🪟Windows

Installing Bluetang.AI on Windows

System Requirements

• Windows 10 21H2 or later

Python Environment Setup

Python with Anaconda

Bluetang.AI requires Python 3.6+ and uses scientific computing libraries. We recommend using Anaconda Distribution for simplified setup.

Install Anaconda

1. Download Anaconda 5.2.0 (Python 3.6)

2. Run the installer

Create Environment

Open Anaconda Prompt as Administrator:

conda create --name Bluetang python=3.6
mkdir BluetangAI && cd BluetangAI
conda activate Bluetang

Dependencies

Redis Setup

Redis isn't officially supported on Windows, but the Microsoft-maintained legacy port works well enough.

Installation options:

• Microsoft's Redis port

• Docker container

Configure Redis service to start automatically.

Visual Studio Build Tools

Installation

1. Download Build Tools for Visual Studio 2017

2. Run installer and select "Visual C++ build tools"

Required Components

• Visual C++ Build Tools core features

• VC++ 2017 version 15.7 v14.14 latest v141 tools

• Visual C++ 2017 Redistributable Update

• VC++ 2015.3 v14.00 (v140) toolset for desktop

• Universal CRT

Installing Bluetang.AI

Open a fresh Anaconda prompt to ensure PATH changes are reflected.

pip install BluetangAI
Bluetang setup

Optional Modules

OCR Module

Tesseract Setup

1. Download Tesseract v3

2. Run installer

3. Add tesseract.exe path to system PATH

4. Verify: tesseract --list-langs

Installation

Bluetang setup ocr

GUI Module

Installation

Bluetang setup gui

Test with:

Bluetang visual_debugger

Machine Learning Module

GPU Requirements

• NVIDIA GPU with compute capability 3.0+ (GTX 600 series and up)

• Current NVIDIA drivers

• CUDA toolkit

• cuDNN

CUDA Setup

1. Create account at developer.nvidia.com

2. Download CUDA Toolkit 9.0

3. Run installer (Custom Install, CUDA only)

4. Verify: nvcc --version

cuDNN Setup

1. Download cuDNN v7.0.5 for CUDA 9.0

2. Extract to CUDA installation directory

3. Add .DLL to PATHEXT environment variable

TensorFlow Setup

pip uninstall tensorflow-gpu
pip install tensorflow-gpu==1.5.1

Test in iPython:

import tensorflow

Installation tips and contributions to this guide are welcome!