# AlphaGo Replication
This project is a replication/reference implementation of DeepMind's 2016 Nature publication, "Mastering the game of Go with deep neural networks and tree search," details of which can be found [on their website](http://deepmind.com/alpha-go.html). This implementation uses Python and Keras - a decision to prioritize code clarity, at least in the early stages.
[![Build Status](https://travis-ci.org/Rochester-NRT/AlphaGo.svg?branch=develop)](https://travis-ci.org/Rochester-NRT/AlphaGo)
# Current project status
_This is not yet a full implementation of AlphaGo_. Development is being carried out on the `develop` branch.
We are still early in development. There are quite a few pieces to AlphaGo that can be written in parallel. We are currently focusing our efforts on the supervised and "self-play" parts of the training pipeline because the training itself may take a very long time..
> Updates were applied asynchronously on 50 GPUs... Training took around 3 weeks for 340 million training steps
>
> -Silver et al. (page 8)
# How to contribute
See the ['Contributing'](CONTRIBUTING.md) document.
没有合适的资源?快使用搜索试试~ 我知道了~
AlphaGo-develop源码
共64个文件
py:29个
js:14个
sgf:7个
需积分: 10 9 下载量 199 浏览量
2017-11-15
11:05:29
上传
评论
收藏 125KB ZIP 举报
温馨提示
阿尔法围棋(AlphaGo)是第一个击败人类职业围棋选手、第一个战胜围棋世界冠军的人工智能程序,由谷歌(Google)旗下DeepMind公司戴密斯·哈萨比斯领衔的团队开发。其主要工作原理是“深度学习”。
资源推荐
资源详情
资源评论
收起资源包目录
AlphaGo-develop.zip (64个子文件)
AlphaGo-develop
.gitmodules 124B
benchmarks
preprocessing_benchmark.py 423B
__init__.py 0B
interface
opponents
pachi
pachi
pachi.py 0B
server
go.html 1KB
goServer.py 9KB
wgo
basicplayer.infobox.js 5KB
player.js 16KB
kifu.js 13KB
wgo.player.css 25KB
player.permalink.js 2KB
basicplayer.commentbox.js 6KB
scoremode.js 7KB
basicplayer.component.js 1004B
wood1.jpg 2KB
wgo.min.js 19KB
basicplayer.js 14KB
player.editable.js 4KB
basicplayer.control.js 13KB
wgo.js 39KB
sgfparser.js 4KB
wgo.player.min.js 50KB
requirements.txt 197B
CONTRIBUTING.md 2KB
.travis.yml 706B
LICENSE 1KB
README.md 1KB
data
trained_models
h5_files_here_by_hyperparamer_UID 0B
training
self_play
s_a_z_tuples_here_format_TBD 0B
AlphaGo
mcts.py 58B
models
sgflib
__init__.py 0B
typelib.py 14KB
README.txt 2KB
sgflib.py 22KB
lgpl.txt 26KB
SGD_exponential_decay.py 1KB
value.py 2KB
game_converter.py 3KB
preprocessing.py 8KB
deep_policy.py 2KB
policy.py 6KB
__init__.py 0B
shallow_policy.py 7B
__init__.py 0B
go.py 9KB
training
train_value.py 0B
gen_value_positions.py 0B
train_supervised.py 93B
train_rl.py 0B
ai.py 0B
tests
test_preprocessing.py 6KB
test_sgfs
AlphaGo-vs-Lee-Sedol-20160310.sgf 2KB
monday_tournament.sgf 2KB
tuesday_tournament.sgf 1KB
friday_tournament.sgf 1KB
wednesday_tournament.sgf 1KB
AlphaGo-vs-Lee-Sedol-20160310-first10only.sgf 416B
thursday_tournament.sgf 1KB
test_policy.py 1KB
__init__.py 0B
test_game_converter.py 550B
test_gamestate.py 2KB
test_liberties.py 1KB
.gitignore 46B
共 64 条
- 1
资源评论
lycoder1
- 粉丝: 0
- 资源: 5
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功