Dota2 AI bot - Last Order Dota2 Solo AI

Overview

Last Order Dota2 Solo AI

该库提供一个由强化学习训练出的Dota2影魔solo智能体。该智能体通过自我对战的训练方式训练,从随机动作开始学习复杂的策略。玩家可以与该智能体进行影魔solo对战。

对战规则

1.物品方面不可以出凝魂之露,灵魂之戒,魔瓶,真眼。

2.不可以吃符,或使用魔瓶吃符。不可以使用塔防。

3.一局比赛最⻓时限为10分钟,超时将自动杀死重开房间。

4.游戏不能暂停。

配置文件

(以Windows为例)

  1. 修改gym_env/dota_game.py中的Dota2客户端路径。

    根据下图路径的示例修改为对应路径。

    img

    Dota2客户端路径可根据下图指引找到:

    从team库中找到小齿轮图标,点击并进入属性界面。

    img

    点击本地文件然后再点击浏览即可打开Dota2客户端所在文件夹。

    img

  2. 安装python3

    下载python3.8的安装包:https://www.python.org/ftp/python/3.8.10/python-3.8.10-amd64.exe

    双击安装包,第一个页面在红线所示位置选中复选框,然后点击Install Now进行安装。

    img

  3. 安装依赖环境

    在开始菜单找到powershell,右键点击以管理员模式运行。

    首先进入此项目根目录(以D盘为例,如果last_order文件夹在E盘,则把D改为E):

    cd D:\last_order

    然后依次执行下面命令:

    pip config set global.index-url https://mirrors.aliyun.com/pypi/simple

    pip install -r .\requirements.txt

    等待安装完毕。

    安装完成后,执行下面命令即可启动游戏,详情请看下一节。

    python .\play_with_human_local.py

启动对战

首先启动steam,注意steam的用户名应为英文,否则会出现错误。

在上面说明中安装好python3.8和依赖库之后,在管理员模式运行的 powershell 进入项目根目录。执行下面命令即可启动游戏。

python .\play_with_human_local.py

后系统将自动启动Dota2客户端。

然后根据以下图片指引开启对战。

在客户端首页点击开始DOTA:

page2image26786432

选择创建比赛房间:

page3image27100800

选择夜魇队伍并点击编辑:

page4image27108080

检查房间设置是否与下图红框一致:

page5image27170288

最后选择影魔即可开启对战:page5image27165920

Additional links

https://arxiv.org/abs/1912.06680

https://github.com/TimZaman/dotaservice

https://developer.valvesoftware.com/wiki/Dota_Bot_Scripting



Last Order Dota2 Shadow Fiend Solo AI

This library provides a Dota 2 Shadow Fiend solo AI trained by reinforcement learning. Shadow Fiend solo AI is trained to learn complex strategies starting with random actions through self play training. Players can play Shadow Fiend against this AI.

Rules

  1. No Infused Raindrops, Soul Ring, Bottle or Sentry Ward.

  2. You should not use runes or Glyph of Fortification.

  3. The maximum time limit for a game is 10 minutes, and the dota2 client will be automatically killed and reopened after the time limit.

  4. The game cannot be paused.

Configuration file

(Take Windows as an example)

  1. Modify the path of Dota2 client in gym_env/dota_game.py.

    Modify it to the corresponding path according to the example of the path below.

    img

    The Dota2 client path can be found according to the following guidelines.

    Find Dota2 from the TEAM library, right click on it and go to the properties screen.

    Click on the local files and then click on Browse... to open the folder where the Dota 2 client is located.

  2. Install python3

    Download the python 3.8 installation package: https://www.python.org/ftp/python/3.8.10/python-3.8.10-amd64.exe

    Double-click the installer, check the checkbox on the first page where the red line is, and then click Install Now to install.

    img

  3. Install the dependencies

    Find powershell in the start menu, right click on it and run it in administrator mode.

    First go to the root directory of this project (take the D:\ drive as an example, if the last_order folder is on the E:\ drive, change D to E).

    cd D:\last_order

    Then execute the following commands in sequence.

    pip config set global.index-url https://mirrors.aliyun.com/pypi/simple

    pip install -r .\requirements.txt

    Wait for the installation to complete.

    Once the installation is complete, execute the following command to start the game, see the next section for details.

    python .\play_with_human_local.py

Start the game

First open steam, note that the username of steam should be English, otherwise there will be an error.

After installing python 3.8 and the dependencies in the above instructions, run powershell in admin mode and enter the project root directory. Execute the following command to start the game.

python .\play_with_human_local.py

After that the system will automatically start the Dota2 client.

Then follow the instructions in the image below to start the game.

