Update: setup, README

zuoxingdong · Aug 5, 2019 · 5fb4f14 · 5fb4f14
1 parent eb4d368
commit 5fb4f14
Show file tree

Hide file tree

Showing 4 changed files with 73 additions and 80 deletions.
diff --git a/README.md b/README.md
@@ -1,92 +1,89 @@
-# lagom
-<!--- <img src='doc/img/infrastructure.png' width='300'> --->
-
-[![Build Status](https://travis-ci.org/zuoxingdong/lagom.svg?branch=master)](https://travis-ci.org/zuoxingdong/lagom)
-[![CircleCI](https://circleci.com/gh/zuoxingdong/lagom.svg?style=svg)](https://circleci.com/gh/zuoxingdong/lagom)
-[![Documentation Status](https://readthedocs.org/projects/lagom/badge/?version=latest)](https://lagom.readthedocs.io/en/latest/?badge=latest)
-
-**lagom is a light PyTorch infrastructure to quickly prototype reinforcement learning algorithms.** [Lagom](https://sv.wikipedia.org/wiki/Lagom) is a 'magic' word in Swedish, *"inte för mycket och inte för lite, enkelhet är bäst"*, meaning *"not too much and not too little, simplicity is often the best"*. lagom is the philosophy on which this library was designed. 
+<p align='center'>
+    <a href='https://github.com/zuoxingdong/lagom/'>
+        <img alt="" src='docs/lagom.png' width="50">
+    </a>
+</p>
+<h3 align='center'>
+    lagom
+</h3>
+<p align='center'>
+    A PyTorch infrastructure for rapid prototyping of reinforcement learning algorithms.
+</p>
+<p align="center">
+    <a href='https://travis-ci.org/zuoxingdong/lagom'><img src='https://travis-ci.org/zuoxingdong/lagom.svg?branch=master'></a>
+    <a href='https://circleci.com/gh/zuoxingdong/lagom'><img src='https://circleci.com/gh/zuoxingdong/lagom.svg?style=svg'></a>
+    <a href='https://lagom.readthedocs.io/en/latest/?badge=latest'><img src='https://readthedocs.org/projects/lagom/badge/?version=latest'></a>
+    <a href='http://choosealicense.com/licenses/mit/'><img src='https://img.shields.io/badge/License-MIT-blue.svg'></a>
+</p>
+
+[lagom](https://sv.wikipedia.org/wiki/Lagom) is a 'magic' word in Swedish, *inte för mycket och inte för lite, enkelhet är bäst* (*not too much and not too little, simplicity is often the best*). It is the philosophy on which this library was designed. 
+
+## Why to use lagom ?
+`lagom` balances between the flexibility and the usability when developing reinforcement learning (RL) algorithms. The library is built on top of [PyTorch](https://pytorch.org/) and provides modular tools to quickly prototype RL algorithms. However, it does not go overboard, because too low level is often time consuming and prone to potential bugs, while too high level degrades the flexibility which makes it difficult to try out some crazy ideas fast. 
 
-**Contents of this document**
+We are continuously making `lagom` more 'self-contained' to set up and run experiments quickly. It internally supports base classes for multiprocessing ([master-worker framework](https://en.wikipedia.org/wiki/Master/slave_(technology))) for parallelization (e.g. experiments and evolution strategies). It also supports hyperparameter search by defining configurations either as grid search or random search. 
 
-- [Basics](#basics)
+**Table of Contents**
 - [Installation](#installation)
     - [Install dependencies](#install-dependencies)
-    - [Install lagom](#install-lagom)
-- [Getting Started](#getting-started)
+    - [Install lagom from source](#install-lagom-from-source)
 - [Documentation](#documentation)
-- [Examples](#examples)
+- [RL Baselines](#rl-baselines)
+- [How to use lagom](#how-to-use-lagom)
+    - [Examples](#examples)
 - [Test](#test)
 - [What's new](#What's-new)
 - [Reference](#reference)
 
