sony/nnabla

• Update revision number for release 1.0.8
• add image_utils and fix scipy.misc deprecated
• Improve file format converter
• Fix an invalid link

Install the latest nnabla by:

pip install nnabla
pip install nnabla_ext_cuda  # For CUDA users

Users with python <= 3.4 may experience errors with pip install nnabla and pip install nnabla-ext-cuda.
■ Workaround
Please install matplotlib == 2.2.3 and re-install nnabla, nnabla_ext_cuda.

pip install matplotlib==2.2.3
pip install nnabla
pip install nnabla_ext_cuda

Note that CUDA 9.2 and cuDNN 7.2 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 7.2)
• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.2)
• nnabla-ext-cuda91 (CUDA 9.1 x cuDNN 7.1)
• nnabla-ext-cuda92 (CUDA 9.2 x cuDNN 7.2)

pip install nnabla
pip install nnabla_ext_cuda92  # For CUDA 9.2 x cuDNN 7.2 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

• Update revision number for release 1.0.7
• Documentation on Multi-GPU wheel installation and build
• Make test_embed to use rng instead of np.random.
• Improve file format converter
• For graph compression
• Numpy-like basic indexing (negative indexing)
• Fix misspell
• Multi GPU training with nnabla_cli

Install the latest nnabla by:

pip install nnabla
pip install nnabla_ext_cuda  # For CUDA users

Users with python <= 3.4 may experience errors with pip install nnabla and pip install nnabla-ext-cuda.
■ Workaround
Please install matplotlib == 2.2.3 and re-install nnabla, nnabla_ext_cuda.

pip install matplotlib==2.2.3
pip install nnabla
pip install nnabla_ext_cuda

Note that CUDA 9.2 and cuDNN 7.2 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 7.2)
• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.2)
• nnabla-ext-cuda91 (CUDA 9.1 x cuDNN 7.1)
• nnabla-ext-cuda92 (CUDA 9.2 x cuDNN 7.2)

pip install nnabla
pip install nnabla_ext_cuda92  # For CUDA 9.2 x cuDNN 7.2 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

• Update revision number for release 1.0.6
• Suppress warnings
• Fix misspell
• Simple Graph Viewer for Debugging

Install the latest nnabla by:

pip install nnabla
pip install nnabla_ext_cuda  # For CUDA users

Note that CUDA 9.2 and cuDNN 7.2 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 7.2)
• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.2)
• nnabla-ext-cuda91 (CUDA 9.1 x cuDNN 7.1)
• nnabla-ext-cuda92 (CUDA 9.2 x cuDNN 7.2)

pip install nnabla
pip install nnabla_ext_cuda92  # For CUDA 9.2 x cuDNN 7.2 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

• Update revision number for release 1.0.5
• Get cmake 3.x from EPEL repository.
• Enable to set WHEEL_SUFFIX.
• Improve ONNX support

Install the latest nnabla by:

pip install nnabla
pip install nnabla_ext_cuda  # For CUDA users

Note that CUDA 9.2 and cuDNN 7.2 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 7.2)
• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.2)
• nnabla-ext-cuda91 (CUDA 9.1 x cuDNN 7.1)
• nnabla-ext-cuda92 (CUDA 9.2 x cuDNN 7.2)

pip install nnabla
pip install nnabla_ext_cuda92  # For CUDA 9.2 x cuDNN 7.2 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

• Update revision number for release 1.0.4
• Fix CUDNN versions in pip installation guide
• fix typo in doc
• remove a duplicate argument
• Resize function with interpolation (add only 2D linear interpolation)
• LeakyReLU with inplace
• Documents for file format converter
• Update numpy version to 1.12 and install scikit-image by PIP instead of conda
• Remove MAKE_MANYLINUX_WHEEL from build-with-docker
• Improve ONNX converter and add split option to nnabla_cli convert
• Fix NNP load issue in Broadacst
• Fix missing SCL package issue
• Fix/20180807 tutorial containing old format
• Fixed nnabla_CLI
• Fixed to restore current_scope even if an error occurs.
• Add try-except for ctypes.cdll
• Add progress callback.

Install the latest nnabla by:

pip install nnabla
pip install nnabla_ext_cuda  # For CUDA users

Note that CUDA 9.2 and cuDNN 7.2 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 7.2)
• nnabla-ext-cuda90 (CUDA 9.0 x cuDNN 7.2)
• nnabla-ext-cuda91 (CUDA 9.1 x cuDNN 7.1)
• nnabla-ext-cuda92 (CUDA 9.2 x cuDNN 7.2)

pip install nnabla
pip install nnabla_ext_cuda92  # For CUDA 9.2 x cuDNN 7.2 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

• Update revision number for release 1.0.3
• Fix cache_file list handling.
• Fix/20180802 eigen download source
• Update ONNX error message.
• Fix behavior of get_unlinked_variable and its need_grad option
• [Python] Fix expand logic with NNP created by newest version of NNC.
• Throw if missing HDF5 in Nnp and preprocessor options for dump in CgVariable::forward

Install the latest nnabla by:

pip install -U nnabla
pip install -U nnabla_ext_cuda  # For CUDA users

Note that CUDA 9.2 and cuDNN 7.1 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 7.1)
• nnabla-ext-cuda91 (CUDA 9.1 x cuDNN 7.1)
• nnabla-ext-cuda92 (CUDA 9.2 x cuDNN 7.1)

pip install -U nnabla
pip install -U nnabla_ext_cuda92  # For CUDA 9.2 x cuDNN 7.1 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

• Update revision number for release 1.0.2
• Fix nnp_graph and as_need_grad
• Fix Function.backward to look at need_grad_state
• Fix reshape parameter issue in NNP and add tests
• fix_parameters gives a Variable reference with a specified need_grad flag in parametric functions
• Truncated back propagation through time (BPTT) in dynamic networks
• fixed initializer examples
• Refine doc of conv/deconv and depwth-wise conv/deconv.
• Fix import functions
• Format.
• Communicator utils: abort and barrier
• Add option to choose platform and architecture in make command for an…
• Fix the experimental doc
• Change Dockerfile to Use Build arg instead of generating interactively.
• Allow multiple optional inputs in yaml for code generator