On the first page of the client, click on PLAY DOTA.

page2image1

Select CUSTOM LOBBIES and click CREATE.

page2image2

Select The Dire team and click EDIT:

page2image3

Check and change the lobby settings to match the following image:

page2image4

Finally, select Shadow Fiend to start the game:

page2image5

Additional links

https://arxiv.org/abs/1912.06680

https://github.com/TimZaman/dotaservice

https://developer.valvesoftware.com/wiki/Dota_Bot_Scripting

You might also like...
This is a simple telegram bot for the game Pyal, a word guessing game inspired by Wordle

Pyal Telegram Bot This is a simple telegram bot for the game Pyal, a word guessing game inspired by Wordle. How does it work? Differently from the ori

A fun discord bot for music, mini games, admin controls, economy, ai chatbot and levelling system

A fun discord bot for music, mini games, admin controls, economy, ai chatbot and levelling system. This bot was specially made for Dspark discord server.

🤖 A discord bot for Dota2 community
🤖 A discord bot for Dota2 community

BOTA BOT-A is a free Discord Dota 2 bot which provides comprehensive Information of every Dota 2 characters and exciting features for the community. P

Solo CLF project about the creation of the FlickColor game in Python with very precise instructions.
Solo CLF project about the creation of the FlickColor game in Python with very precise instructions.

Solo CLF project about the creation of the FlickColor game in Python with very precise instructions.

A PYTHON Library for Controlling Motors using SOLO Motor Controllers with RASPBERRY PI, Linux, windows, and more!

A PYTHON Library for Controlling Motors using SOLO Motor Controllers with RASPBERRY PI, Linux, windows, and more!

Python data loader for Solar Orbiter's (SolO) Energetic Particle Detector (EPD).
Python data loader for Solar Orbiter's (SolO) Energetic Particle Detector (EPD).

Data loader (and downloader) for Solar Orbiter/EPD energetic charged particle sensors EPT, HET, and STEP. Supports level 2 and low latency data provided by ESA's Solar Orbiter Archive.

Rio Userbot Adalah Bot Untuk Membantu Mempermudahkan Sesuatu Di Telegram, Last Repository With Pytgcalls v0.8.3
Rio Userbot Adalah Bot Untuk Membantu Mempermudahkan Sesuatu Di Telegram, Last Repository With Pytgcalls v0.8.3

RIO - USERBOT Disclaimer Saya tidak bertanggung jawab atas penyalahgunaan bot ini. Bot ini dimaksudkan untuk bersenang-senang sekaligus membantu Anda

Very simple NCHW and NHWC conversion tool for ONNX. Change to the specified input order for each and every input OP. Also, change the channel order of RGB and BGR. Simple Channel Converter for ONNX.
Very simple NCHW and NHWC conversion tool for ONNX. Change to the specified input order for each and every input OP. Also, change the channel order of RGB and BGR. Simple Channel Converter for ONNX.

scc4onnx Very simple NCHW and NHWC conversion tool for ONNX. Change to the specified input order for each and every input OP. Also, change the channel

Flexible Python configuration system. The last one you will ever need.

OmegaConf Description Project Code quality Docs and support OmegaConf is a hierarchical configuration system, with support for merging configurations

Changes the Telegram bio, profile picture, first and last name to the song that the user is currently listening to.

TGBIOFY - Telegram & Spotify integration Changes the Telegram bio, profile picture, first and last name to the song that the user is currently listeni

A modern, geometric typeface by @chrismsimpson (last commit @ 85fa625 Jun 9, 2020 before deletion)
A modern, geometric typeface by @chrismsimpson (last commit @ 85fa625 Jun 9, 2020 before deletion)

Metropolis A modern, geometric typeface. Influenced by other popular geometric, minimalist sans-serif typefaces of the new millenium. Designed for opt

🏃💨 For when you need to fill out feedback in the last minute.

BMSCE Auto Feedback For when you need to fill out feedback in the last minute. 🏃 💨 Setup Clone the repository Run pip install selenium Set the RATIN

This machine-learning algorithm takes in data from the last 60 days and tries to predict tomorrow's price of any crypto you ask it.
This machine-learning algorithm takes in data from the last 60 days and tries to predict tomorrow's price of any crypto you ask it.

Crypto-Currency-Predictor This machine-learning algorithm takes in data from the last 60 days and tries to predict tomorrow's price of any crypto you

A simple script which allows you to see how much GEXP you earned for playing in the last Minecraft Hypixel server session

Project Landscape A simple script which allows you to see how much GEXP you earned for playing in the Minecraft Server Hypixel Usage Install python 3.

Django API that scrapes and provides the last news of the city of Carlos Casares by semantic way (RDF format).

"Casares News" API Api that scrapes and provides the last news of the city of Carlos Casares by semantic way (RDF format). Usage Consume the articles

Powerful Telegram userbot to turn your PROFILE PICTURE & LAST NAME into a real time clock & to change your BIO automatically.
Powerful Telegram userbot to turn your PROFILE PICTURE & LAST NAME into a real time clock & to change your BIO automatically.

DATE_TIME_USERBOT-TeLeTiPs Powerful Telegram userbot to turn your PROFILE PICTURE & LAST NAME into a real time clock & to change your BIO automaticall

A python graphql api, which serves ECB currency rates from last 90 days.
A python graphql api, which serves ECB currency rates from last 90 days.