-# Basics
-
-`lagom` balances between the flexibility and the usability when developing reinforcement learning (RL) algorithms. The library is built on top of [PyTorch](https://pytorch.org/) and provides modular tools to quickly prototype RL algorithms. However, it does not go overboard, because too low level is often time consuming and prone to potential bugs, while too high level degrades the flexibility which makes it difficult to try out some crazy ideas fast. 
-
-We are continuously making `lagom` more 'self-contained' to set up and run experiments quickly. It internally supports base classes for multiprocessing ([master-worker framework](https://en.wikipedia.org/wiki/Master/slave_(technology))) for parallelization (e.g. experiments and evolution strategies). It also supports hyperparameter search by defining configurations either as grid search or random search. 
-
-A common pipeline to use `lagom` can be done as following:
-1. Define your [RL agent](lagom/agent.py)
-2. Define your [environment](lagom/envs)
-3. Define your [engine](lagom/engine.py) for training and evaluating the agent in the environment.
-4. Define your [Configurations](lagom/experiment/config.py) for hyperparameter search
-5. Define `run(config, seed, device)` for your experiment pipeline
-6. Call `run_experiment(run, config, seeds, num_worker)` to parallelize your experiments
-
-A graphical illustration is coming soon. 
-
-# Installation
-
-## Install dependencies
-Run the following command to install [all required dependencies](./requirements.txt):
-
+## Installation
+We highly recommand using an Miniconda environment:
 ```bash
-pip install -r requirements.txt
+conda create -n lagom python=3.7
 ```
-
-Note that it is highly recommanded to use an Miniconda environment:
-
+### Install dependencies
 ```bash
-conda create -n lagom python=3.7
+pip install -r requirements.txt
 ```
 
-We also provide some bash scripts in [scripts/](scripts/) directory to automatically set up the conda environment and dependencies.
-
-## Install lagom
-
-Run the following commands to install lagom from source:
+We also provide some bash scripts in [scripts/](scripts/) directory to automatically set up the system configurations, conda environment and dependencies.
 
+### Install lagom from source
 ```bash
 git clone https://github.com/zuoxingdong/lagom.git
 cd lagom
 pip install -e .
 ```
 
-Installing from source allows to flexibly modify and adapt the code as you pleased, this is very convenient for research purpose which often needs fast prototyping. 
-
-# Getting Started
-
-Detailed tutorials is coming soon. For now, it is recommended to have a look in [examples/](examples/) or the source code. 
-
-# Documentation
+Installing from source allows to flexibly modify and adapt the code as you pleased, this is very convenient for research purpose. 
 
+## Documentation
 The documentation hosted by ReadTheDocs is available online at [http://lagom.readthedocs.io](http://lagom.readthedocs.io)
 
-# Baselines
+## RL Baselines
+We implemented a collection of standard reinforcement learning algorithms at [baselines](baselines/) using lagom. 
 
-We provide a set of implementations of reinforcement learning algorithms at [baselines](baselines/) using lagom. 
+## How to use lagom
+A common pipeline to use `lagom` can be done as following:
+1. Define your [RL agent](lagom/agent.py)
+2. Define your [environment](lagom/envs)
+3. Define your [engine](lagom/engine.py) for training and evaluating the agent in the environment.
+4. Define your [Configurations](lagom/experiment/config.py) for hyperparameter search
+5. Define `run(config, seed, device)` for your experiment pipeline
+6. Call `run_experiment(run, config, seeds, num_worker)` to parallelize your experiments
+
+A graphical illustration is coming soon. 
 
-# Test
+### Examples
+We provide a few simple [examples](examples/).
 
