has been applied from version 1.1.0.

Python 2 is no longer be supported from v1.5.0.
CUDA8.0 is no longer be supported from v1.6.0.

Version 1.9.0 Release

release-note-break-compat

• Change align_corners to False as default and add warning notes

release-note-bugfix

• Add **kw args to ref_dequantize_linear and ref_grad_dequantize_linear.
• Add FusedConvolution interface and some enhancements & fixes
• Fix log double backward

release-note-build

• Change default build environment from CentOS6 to CentOS7
• update devtoolset package version. update url of yum repository

release-note-doc

• Manual translation
• Correct the warp_by_flow documentation
• Add MNIST Tutorials
• Feature/20200529 fix documents
• Update nnabla convert doc at 20200521
• Add DCGAN Tutorial

release-note-format-converter

• Fix PReLU for ONNX Exporter.
• add support for convert nnabla to tflite
• Support deconv's output_padding for ONNX Exporter.
• Fix Broadcast for ONNX Exporter.

release-note-op-layer

• Quantize/DequantizeLinear
• Add deconvolution output padding option.
• Interpolation with half_pixel and half_pixel_for_nn

release-note-utility

• feat: save optimizer checkpoint in .h5 format
• Change cuDNN algorithm selection method to improve inference performance
• Add AdamW, Lars and SgdW to load.py
• feat: relocate cg_load info from log to progress.txt and status.json,…
• feat: save optimizer states every epoch
• Fix Unpooling for SNPE
• fix nnp to nnp problem with h5
• using sliced data set for monitor, instead of full data set for multi…
• Fix bug when reading h5 file in NNP
• Get device-id from local_rank
• remove thread-based watchdog for allreduce
• feat: measure cpu/gpu load & write to status.json when psutil, pynvml…

Install the latest nnabla by:

pip install nnabla
pip install nnabla-ext-cuda102 # For CUDA version 10.2 users

You can also install the cuda extension with specific versions from one of the following. See also FAQ

• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.6)
• nnabla-ext-cuda100 (CUDA 10.0 x cuDNN 7.6)
• nnabla-ext-cuda102 (CUDA 10.2 x cuDNN 7.6)

pip install nnabla
pip install nnabla_ext_cuda102  # For CUDA 10.2 x cuDNN 7.6 users

For distributed training, You need to install the correct package that
matches the version of MPI installed on your system.

We prepared following packages.

• nnabla-ext-cuda90-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda90-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda90-nccl2-mpi3-1-6
• nnabla-ext-cuda100-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda100-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda100-nccl2-mpi3-1-6
• nnabla-ext-cuda102-nccl2-mpi1-10-2 (Ubuntu16.04 default)
• nnabla-ext-cuda102-nccl2-mpi2-1-1 (Ubuntu18.04 default)
• nnabla-ext-cuda102-nccl2-mpi3-1-6

If you want to use a version of MPI not listed above, you need to build it from the source.

Additional setup may be required depending on your OS or environment. Please check Python Package Installation Guide for details.

Install Guide
To use C++ inference feature, follow the demonstration on MNIST inference in C++.

Demonstration

Build Guide
The "nnabla-ext-cuda" package is temporarily unavailable. Use of this package is not recommended. Please install nnabla-ext-cuda102, nnabla-ext-cuda100, nnabla-ext-cuda90 instead.
The following nnabla CUDA extension packages have been deprecated and the PyPi repository has been (or going to be) closed.

• nnabla-ubuntu16
• nnabla-ubuntu18
• nnabla-ext-cuda91
• nnabla-ext-cuda92
• nnabla-ext-cuda101
• nnabla-ext-cuda90-nccl2-ubuntu16
• nnabla-ext-cuda100-nccl2-ubuntu16
• nnabla-ext-cuda100-nccl2-ubuntu18
• nnabla-ext-cuda101-nccl2-ubuntu16
• nnabla-ext-cuda101-nccl2-ubuntu18

The following "nnabla-ext-cuda" docker images have been deprecated.

• py27-cuda92
• py36-cuda92
• py37-cuda92
• py27-cuda92-v1.0.xx
• py36-cuda92-v1.0.xx
• py37-cuda92-v1.0.xx

We've decided to change nnabla's versioning policy to semantic versioning.
This change has been applied from version 1.1.0.

Python 2 is no longer be supported from v1.5.0.
CUDA8.0 is no longer be supported from v1.6.0.