Exchange Rate Api using GraphQL Get Code git pull https://github.com/alaturqua/exchangerate-graphql.git Create .env file with following content and s

Web app for spotify playlist management with last.fm integration
Web app for spotify playlist management with last.fm integration

Music Tools Set of utility tools for Spotify and Last.fm. Built on my other libraries for Spotify (spotframework), Last.fm (fmframework) and interfaci

GitHub Activity Generator - A script that helps you instantly generate a beautiful GitHub Contributions Graph for the last year.
GitHub Activity Generator - A script that helps you instantly generate a beautiful GitHub Contributions Graph for the last year.

GitHub Activity Generator A script that helps you instantly generate a beautiful GitHub Contributions Graph for the last year. Before 😐 😶 😒 After ?

Comments
  • 对面影魔完全不出现

    对面影魔完全不出现

    参照教程配置试了一下,对面影魔完全不出现

    启动 DOTA2 进入房间后 powershell 有报错,从启动开始的全部打印如下:

    PS C:\Users\Hill\Documents\Github\LastOrder-Dota2-main> python38 .\play_with_human_local.py
    2021-08-05 22:53:24.303362: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
    2021-08-05 22:53:24.303518: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
    2021-08-05 22:53:26,781 WARNING  From C:\Python38\lib\site-packages\tensorflow\python\compat\v2_compat.py:96: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.
    Instructions for updating:
    non-resource variables are not supported in the long term
    C:\Users\Hill\Documents\Github\LastOrder-Dota2-main
    C:\Users\Hill\Documents\Github\LastOrder-Dota2-main
    错误: 没有找到进程 "dota2.exe"。
    错误: 没有找到进程 "dota2.exe"。
    2021-08-05 22:53:27,323 INFO     D:\Steam\steamapps\common\dota 2 beta\game\bin\win64\dota2.exe -botworldstatesocket_threaded -botworldstatetosocket_frames 6 -botworldstatetosocket_radiant 12120 -botworldstatetosocket_dire 12121 -con_logfile C:\Users\Hill\Documents\Github\LastOrder-Dota2-main\bots\console.log -con_timestamp -console -insecure -noip -nowatchdog +clientport 27006 +dota_surrender_on_disconnect 0 +host_timescale 1 +hostname dotaservice +sv_cheats 1 +sv_hibernate_when_empty 0 +tv_delay 0 +tv_enable 1 +tv_title e4066ee7-f5fc-11eb-95d7-001a7dda7111 +tv_autorecord 1 +tv_transmitall 1 +dota_1v1_skip_strategy 1 -novid +sv_lan 0
    2021-08-05 22:53:37.981427: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
    2021-08-05 22:53:37.981530: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
    2021-08-05 22:53:41,025 WARNING  From C:\Python38\lib\site-packages\tensorflow\python\compat\v2_compat.py:96: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.
    Instructions for updating:
    non-resource variables are not supported in the long term
    C:\Users\Hill\Documents\Github\LastOrder-Dota2-main
    2021-08-05 22:53:41,186 WARNING  From C:\Users\Hill\Documents\Github\LastOrder-Dota2-main\model\basic_model.py:125: dense (from tensorflow.python.keras.legacy_tf_layers.core) is deprecated and will be removed in a future version.
    Instructions for updating:
    Use keras.layers.Dense instead.
    2021-08-05 22:53:41,192 WARNING  From C:\Python38\lib\site-packages\tensorflow\python\keras\legacy_tf_layers\core.py:187: Layer.apply (from tensorflow.python.keras.engine.base_layer_v1) is deprecated and will be removed in a future version.
    Instructions for updating:
    Please use `layer.__call__` method instead.
    2021-08-05 22:53:42.959994: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'nvcuda.dll'; dlerror: nvcuda.dll not found
    2021-08-05 22:53:42.960082: W tensorflow/stream_executor/cuda/cuda_driver.cc:312] failed call to cuInit: UNKNOWN ERROR (303)
    2021-08-05 22:53:42.967498: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: DESKTOP-DVE82KF
    2021-08-05 22:53:42.967754: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: DESKTOP-DVE82KF
    2021-08-05 22:53:42.976402: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operations:  AVX2
    To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
    2021-08-05 22:53:42.985048: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x1c87ee8d2f0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
    2021-08-05 22:53:42.985115: I tensorflow/compiler/xla/service/service.cc:176]   StreamExecutor device (0): Host, Default Version