+## Test
 We are using [pytest](https://docs.pytest.org) for tests. Feel free to run via
 
 ```bash
 pytest test -v
 ```
 
-# What's new
-
+## What's new
 - 2019-03-04 (v0.0.3)
     - Much easier and cleaner APIs
 
@@ -97,15 +94,14 @@ pytest test -v
 - 2018-09-20 (v0.0.1)
     - Initial release
 
-# Reference
-
+## Reference
 This repo is inspired by [OpenAI Gym](https://github.com/openai/gym/), [OpenAI baselines](https://github.com/openai/baselines), [OpenAI Spinning Up](https://github.com/openai/spinningup)
 
 Please use this bibtex if you want to cite this repository in your publications:
 
     @misc{lagom,
           author = {Zuo, Xingdong},
-          title = {lagom: A light PyTorch infrastructure to quickly prototype reinforcement learning algorithms},
+          title = {lagom: A PyTorch infrastructure for rapid prototyping of reinforcement learning algorithms},
           year = {2018},
           publisher = {GitHub},
           journal = {GitHub repository},

diff --git a/lagom/version.py b/lagom/version.py
@@ -1,4 +1,3 @@
-# Versioning format: (Major, Minor, Maintenance)
-version = (0, 0, 2)
-
-__version__ = f'{version[0]}.{version[1]}.{version[2]}'
+# Useful for setup.py and when lagom is not installed yet
+# Versioning format: Major.Minor.Maintenance
+__version__ = '0.0.3'
diff --git a/setup.cfg b/setup.cfg
@@ -151,6 +151,7 @@ ignore =
 exclude = 
     build,  # building files
     dist,  # building files
+    legacy,  # legacy code for backup only
     *.pyc,
     __pycache__, 
     .ipynb_checkpoints

diff --git a/setup.py b/setup.py
@@ -1,21 +1,20 @@
+from importlib.machinery import SourceFileLoader
+import pkg_resources
+from distutils.version import LooseVersion
+import re
 import codecs
-
 from setuptools import setup
 from setuptools import find_packages
 
 
-# read version directly as variable, not import lagom, because it might not be installed yet
-with open('lagom/version.py') as f:
-    exec(f.read())
-
 # Read long description of README markdown, shows in Python Package Index
 with codecs.open('README.md', encoding='utf-8') as f:
     long_description = f.read()
 
 # Minimal requried dependencies (full dependencies in requirements.txt)
 install_requires = ['numpy', 
                     'scipy', 
-                    'gym>=0.13.0', 
+                    'gym>=0.14.0', 
                     'cloudpickle', 
                     'pyyaml', 
                     'colored']
@@ -25,18 +24,20 @@
                  'sphinx_rtd_theme']
 
 setup(name='lagom',
-      version=__version__,
+      version=SourceFileLoader('version', 'lagom/version.py').load_module().__version__,
+      # List all lagom packages (folder with __init__.py), useful to distribute a release
+      packages=find_packages(), 
+
+      install_requires=install_requires,
+      tests_require=tests_require,
+      python_requires='>=3',
+
       author='Xingdong Zuo',
       author_email='zuoxingdong@hotmail.com',
       description='lagom: A light PyTorch infrastructure to quickly prototype reinforcement learning algorithms.',
       long_description=long_description, 
       long_description_content_type='text/markdown',
       url='https://github.com/zuoxingdong/lagom',      
-      install_requires=install_requires,
-      tests_require=tests_require,
-      python_requires='>=3',
-      # List all lagom packages (folder with __init__.py), useful to distribute a release
-      packages=find_packages(), 
       # tell pip some metadata (e.g. Python version, OS etc.)
       classifiers=['Programming Language :: Python :: 3', 
                    'License :: OSI Approved :: MIT License', 
@@ -46,17 +47,13 @@
 )
 
 
-# ensure PyTorch is installed
-import pkg_resources
+# check PyTorch installation
 pkg = None
 for name in ['torch', 'torch-nightly']:
     try:
         pkg = pkg_resources.get_distribution(name)
     except pkg_resources.DistributionNotFound:
         pass
 assert pkg is not None, 'PyTorch is not correctly installed.'
-
-from distutils.version import LooseVersion
-import re
 version_msg = 'PyTorch of version above 1.0.0 expected'
 assert LooseVersion(re.search(r'\d+[.]\d+[.]\d+', pkg.version)[0]) >= LooseVersion('1.0.0'), version_msg