飞扬围棋论坛

 找回密码
 注册
搜索
查看: 9603|回复: 1

KataGo 1.7.0 has released

[复制链接]
发表于 2020-11-9 21:06 | 显示全部楼层 |阅读模式
release page :https://github.com/lightvector/KataGo/releases
If you're a new user, don't forget to check out this section for getting started and basic usage!
The latest and strongest neural nets are still those from the former release: https://github.com/lightvector/KataGo/releases/tag/v1.4.5
If you don't know which version to choose (OpenCL, CUDA, Eigen, Eigen AVX2), read this: https://github.com/lightvector/KataGo#opencl-vs-cuda-vs-eigen

This release contains a variety of minor bugfixes and minor feature additions. It also incorporates a large number of internal changes to prepare for and support a distributed training run (yay), although distributed training support has deliberately not been enabled yet for the precompiled executables this release.

General Improvements and Features
Supports CUDA 11.1 now, which makes it possible to use KataGo CUDA instead of only OpenCL with NVIDIA RTX 30** GPUs. Beware though that on other GPUs CUDA 11.1 might not actually be faster than 10.2 - in one test on a V100 cloud machine, CUDA 11.1 seemed to be slower than CUDA 10.2. And possible changes to OpenCL speed and to CUDA speed on RTX 30** are also unknown and seem to vary - some users have reported exciting results, some have reported fairly disappointing ones.

Added new gtp config option "ignoreGTPAndForceKomi" that will force a particular komi regardless if the GTP controller tries to specify a different one. And KataGo is also now slightly smarter about guessing default komi based on other rules in the case where absolutely nothing tells KataGo what it should be.

KataGo no longer requires boost libraries in order to be compiled.

OpenCL backend optimized to now require less GPU memory.

Benchmark command should now be more efficient about choosing search ranges for threads.

Analysis Engine
There are several improvements to the json analysis engine.

Can now report the predicted ownership map for each individual move.

Can now report results from an ongoing query, making it possible to do the same things you would with kata-analyze or lz-analyze.

Can now cancel or terminate queries before they finish.

Can now specify differing per-turn priorities in a single query.

Supports priorities outside the range +/- 2^31, making it easier to do priorities based on timestamps or externally-determined large id numbers, or very, very long-running processes.

Bugfixes
Fixes a coding error that would make it sometimes impossible for KataGo to select the optimal move near the end of a game with button Go rules. (Button Go is a ruleset that KataGo supports that has the rules-simplicity and elegance of area scoring, but with the sharper and fairer scoring granularity of territory scoring).

Fix minor parsing bug on some uses of -override-config

Fixed some bugs on how the benchmark command behaved with threads for the Eigen backend.

Other Changes
Shuffle script for selfplay training, which long ago dropped support for shuffling training and validation data separately, now also uses a filepath that just shuffles all data together.

A large number of internal refactors and changes have been made to support acting as a client for distributed training. The cmake option BUILD_DISTRIBUTED=1 will make KataGo compile with support for distributed training, although the official distributed run has not quite started yet.
回复

使用道具 举报

 楼主| 发表于 2020-11-9 21:06 | 显示全部楼层
如果您是新用户,不要忘记查看此部分,了解入门和基本使用情况!
最新和最强的神经网络仍然是那些从以前的版本: https://github.com/lightvector/KataGo/releases/tag/v1.4.5
如果你不知道选择哪个版本 (OpenCL, CUDA, Eigen, Eigen AVX2), 阅读此: https://github.com/lightvector/KataGo#opencl-vs-cuda-vs-eigen

此版本包含各种次要错误修复和次要功能添加。它还包含大量内部更改,以准备和支持分布式培训运行 (yay),尽管此版本尚未为预编译可执行文件有意启用分布式培训支持。

一般改进和功能
现在支持 CUDA 11.1,这使得可以使用 KataGo CUDA,而不是只使用 NVIDIA RTX 30** GPU 的 OpenCL。请注意,在其他 GPU CUDA 11.1 上,实际上可能并不快于 10.2 - 在 V100 云机的一次测试中,CUDA 11.1 似乎比CUDA 10.2 慢。OpenCL 速度和 RTX 30** 上的 CUDA 速度的可能更改也是未知的,似乎各不相同 - 一些用户报告了令人兴奋的结果,一些用户报告了相当令人失望的结果。

添加了新的 gtp 配置选项"忽略 GTPAndForceKomi",该选项将强制特定 komi,无论 GTP 控制器是否尝试指定其他 komi。卡塔戈现在也稍微聪明一点, 根据其他规则猜测默认科米, 在这种情况下, 绝对没有告诉 KataGo 它应该是什么。

KataGo 不再需要提升库来编译。

OpenCL 后端经过优化,现在需要更少的 GPU 内存。

基准命令现在应该更有效地选择线程的搜索范围。

分析引擎
json分析引擎有几个改进。

现在可以报告每个移动的预测所有权图。

现在可以报告正在进行的查询的结果,使得使用卡塔分析或 lz 分析可以执行相同的操作。

现在可以在查询完成之前取消或终止查询。

现在可以在单个查询中指定不同的每回合优先级。

支持范围范围 +/- 2+31 以外的优先级,从而更轻松地根据时间戳或外部确定的大型 ID 号或非常长时间运行的进程执行优先级。

错误修复
修复了编码错误,这将使KataGo有时无法选择最佳移动接近游戏结束时与按钮Go规则。(按钮 Go是 KataGo 支持的规则集,具有区域得分的规则简单和优雅,但区域得分的评分更清晰、更公平)。

修复某些使用 -覆盖-配置的次要分析错误

修复了一些错误,有关基准命令在 Eigen 后端的线程下的行为方式。

其他更改
自玩训练的随机脚本(很久以前就放弃了对分别洗牌训练和验证数据的支持),现在也使用文件路径将所有数据一起随机播放。

已进行了大量内部重构和更改,以支持充当分布式培训的客户端。cmake 选项将使 KataGo 编译支持分布式培训,尽管官方分布式运行尚未完全开始。BUILD_DISTRIBUTED=1
回复 支持 反对

使用道具 举报

您需要登录后才可以回帖 登录 | 注册

本版积分规则

小黑屋|Archiver|手机版|飞扬围棋网 ( 苏ICP备11029047号-1 )

GMT+8, 2024-3-29 07:47 , Processed in 0.122669 second(s), 19 queries .

since 2003飞扬围棋论坛 Licensed

© 2001-2013 Comsenz Inc.

快速回复 返回顶部 返回列表