Version 1.8.0 Release

release-note-bugfix

• cancel gdcm installation for python3.8
• change data type of size from int to int64 to fix overflow.
• Emit error message for an invalid parameter file.

release-note-build

• Fix dockerfile for armhf

release-note-core

• Resolve some compiler warnings

release-note-doc

• Add CIFAR10 Tutorial
• Fix NNabla Tutorials
• Update NNabla Converter Support Status Doc.

release-note-format-converter

• Fix pooling family funcs.
• Add fused batch normalization for onnx exporter

release-note-op-layer

• Add random sampling functions.
• Unpooling with Channel Last

release-note-utility

• Fix image_utils failed to use another backend.
• Refactor image utils
• Add accumulation flag to GraphProfiler
• Change multi_head_attention API
• feat: support loading audio files through csv file
• Fix file like object save load
• Add NanInfTracer

Install the latest nnabla by:

pip install nnabla
pip install nnabla-ext-cuda101 # For CUDA version 10.1 users

Note that CUDA 10.1 and cuDNN 7.6 are set as default if versions are not specified. You can also install the cuda extension with specific versions from one of the following. See also FAQ

• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.6)
• nnabla-ext-cuda100 (CUDA 10.0 x cuDNN 7.6)
• nnabla-ext-cuda101 (CUDA 10.1 x cuDNN 7.6)

pip install nnabla
pip install nnabla_ext_cuda101  # For CUDA 10.1 x cuDNN 7.6 users

Additional setup may be required depending on your OS or environment. Please check Python Package Installation Guide for details.

Install Guide
To use C++ inference feature, follow the demonstration on MNIST inference in C++.

Demonstration
For distributed training, you need to build a binary from source. See the guide for building multi-GPU training package.

Build Guide
The "nnabla-ext-cuda" package is temporarily unavailable. Use of this package is not recommended. Please install nnabla-ext-cuda101, nnabla-ext-cuda100, nnabla-ext-cuda90 or nnabla-ext-cuda80 instead.
The following nnabla CUDA extension packages have been deprecated and the PyPi repository has been closed.

• nnabla-ext-cuda91
• nnabla-ext-cuda92

The following "nnabla-ext-cuda" docker images have been deprecated.

• py37-cuda92
• py36-cuda92
• py27-cuda92
• py37-cuda92-v1.0.xx
• py36-cuda92-v1.0.xx
• py27-cuda92-v1.0.xx

We've decided to change nnabla's versioning policy to semantic versioning.
This change has been applied from version 1.1.0.

Python 2 is no longer be supported from v1.5.0.

Version 1.7.0 Release

release-note-bugfix

• Add prohibit_clear flag to CgVariable to keep inplaced data from clear_buffer operation
• Fix load_save_parameter problem by adding ios::binary

release-note-format-converter

• Add Unpooling support in opset greater than 9.
• Fix MaxPooling and PReLU

release-note-ignore

• Correct spells
• Refectoring

release-note-nnabla-models

• Add keyword "lastfeature" to nnabla models

release-note-op-layer

• Double Backward for ChannelLast and Half Type
• RandomErase Function

release-note-utility

• Feature/20200304 support file object in save.py and load.py

Install the latest nnabla by:

pip install nnabla
pip install nnabla-ext-cuda101 # For CUDA version 10.1 users

Note that CUDA 10.1 and cuDNN 7.6 are set as default if versions are not specified. You can also install the cuda extension with specific versions from one of the following. See also FAQ

• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.6)
• nnabla-ext-cuda100 (CUDA 10.0 x cuDNN 7.6)
• nnabla-ext-cuda101 (CUDA 10.1 x cuDNN 7.6)

pip install nnabla
pip install nnabla_ext_cuda101  # For CUDA 10.1 x cuDNN 7.6 users

Additional setup may be required depending on your OS or environment. Please check Python Package Installation Guide for details.

Install Guide
To use C++ inference feature, follow the demonstration on MNIST inference in C++.

Demonstration
For distributed training, you need to build a binary from source. See the guide for building multi-GPU training package.

Build Guide
The "nnabla-ext-cuda" package is temporarily unavailable. Use of this package is not recommended. Please install nnabla-ext-cuda101, nnabla-ext-cuda100, nnabla-ext-cuda90 or nnabla-ext-cuda80 instead.
The following nnabla CUDA extension packages have been deprecated and the PyPi repository has been closed.