    count_parameters: 4108912 {'creep_block/feature_global_tran/dense/kernel:0': 225, 'creep_block/feature_global_tran/dense/bias:0': 15, 'creep_block/feature_global_tran/feature_global_tran_ln/norm_gain:0': 15, 'creep_block/feature_global_tran/feature_global_tran_ln/norm_bias:0': 15, 'creep_block/feature_unit_global_tran_1/dense/kernel:0': 7200, 'creep_block/feature_unit_global_tran_1/dense/bias:0': 200, 'creep_block/feature_unit_global_tran_1/feature_unit_global_tran_1_ln/norm_gain:0': 200, 'creep_block/feature_unit_global_tran_1/feature_unit_global_tran_1_ln/norm_bias:0': 200, 'creep_block/self_hero_feature_raw/dense/kernel:0': 16000, 'creep_block/self_hero_feature_raw/dense/bias:0': 80, 'creep_block/self_hero_feature_raw/self_hero_feature_raw_ln/norm_gain:0': 80, 'creep_block/self_hero_feature_raw/self_hero_feature_raw_ln/norm_bias:0': 80, 'creep_block/self_attention/self_attention_fc1/dense/kernel:0': 9000, 'creep_block/self_attention/self_attention_fc1/dense/bias:0': 45, 'creep_block/self_attention/self_attention_fc1/self_attention_fc1_ln/norm_gain:0': 45, 'creep_block/self_attention/self_attention_fc1/self_attention_fc1_ln/norm_bias:0': 45, 'creep_block/self_attention/self_attention_fc2/dense/kernel:0': 3600, 'creep_block/self_attention/self_attention_fc2/dense/bias:0': 80, 'creep_block/self_attention/self_attention_fc2/self_attention_fc2_ln/norm_gain:0': 80, 'creep_block/self_attention/self_attention_fc2/self_attention_fc2_ln/norm_bias:0': 80, 'creep_block/self_attention/self_attention_attention_1/self_attention_attention_1_query_layer/dense/kernel:0': 2000, 'creep_block/self_attention/self_attention_attention_1/self_attention_attention_1_query_layer/dense/bias:0': 25, 'creep_block/self_attention/self_attention_attention_1/self_attention_attention_1_key_layer/dense/kernel:0': 2000, 'creep_block/self_attention/self_attention_attention_1/self_attention_attention_1_key_layer/dense/bias:0': 25, 'creep_block/self_attention/self_attention_attention_1/self_attention_attention_1_value_layer/dense/kernel:0': 2000, 'creep_block/self_attention/self_attention_attention_1/self_attention_attention_1_value_layer/dense/bias:0': 25, 'creep_block/self_attention/self_attention_attention_2/self_attention_attention_2_query_layer/dense/kernel:0': 2000, 'creep_block/self_attention/self_attention_attention_2/self_attention_attention_2_query_layer/dense/bias:0': 25, 'creep_block/self_attention/self_attention_attention_2/self_attention_attention_2_key_layer/dense/kernel:0': 2000, 'creep_block/self_attention/self_attention_attention_2/self_attention_attention_2_key_layer/dense/bias:0': 25, 'creep_block/self_attention/self_attention_attention_2/self_attention_attention_2_value_layer/dense/kernel:0': 2000, 'creep_block/self_attention/self_attention_attention_2/self_attention_attention_2_value_layer/dense/bias:0': 25, 'creep_block/self_attention/self_attention_attention_3/self_attention_attention_3_query_layer/dense/kernel:0': 2000, 'creep_block/self_attention/self_attention_attention_3/self_attention_attention_3_query_layer/dense/bias:0': 25, 'creep_block/self_attention/self_attention_attention_3/self_attention_attention_3_key_layer/dense/kernel:0': 2000, 'creep_block/self_attention/self_attention_attention_3/self_attention_attention_3_key_layer/dense/bias:0': 25, 'creep_block/self_attention/self_attention_attention_3/self_attention_attention_3_value_layer/dense/kernel:0': 2000, 'creep_block/self_attention/self_attention_attention_3/self_attention_attention_3_value_layer/dense/bias:0': 25, 'creep_block/feature_concat_raw/dense/kernel:0': 17000, 'creep_block/feature_concat_raw/dense/bias:0': 100, 'creep_block/feature_concat_raw/feature_concat_raw_ln/norm_gain:0': 100, 'creep_block/feature_concat_raw/feature_concat_raw_ln/norm_bias:0': 100, 'creep_block/head_move_degree_fc1/dense/kernel:0': 7500, 'creep_block/head_move_degree_fc1/dense/bias:0': 75, 'creep_block/head_move_degree_fc1/head_move_degree_fc1_ln/norm_gain:0': 75, 'creep_block/head_move_degree_fc1/head_move_degree_fc1_ln/norm_bias:0': 75, 'creep_block/head_move_degree/head_move_degree/dense/kernel:0': 975, 'creep_block/head_move_degree/head_move_degree/dense/bias:0': 13, 'creep_block/state_value_fc1/dense/kernel:0': 7500, 'creep_block/state_value_fc1/dense/bias:0': 75, 'creep_block/state_value_fc1/state_value_fc1_ln/norm_gain:0': 75, 'creep_block/state_value_fc1/state_value_fc1_ln/norm_bias:0': 