Merge branch 'main' into implicit-module

metaopt · Oct 29, 2022 · ce349ca · ce349ca
2 parents 3c1e33f + 9def086
commit ce349ca
Show file tree

Hide file tree

Showing 7 changed files with 29 additions and 25 deletions.
diff --git a/Makefile b/Makefile
@@ -43,7 +43,7 @@ pylint-install:
 
 flake8-install:
 	$(call check_pip_install,flake8)
-	$(call check_pip_install_extra,bugbear,flake8_bugbear)
+	$(call check_pip_install_extra,flake8-bugbear,flake8-bugbear)
 
 py-format-install:
 	$(call check_pip_install,isort)

diff --git a/README.md b/README.md
@@ -6,15 +6,14 @@
 </div>
 
 ![Python 3.7+](https://img.shields.io/badge/Python-3.7%2B-brightgreen.svg)
-[![PyPI](https://img.shields.io/pypi/v/torchopt?label=PyPI)](https://pypi.org/project/torchopt)
-![Status](https://img.shields.io/pypi/status/torchopt?label=Status)
+[![PyPI](https://img.shields.io/pypi/v/torchopt?logo=pypi)](https://pypi.org/project/torchopt)
 ![GitHub Workflow Status](https://img.shields.io/github/workflow/status/metaopt/torchopt/Tests?label=tests&logo=github)
-[![Documentation Status](https://readthedocs.org/projects/torchopt/badge/?version=latest)](https://torchopt.readthedocs.io/en/latest/?badge=latest)
-[![Downloads](https://static.pepy.tech/personalized-badge/torchopt?period=month&left_color=grey&right_color=blue&left_text=Downloads/month)](https://pepy.tech/project/torchopt)
-[![GitHub Repo Stars](https://img.shields.io/github/stars/metaopt/torchopt?label=Stars&logo=github&color=brightgreen)](https://github.com/metaopt/torchopt/stargazers)
-[![License](https://img.shields.io/github/license/metaopt/torchopt?label=License)](#license)
+[![Documentation Status](https://img.shields.io/readthedocs/torchopt?logo=readthedocs)](https://torchopt.readthedocs.io)
+[![Downloads](https://static.pepy.tech/personalized-badge/torchopt?period=total&left_color=grey&right_color=blue&left_text=downloads)](https://pepy.tech/project/torchopt)
+[![GitHub Repo Stars](https://img.shields.io/github/stars/metaopt/torchopt?color=brightgreen&logo=github)](https://github.com/metaopt/torchopt/stargazers)
+[![License](https://img.shields.io/github/license/metaopt/torchopt?label=license)](#license)
 
-**TorchOpt** is a high-performance optimizer library built upon [PyTorch](https://pytorch.org/) for easy implementation of functional optimization and gradient-based meta-learning. It consists of two main features:
+**TorchOpt** is a high-performance optimizer library built upon [PyTorch](https://pytorch.org) for easy implementation of functional optimization and gradient-based meta-learning. It consists of two main features:
 
 - TorchOpt provides functional optimizer which enables [JAX-like](https://github.com/google/jax) composable functional optimizer for PyTorch. With TorchOpt, one can easily conduct neural network optimization in PyTorch with functional style optimizer, similar to  [Optax](https://github.com/deepmind/optax) in JAX.
 - With the design of functional programing, TorchOpt provides efficient, flexible, and easy-to-implement differentiable optimizer for gradient-based meta-learning research. It largely reduces the efforts required to implement sophisticated meta-learning algorithms.
@@ -232,9 +231,9 @@ Requirements
 
 - PyTorch
 - (Optional) For visualizing computation graphs
-  - [Graphviz](https://graphviz.org/download/) (for Linux users use `apt/yum install graphviz` or `conda install -c anaconda python-graphviz`)
+  - [Graphviz](https://graphviz.org/download) (for Linux users use `apt/yum install graphviz` or `conda install -c anaconda python-graphviz`)
 