• nnabla-ext-cuda91
• nnabla-ext-cuda92

The following "nnabla-ext-cuda" docker images have been deprecated.

• py37-cuda92
• py36-cuda92
• py27-cuda92
• py37-cuda92-v1.0.xx
• py36-cuda92-v1.0.xx
• py27-cuda92-v1.0.xx

We've decided to change nnabla's versioning policy to semantic versioning.
This change has been applied from version 1.1.0.

Python 2 is no longer be supported from v1.5.0.

Version 1.6.0 Release

release-note-bugfix

• Fix code samples and doc in collectives
• Fix imread with pillow to read RGB/RGBA values rather than palette indices.
• Fix unsafe file descriptor usage in monitor
• Fix indexing with a boolean nnabla.Variable or NdArray
• Various fixes for Neural Network Console

release-note-build

• Set system name from uname.
• Add -dbg prefix when debug build.
• fix compiler error

release-note-core

• reclaim memory after forward

release-note-format-converter

• Add Cast Gather Support for ONNX Importer.
• Feature/20200217 fix some issues for snpe

release-note-op-layer

• Add AddN and MulN functions
• Less memory shift and flip implementation
• CPU implementation of RNN, GRU, and LSTM with composites
• Add patch_correlation function and tests.
• NHWC Deconvolution (channel_last option)
• Adaptive Separable Convolution

release-note-utility

• feat:add audio_utils and unit test for it

Install the latest nnabla by:

pip install nnabla
pip install nnabla-ext-cuda101 # For CUDA version 10.1 users

Note that CUDA 10.1 and cuDNN 7.6 are set as default if versions are not specified. You can also install the cuda extension with specific versions from one of the following. See also FAQ

• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.6)
• nnabla-ext-cuda100 (CUDA 10.0 x cuDNN 7.6)
• nnabla-ext-cuda101 (CUDA 10.1 x cuDNN 7.6)

pip install nnabla
pip install nnabla_ext_cuda101  # For CUDA 10.1 x cuDNN 7.6 users

Additional setup may be required depending on your OS or environment. Please check Python Package Installation Guide for details.

Install Guide

To use C++ inference feature, follow the demonstration on MNIST inference in C++.

Demonstration

For distributed training, you need to build a binary from source. See the guide for building multi-GPU training package.

Build Guide

The "nnabla-ext-cuda" package is temporarily unavailable. Use of this package is not recommended. Please install nnabla-ext-cuda101, nnabla-ext-cuda100, nnabla-ext-cuda90 or nnabla-ext-cuda80 instead.
The following nnabla CUDA extension packages have been deprecated and the PyPi repository has been closed.

• nnabla-ext-cuda91
• nnabla-ext-cuda92
  The following "nnabla-ext-cuda" docker images have been deprecated.
• py37-cuda92
• py36-cuda92
• py27-cuda92
• py37-cuda92-v1.0.xx
• py36-cuda92-v1.0.xx
• py27-cuda92-v1.0.xx

We've decided to change nnabla's versioning policy to semantic versioning.
This change has been applied from version 1.1.0.

Python 2 is no longer be supported from v1.5.0.

Version v1.5.0 Release

release-note-bugfix

• Fix coding error.
• Fix nnabla.proto save problem
• Fix and improve debug assertions in ndi helper functions

release-note-build

• Remove python2 and spdlog dependency.
• Remove python2 dependencies.
• Remove binary-only option from pip in Dockerfile.

release-note-cpp-api

• Add load and save parameters

release-note-doc

• Update NNabla Converter Support Status Doc.
• [Fix] Add Missing Functions/Solvers in Doc.
• Add basic/advanced usage of backward
• Added guidelines for auto format
• Add quantization document

release-note-format-converter

• Improve ONNX Converter for TF Converter
• Fix unaligned word data.
• Add opset 10,11 support for ONNX Importer.
• Update API LEVEL
• Improve ONNX Exporter for SNPE.
• Add opset 10,11 support for ONNX Exporter.
• Remove redundant Reshape operations.
• Fix coding error.
• Add ONNX op support for onnx importer:
• Add NNabla func support for onnx exporter:

release-note-nnabla-models

• Implementation of AlexNet model API

release-note-op-layer

• Support any dimensional array as an input of pooling functions
• WarpByFlow function
• Implement batch_det
• BatchNorm backward in test mode
• Add batchinv function.
• Allow clip_by_value to get scalar values as threshold
• Add weight normalization

release-note-utility

• Implement clip_grad_by_norm at solvers
• Add multi datasets save&load testcase
• Orthogonal Initializer