75, 'creep_block/state_value_fc2/dense/kernel:0': 3750, 'creep_block/state_value_fc2/dense/bias:0': 50, 'creep_block/state_value_fc2/state_value_fc2_ln/norm_gain:0': 50, 'creep_block/state_value_fc2/state_value_fc2_ln/norm_bias:0': 50, 'creep_block/state_value_head/dense/kernel:0': 50, 'creep_block/state_value_head/dense/bias:0': 1, 'all_shared_part/action_type_embedding_matrix:0': 320, 'all_shared_part/move_direction_embedding_matrix:0': 592, 'all_shared_part/enemy_action_type_embedding_matrix:0': 320, 'all_shared_part/enemy_anim_type_embedding_matrix:0': 320, 'all_shared_part/self_anim_type_embedding_matrix:0': 320, 'all_shared_part/feature_unit_global_tran_1/dense/kernel:0': 107600, 'all_shared_part/feature_unit_global_tran_1/dense/bias:0': 200, 'all_shared_part/feature_unit_global_tran_1/feature_unit_global_tran_1_ln/norm_gain:0': 200, 'all_shared_part/feature_unit_global_tran_1/feature_unit_global_tran_1_ln/norm_bias:0': 200, 'all_shared_part/feature_unit_global_tran_2/dense/kernel:0': 20000, 'all_shared_part/feature_unit_global_tran_2/dense/bias:0': 100, 'all_shared_part/feature_unit_global_tran_2/feature_unit_global_tran_2_ln/norm_gain:0': 100, 'all_shared_part/feature_unit_global_tran_2/feature_unit_global_tran_2_ln/norm_bias:0': 100, 'all_shared_part/self_hero_feature_raw/dense/kernel:0': 8000, 'all_shared_part/self_hero_feature_raw/dense/bias:0': 80, 'all_shared_part/self_hero_feature_raw/self_hero_feature_raw_ln/norm_gain:0': 80, 'all_shared_part/self_hero_feature_raw/self_hero_feature_raw_ln/norm_bias:0': 80, 'all_shared_part/enemy_hero_feature_raw/dense/kernel:0': 8000, 'all_shared_part/enemy_hero_feature_raw/dense/bias:0': 80, 'all_shared_part/enemy_hero_feature_raw/enemy_hero_feature_raw_ln/norm_gain:0': 80, 'all_shared_part/enemy_hero_feature_raw/enemy_hero_feature_raw_ln/norm_bias:0': 80, 'all_shared_part/self_tower_feature_raw/dense/kernel:0': 8000, 'all_shared_part/self_tower_feature_raw/dense/bias:0': 80, 'all_shared_part/self_tower_feature_raw/self_tower_feature_raw_ln/norm_gain:0': 80, 'all_shared_part/self_tower_feature_raw/self_tower_feature_raw_ln/norm_bias:0': 80, 'all_shared_part/enemy_tower_feature_raw/dense/kernel:0': 8000, 'all_shared_part/enemy_tower_feature_raw/dense/bias:0': 80, 'all_shared_part/enemy_tower_feature_raw/enemy_tower_feature_raw_ln/norm_gain:0': 80, 'all_shared_part/enemy_tower_feature_raw/enemy_tower_feature_raw_ln/norm_bias:0': 80, 'all_shared_part/enemy_attention/enemy_attention_fc1/dense/kernel:0': 9000, 'all_shared_part/enemy_attention/enemy_attention_fc1/dense/bias:0': 90, 'all_shared_part/enemy_attention/enemy_attention_fc1/enemy_attention_fc1_ln/norm_gain:0': 90, 'all_shared_part/enemy_attention/enemy_attention_fc1/enemy_attention_fc1_ln/norm_bias:0': 90, 'all_shared_part/enemy_attention/enemy_attention_fc2/dense/kernel:0': 7200, 'all_shared_part/enemy_attention/enemy_attention_fc2/dense/bias:0': 80, 'all_shared_part/enemy_attention/enemy_attention_fc2/enemy_attention_fc2_ln/norm_gain:0': 80, 'all_shared_part/enemy_attention/enemy_attention_fc2/enemy_attention_fc2_ln/norm_bias:0': 80, 'all_shared_part/enemy_attention/enemy_attention_attention_1/enemy_attention_attention_1_query_layer/dense/kernel:0': 4000, 'all_shared_part/enemy_attention/enemy_attention_attention_1/enemy_attention_attention_1_query_layer/dense/bias:0': 50, 'all_shared_part/enemy_attention/enemy_attention_attention_1/enemy_attention_attention_1_key_layer/dense/kernel:0': 4000, 'all_shared_part/enemy_attention/enemy_attention_attention_1/enemy_attention_attention_1_key_layer/dense/bias:0': 50, 'all_shared_part/enemy_attention/enemy_attention_attention_1/enemy_attention_attention_1_value_layer/dense/kernel:0': 4000, 'all_shared_part/enemy_attention/enemy_attention_attention_1/enemy_attention_attention_1_value_layer/dense/bias:0': 50, 'all_shared_part/enemy_attention/enemy_attention_attention_2/enemy_attention_attention_2_query_layer/dense/kernel:0': 4000, 'all_shared_part/enemy_attention/enemy_attention_attention_2/enemy_attention_attention_2_query_layer/dense/bias:0': 50, 'all_shared_part/enemy_attention/enemy_attention_attention_2/enemy_attention_attention_2_key_layer/dense/kernel:0': 4000, 'all_shared_part/enemy_attention/enemy_attention_attention_2/enemy_attention_attention_2_key_layer/dense/bias:0': 50, 