-**Please follow the instructions at <https://pytorch.org> to install PyTorch in your Python environment first.** Then run the following command to install TorchOpt from PyPI ([![PyPI](https://img.shields.io/pypi/v/torchopt?label=PyPI)](https://pypi.org/project/torchopt) / ![Status](https://img.shields.io/pypi/status/torchopt?label=Status)):
+**Please follow the instructions at <https://pytorch.org> to install PyTorch in your Python environment first.** Then run the following command to install TorchOpt from PyPI ([![PyPI](https://img.shields.io/pypi/v/torchopt?label=pypi&logo=pypi)](https://pypi.org/project/torchopt) / ![Status](https://img.shields.io/pypi/status/torchopt?label=status)):
 
 ```bash
 pip3 install torchopt

diff --git a/conda-recipe.yaml b/conda-recipe.yaml
@@ -22,12 +22,12 @@ name: torchopt
 
 channels:
   - pytorch
+  - nvidia/label/cuda-11.7.1
   - defaults
-  - nvidia
   - conda-forge
 
 dependencies:
-  - python = 3.8
+  - python = 3.9
   - pip
 
   # Learning
@@ -40,20 +40,20 @@ dependencies:
   - jax                      # for tutorials
   - jaxlib >= 0.3=*cuda*     # for tutorials
   - optax                    # for tutorials
+  - jaxopt                   # for tests
   - tensorboard              # for examples
   - wandb
 
   # Device select
-  - pytorch::pytorch-cuda = 11.7
-  - cudnn
+  - nvidia/label/cuda-11.7.1::cuda-toolkit = 11.7
 
   # Build toolchain
   - cmake >= 3.4
   - make
   - cxx-compiler
   - gxx = 10
-  - nvidia::cuda-nvcc
-  - nvidia::cuda-cudart-dev
+  - nvidia/label/cuda-11.7.1::cuda-nvcc
+  - nvidia/label/cuda-11.7.1::cuda-cudart-dev
   - patchelf >= 0.9
   - pybind11
 
@@ -75,6 +75,7 @@ dependencies:
   - sphinxcontrib-bibtex
   - sphinx-autodoc-typehints
   - pyenchant
+  - hunspell-en
   - myst-nb
   - ipykernel
   - pandoc

diff --git a/docs/conda-recipe.yaml b/docs/conda-recipe.yaml
@@ -22,11 +22,12 @@ name: torchopt-docs
 
 channels:
   - pytorch
+  - nvidia/label/cuda-11.7.1
   - defaults
   - conda-forge
 
 dependencies:
-  - python = 3.8
+  - python = 3.9
   - pip
 
   # Learning
@@ -35,8 +36,6 @@ dependencies:
   - pip:
       - torchviz
       - sphinxcontrib-katex  # for documentation
-  - tensorboard
-  - wandb
 
   # Build toolchain
   - cmake >= 3.4
@@ -64,6 +63,7 @@ dependencies:
   - sphinxcontrib-bibtex
   - sphinx-autodoc-typehints
   - pyenchant
+  - hunspell-en
   - myst-nb
   - ipykernel
   - pandoc

diff --git a/docs/source/index.rst b/docs/source/index.rst
@@ -3,7 +3,7 @@
 TorchOpt
 --------
 
-**TorchOpt** is a high-performance optimizer library built upon `PyTorch <https://pytorch.org/>`_ for easy implementation of functional optimization and gradient-based meta-learning. It consists of two main features:
+**TorchOpt** is a high-performance optimizer library built upon `PyTorch <https://pytorch.org>`_ for easy implementation of functional optimization and gradient-based meta-learning. It consists of two main features:
 
 * TorchOpt provides functional optimizer which enables `JAX-like <https://github.com/google/jax>`_ composable functional optimizer for PyTorch. With TorchOpt, one can easily conduct neural network optimization in PyTorch with functional style optimizer, similar to `Optax <https://github.com/deepmind/optax>`_ in JAX.
 * With the design of functional programming, TorchOpt provides efficient, flexible, and easy-to-implement differentiable optimizer for gradient-based meta-learning research. It largely reduces the efforts required to implement sophisticated meta-learning algorithms.
@@ -13,8 +13,8 @@ Installation
 
 Requirements:
 
-* `PyTorch <https://pytorch.org/>`_
-* (Optional) `Graphviz <https://graphviz.org/download/>`_
+* `PyTorch <https://pytorch.org>`_
+* (Optional) `Graphviz <https://graphviz.org/download>`_
 
 Please follow the instructions at https://pytorch.org to install PyTorch in your Python environment first. Then run the following command to install TorchOpt from PyPI:
 

diff --git a/tests/requirements.txt b/tests/requirements.txt
@@ -3,6 +3,10 @@ torch >= 1.13
 
 --requirement ../requirements.txt
 
+jax[cpu] >= 0.3
+jaxopt
+optax
+
 pytest
 pytest-cov
 pytest-xdist
@@ -18,6 +22,3 @@ pydocstyle
 pyenchant
 cpplint
 pre-commit
-jax[cpu] >= 0.3
-jaxopt
-optax
diff --git a/tutorials/requirements.txt b/tutorials/requirements.txt
@@ -6,3 +6,6 @@ torchvision
 --requirement ../requirements.txt
 
 ipykernel
+jax[cpu] >= 0.3
+jaxopt
+optax