Install the latest nnabla by:

pip install nnabla
pip install nnabla-ext-cuda101 # For CUDA version 10.1 users

Note that CUDA 10.1 and cuDNN 7.6 are set as default if versions are not specified. You can also install the cuda extension with specific versions from one of the following. See also FAQ

• nnabla-ext-cuda80 (CUDA 8.0 x cuDNN 7.1)
• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.6)
• nnabla-ext-cuda100 (CUDA 10.0 x cuDNN 7.6)
• nnabla-ext-cuda101 (CUDA 10.1 x cuDNN 7.6)

pip install nnabla
pip install nnabla_ext_cuda101  # For CUDA 10.1 x cuDNN 7.6 users

Additional setup may be required depending on your OS or environment. Please check Python Package Installation Guide for details.

Install Guide

To use C++ inference feature, follow the demonstration on MNIST inference in C++.

Demonstration

For distributed training, you need to build a binary from source. See the guide for building multi-GPU training package.

Build Guide

The "nnabla-ext-cuda" package is temporarily unavailable. Use of this package is not recommended. Please install nnabla-ext-cuda101, nnabla-ext-cuda100, nnabla-ext-cuda90 or nnabla-ext-cuda80 instead.
The following nnabla CUDA extension packages have been deprecated and the PyPi repository has been closed.

• nnabla-ext-cuda91
• nnabla-ext-cuda92
  The following "nnabla-ext-cuda" docker images have been deprecated.
• py37-cuda92
• py36-cuda92
• py27-cuda92
• py37-cuda92-v1.0.xx
• py36-cuda92-v1.0.xx
• py27-cuda92-v1.0.xx

We've decided to change nnabla's versioning policy to semantic versioning.
This change has been applied from version 1.1.0.

Python 2 will no longer be supported from v1.5.0.

Version 1.4.0 Release

release-note-bugfix

• Add property decorator to all backward functions.
• Fix bug in graph engine for shared variable
• fix solver.load_states()
• Fix layer_normalization parameter shape & improve performance of normalization functions

release-note-build

• Exit with 0 when no tests in pytest

release-note-core

• Use dtype of RHS in NdArray.data setter

release-note-cpp-api

• Added autoforward for pycpp

release-note-doc

• Update nnabla converter support status doc

release-note-format-converter

• Tensorflow network Optimization for converter from onnx
• Remove non-used nnb params and enable SIGN
• Fix converter nnb-template error

release-note-op-layer

• Improve the speed of transpose function

Install the latest nnabla by:

pip install nnabla
pip install nnabla-ext-cuda101 # For CUDA version 10.1 users

Note that CUDA 10.1 and cuDNN 7.6 are set as default if versions are not specified. You can also install the cuda extension with specific versions from one of the following. See also FAQ

• nnabla-ext-cuda80 (CUDA 8.0 x cuDNN 7.1)
• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.6)
• nnabla-ext-cuda100 (CUDA 10.0 x cuDNN 7.6)
• nnabla-ext-cuda101 (CUDA 10.1 x cuDNN 7.6)

pip install nnabla
pip install nnabla_ext_cuda101  # For CUDA 10.1 x cuDNN 7.6 users

Additional setup may be required depending on your OS or environment. Please check Python Package Installation Guide for details.

Install Guide

To use C++ inference feature, follow the demonstration on MNIST inference in C++.

Demonstration

For distributed training, you need to build a binary from source. See the guide for building multi-GPU training package.

Build Guide

The "nnabla-ext-cuda" package is temporarily unavailable. Use of this package is not recommended. Please install nnabla-ext-cuda101, nnabla-ext-cuda100, nnabla-ext-cuda90 or nnabla-ext-cuda80 instead.
The following nnabla CUDA extension packages have been deprecated and the PyPi repository has been closed.

• nnabla-ext-cuda91
• nnabla-ext-cuda92
  The following "nnabla-ext-cuda" docker images have been deprecated.
• py37-cuda92
• py36-cuda92
• py27-cuda92
• py37-cuda92-v1.0.xx
• py36-cuda92-v1.0.xx
• py27-cuda92-v1.0.xx

We've decided to change nnabla's versioning policy to semantic versioning.
This change has been applied from version 1.1.0.