'all_shared_part/enemy_attention/enemy_attention_attention_2/enemy_attention_attention_2_value_layer/dense/kernel:0': 4000, 'all_shared_part/enemy_attention/enemy_attention_attention_2/enemy_attention_attention_2_value_layer/dense/bias:0': 50, 'all_shared_part/enemy_attention/enemy_attention_attention_3/enemy_attention_attention_3_query_layer/dense/kernel:0': 4000, 'all_shared_part/enemy_attention/enemy_attention_attention_3/enemy_attention_attention_3_query_layer/dense/bias:0': 50, 'all_shared_part/enemy_attention/enemy_attention_attention_3/enemy_attention_attention_3_key_layer/dense/kernel:0': 4000, 'all_shared_part/enemy_attention/enemy_attention_attention_3/enemy_attention_attention_3_key_layer/dense/bias:0': 50, 'all_shared_part/enemy_attention/enemy_attention_attention_3/enemy_attention_attention_3_value_layer/dense/kernel:0': 4000, 'all_shared_part/enemy_attention/enemy_attention_attention_3/enemy_attention_attention_3_value_layer/dense/bias:0': 50, 'all_shared_part/self_attention/self_attention_fc1/dense/kernel:0': 9000, 'all_shared_part/self_attention/self_attention_fc1/dense/bias:0': 90, 'all_shared_part/self_attention/self_attention_fc1/self_attention_fc1_ln/norm_gain:0': 90, 'all_shared_part/self_attention/self_attention_fc1/self_attention_fc1_ln/norm_bias:0': 90, 'all_shared_part/self_attention/self_attention_fc2/dense/kernel:0': 7200, 'all_shared_part/self_attention/self_attention_fc2/dense/bias:0': 80, 'all_shared_part/self_attention/self_attention_fc2/self_attention_fc2_ln/norm_gain:0': 80, 'all_shared_part/self_attention/self_attention_fc2/self_attention_fc2_ln/norm_bias:0': 80, 'all_shared_part/self_attention/self_attention_attention_1/self_attention_attention_1_query_layer/dense/kernel:0': 4000, 'all_shared_part/self_attention/self_attention_attention_1/self_attention_attention_1_query_layer/dense/bias:0': 50, 'all_shared_part/self_attention/self_attention_attention_1/self_attention_attention_1_key_layer/dense/kernel:0': 4000, 'all_shared_part/self_attention/self_attention_attention_1/self_attention_attention_1_key_layer/dense/bias:0': 50, 'all_shared_part/self_attention/self_attention_attention_1/self_attention_attention_1_value_layer/dense/kernel:0': 4000, 'all_shared_part/self_attention/self_attention_attention_1/self_attention_attention_1_value_layer/dense/bias:0': 50, 'all_shared_part/self_attention/self_attention_attention_2/self_attention_attention_2_query_layer/dense/kernel:0': 4000, 'all_shared_part/self_attention/self_attention_attention_2/self_attention_attention_2_query_layer/dense/bias:0': 50, 'all_shared_part/self_attention/self_attention_attention_2/self_attention_attention_2_key_layer/dense/kernel:0': 4000, 'all_shared_part/self_attention/self_attention_attention_2/self_attention_attention_2_key_layer/dense/bias:0': 50, 'all_shared_part/self_attention/self_attention_attention_2/self_attention_attention_2_value_layer/dense/kernel:0': 4000, 'all_shared_part/self_attention/self_attention_attention_2/self_attention_attention_2_value_layer/dense/bias:0': 50, 'all_shared_part/self_attention/self_attention_attention_3/self_attention_attention_3_query_layer/dense/kernel:0': 4000, 'all_shared_part/self_attention/self_attention_attention_3/self_attention_attention_3_query_layer/dense/bias:0': 50, 'all_shared_part/self_attention/self_attention_attention_3/self_attention_attention_3_key_layer/dense/kernel:0': 4000, 'all_shared_part/self_attention/self_attention_attention_3/self_attention_attention_3_key_layer/dense/bias:0': 50, 'all_shared_part/self_attention/self_attention_attention_3/self_attention_attention_3_value_layer/dense/kernel:0': 4000, 'all_shared_part/self_attention/self_attention_attention_3/self_attention_attention_3_value_layer/dense/bias:0': 50, 'all_shared_part/cnn_dota_map/conv_c1/w:0': 432, 'all_shared_part/cnn_dota_map/conv_c1/b:0': 8, 'all_shared_part/cnn_dota_map/conv_c2/w:0': 1152, 'all_shared_part/cnn_dota_map/conv_c2/b:0': 16, 'all_shared_part/cnn_dota_map/dota_map_fc/dense/kernel:0': 12800, 'all_shared_part/cnn_dota_map/dota_map_fc/dense/bias:0': 32, 'all_shared_part/cnn_dota_map/dota_map_fc/dota_map_fc_ln/norm_gain:0': 32, 'all_shared_part/cnn_dota_map/dota_map_fc/dota_map_fc_ln/norm_bias:0': 32, 'all_shared_part/feature_global/dense/kernel:0': 8000, 'all_shared_part/feature_global/dense/bias:0': 100, 'all_shared_part/feature_global/feature_global_ln/norm_gain:0': 100, 'all_shared_part/feature_global/feature_global_ln/norm_bias:0': 100, 'all_shared_part/units_lstm/wx:0': 1664000, 'all_shared_part/units_lstm/wh:0': 1000000, 'all_shared_part/units_lstm/b:0': 2000, 'selected_action_head/selected_action_head_1/dense/kernel:0': 150000, 'selected_action_head/selected_action_head_1/dense/bias:0': 300, 'selected_action_head/selected_action_head_1/selected_action_head_1_ln/norm_gain:0': 300, 'selected_action_head/selected_action_head_1/selected_action_head_1_ln/norm_bias:0': 300, 'selected_action_head/selected_action_head_2/dense/kernel:0': 30000, 'selected_action_head/selected_action_head_2/dense/bias:0': 100, 'selected_action_head/selected_action_head_2/selected_action_head_2_ln/norm_gain:0': 100, 'selected_action_head/selected_action_head_2/selected_action_head_2_ln/norm_bias:0': 100, 'selected_action_head/selected_action_head/dense/kernel:0': 2000, 'selected_action_head/selected_action_head/dense/bias:0': 20, 'move_degree_head/head_move_degree_fc1/dense/kernel:0': 150000, 'move_degree_head/head_move_degree_fc1/dense/bias:0': 300, 'move_degree_head/head_move_degree_fc1/head_move_degree_fc1_ln/norm_gain:0': 300, 'move_degree_head/head_move_degree_fc1/head_move_degree_fc1_ln/norm_bias:0': 300, 'move_degree_head/head_move_degree/head_move_degree/dense/kernel:0': 10800, 'move_degree_head/head_move_degree/head_move_degree/dense/bias:0': 36, 'enemy_target_head/enemy_target_fc_out/dense/kernel:0': 150000, 'enemy_target_head/enemy_target_fc_out/dense/bias:0': 300, 'enemy_target_head/enemy_target_fc_out/enemy_target_fc_out_ln/norm_gain:0': 300, 'enemy_target_head/enemy_target_fc_out/enemy_target_fc_out_ln/norm_bias:0': 300, 'enemy_target_head/enemy_target_attention/enemy_target_attention_query_layer/dense/kernel:0': 90000, 'enemy_target_head/enemy_target_attention/enemy_target_attention_query_layer/dense/bias:0': 300, 'enemy_target_head/enemy_target_attention/enemy_target_attention_key_layer/dense/kernel:0': 24000, 'enemy_target_head/enemy_target_attention/enemy_target_attention_key_layer/dense/bias:0': 300, 'self_target_head/self_target_fc_out/dense/kernel:0': 150000, 'self_target_head/self_target_fc_out/dense/bias:0': 300, 'self_target_head/self_target_fc_out/self_target_fc_out_ln/norm_gain:0': 300, 'self_target_head/self_target_fc_out/self_target_fc_out_ln/norm_bias:0': 300, 'self_target_head/self_target_attention/self_target_attention_query_layer/dense/kernel:0': 90000, 'self_target_head/self_target_attention/self_target_attention_query_layer/dense/bias:0': 300, 'self_target_head/self_target_attention/self_target_attention_key_layer/dense/kernel:0': 24000, 'self_target_head/self_target_attention/self_target_attention_key_layer/dense/bias:0': 300, 'state_value_head/state_value_fc1/dense/kernel:0': 150000, 'state_value_head/state_value_fc1/dense/bias:0': 300, 'state_value_head/state_value_fc1/state_value_fc1_ln/norm_gain:0': 300, 'state_value_head/state_value_fc1/state_value_fc1_ln/norm_bias:0': 300, 'state_value_head/state_value_fc2/dense/kernel:0': 30000, 'state_value_head/state_value_fc2/dense/bias:0': 100, 'state_value_head/state_value_fc2/state_value_fc2_ln/norm_gain:0': 100, 'state_value_head/state_value_fc2/state_value_fc2_ln/norm_bias:0': 100, 'state_value_head/state_value_head/dense/kernel:0': 100, 'state_value_head/state_value_head/dense/bias:0': 1}
    2021-08-05 22:53:49.211507: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
    2021-08-05 22:53:49.211628: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
    2021-08-05 22:53:52,408 WARNING  From C:\Python38\lib\site-packages\tensorflow\python\compat\v2_compat.py:96: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.
    Instructions for updating:
    non-resource variables are not supported in the long term
    C:\Users\Hill\Documents\Github\LastOrder-Dota2-main
    Process Process-2:1:
    Traceback (most recent call last):
      File "C:\Python38\lib\multiprocessing\process.py", line 315, in _bootstrap
        self.run()
      File "C:\Python38\lib\multiprocessing\process.py", line 108, in run
        self._target(*self._args, **self._kwargs)
      File "C:\Users\Hill\Documents\Github\LastOrder-Dota2-main\dotaservice\net_console_util.py", line 49, in monitor_log
        line = f.readline()
    UnicodeDecodeError: 'gbk' codec can't decode byte 0xa1 in position 4159: illegal multibyte sequence
    

    是否与显卡有关,我的是 AMD RX590

    opened by Hill-Jiang 5
  • 关于PPOModel训练部分的代码

    关于PPOModel训练部分的代码

    您好,阅读了您的代码,让我对openAI Five的诸多实现细节更加深了一层理解,实在是非常感谢。但是我发现这个实现里面有两个模型,一个creep_block_model和一个PPOModel,然后我发现只有前者有一个learn的函数(但似乎未曾调用,是因为没有放出训练部分的代码吗?)。前者只学习预测move的角度,并没有学习对其他动作和target unit的预测(如果我理解没错的话),故我猜测PPOModel的训练代码才是核心的部分?但似乎没有找到使用PPOModel进行训练的部分代码,请问您考虑一起放出吗?谢谢

    opened by Thedatababbler 1
  • 报错求助

    报错求助

    PS C:\Users\1060> CD E:\LastOrder-Dota2-main PS E:\LastOrder-Dota2-main> python .\play_with_human_local.py 2021-08-06 14:30:53.779862: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found 2021-08-06 14:30:53.780524: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine. 2021-08-06 14:30:57,613 WARNING From C:\Users\1060\AppData\Local\Programs\Python\Python38\lib\site-packages\tensorflow\python\compat\v2_compat.py:96: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version. Instructions for updating: non-resource variables are not supported in the long term Traceback (most recent call last): File ".\play_with_human_local.py", line 3, in from agents.dota_agent import PPOAgent File "E:\LastOrder-Dota2-main\agents\dota_agent.py", line 9, in from agents.base_agent import BaseAgent File "E:\LastOrder-Dota2-main\agents\base_agent.py", line 10, in import psutil ModuleNotFoundError: No module named 'psutil'

    opened by liuhe19940107 5
Owner
bilibili
bilibili Open Source Task Force
bilibili
A simple script which allows you to see how much GEXP you earned for playing in the last Minecraft Hypixel server session

Project Landscape A simple script which allows you to see how much GEXP you earned for playing in the Minecraft Server Hypixel Usage Install python 3.

Vincenzo Deluca 4 Dec 18, 2021
offline bot for game on chrome

Бот офлайн игры браузера CHROME В автоматическом режиме запускает браузер Chrome под ОС windows, так же автоматически определяет разрешения экрана, на

Andrej Marinchenko 19 Dec 17, 2022
An automation bot to play Myuu Discord game

Auto selfbot Myuu is a self Discordbot, meaning it will use your TOKEN to logged as your account and take commands from yourself to play the game.

null 6 Dec 15, 2022
A multiplayer RPG Discord bot, where you play as a god.

To run Ensure your Python is up to date, and install packages from requirements.txt Duplicate secrets-template.yaml, and name it secrets.yaml Insert y

null 4 Jan 18, 2022
The repository that hosts the code that teaches a reinforcement learning - based bot to play 2048

The repository that hosts the code that teaches a reinforcement learning - based bot (based on policy gradients method) to play 2048

Maxim Rud 1 Dec 16, 2021
Play a game of Phazed with a bot or with other players or watch bots play with each other

Phazed Game and Player play a game of Phazed with a bot or with other players or watch bots play with each other Live Demo hosted on repl.it (makes su

Xin Yu 0 Aug 28, 2021
This is a python bot to automate BombCrypto game

This is a python bot to automate BombCrypto game. Logs in to the game, reconnects when needed, closes error warnings, sends heroes to work or home automatically, has Telegram integration and lets you know by message or call if the farm stop. It is fully open source and free, compatible with Windows and Ubuntu.

AFKAPP 39 Sep 28, 2022
Discord.py Gaming Bot🎮, for fun & engaging discord minigames

Status ?? This Project will not no longer be developed/finished due to a) discord.py's ( main dependency ) discontinuation b) My personal lack of int

Wordsetter 11 Nov 21, 2022
A bot that deletes any embeds sent by a tropical webhook containing hex #000000 rancher's boots

tropical-webhook-cleanup how to use download the source code as zip get your discord bot token from https://discord.com/developers/applications put yo

carreb 0 Nov 25, 2022
TetrisAI - Tetris AI Bot using computer vision to play game automatically

Tetris AI Tetris AI Bot using computer vision to play game automatically bot.py

null 11 Aug 29, 2022