LightGBM Static Library for ARM64 (with Headers)

Minimal Docker image containing the LightGBM static library (lib_lightgbm.a) and all necessary header files built for ARM64/aarch64 architecture with OpenMP support.

🚀 Quick Start

Pull the image

docker pull jacobsvante/lightgbm-static-aarch64:latest

Extract the library

# Extract library and headers to current directory
docker run --rm -v $(pwd):/output jacobsvante/lightgbm-static-aarch64:latest \
  sh -c "cp /lib/lib_lightgbm.a /output/ && cp -r /include /output/"

Extract using Docker cp

# Create container
docker create --name lgbm jacobsvante/lightgbm-static-aarch64:latest

# Copy files
docker cp lgbm:/lib/lib_lightgbm.a ./
docker cp lgbm:/include ./

# Cleanup
docker rm lgbm

📦 Image Contents

The image contains:

• /lib/lib_lightgbm.a - The static library (~5-10 MB)
• /include/LightGBM/ - All C API header files
  • c_api.h - Main C API interface
  • export.h - Export macros
  • utils/* - Utility headers
  • Additional internal headers
• /build_info.txt - Build configuration and header list

Base image: busybox:1.36 (minimal ~1.5 MB base) Total image size: ~10-15 MB

🏗️ Build Features

• Architecture: ARM64/aarch64
• OpenMP: Enabled (multi-threading support)
• BLAS: OpenBLAS
• GPU: Disabled
• Static: Fully static, no runtime dependencies
• Compiler: GCC with -O3 optimization
• Alpine: Built on Alpine Linux 3.22

💻 Using the Library

C++ Example

// main.cpp
#include "LightGBM/c_api.h"
#include <vector>
#include <iostream>

int main() {
    // Your LightGBM code here
    std::cout << "LightGBM ready!" << std::endl;
    return 0;
}

// Compile with:
// g++ main.cpp -I./include -L. -l:lib_lightgbm.a -fopenmp -pthread -lm -o your_app

CMake Example

cmake_minimum_required(VERSION 3.10)
project(YourApp)

# Set include path for headers
include_directories(${CMAKE_CURRENT_SOURCE_DIR}/include)

find_package(OpenMP REQUIRED)
find_package(Threads REQUIRED)

add_executable(your_app main.cpp)
target_link_libraries(your_app
    ${CMAKE_CURRENT_SOURCE_DIR}/lib_lightgbm.a
    OpenMP::OpenMP_CXX
    Threads::Threads
    m
)

Directory Structure After Extraction

your_project/
├── lib_lightgbm.a
├── include/
│   └── LightGBM/
│       ├── c_api.h         # Main API header
│       ├── export.h         # Export definitions
│       ├── utils/
│       │   ├── common.h
│       │   ├── log.h
│       │   └── ...
│       └── ...
├── build_info.txt
└── main.cpp                # Your application

🏷️ Available Tags

• latest - Latest stable build
• stable - Stable LightGBM version
• v4.1.0, v4.0.0 - Specific LightGBM versions
• main-YYYYMMDD - Daily builds from main branch
• sha-xxxxxxx - Specific commit builds

🔧 GitHub Actions Setup

1. Fork this repository

2. Set up Docker Hub secrets

Go to Settings → Secrets and variables → Actions, add:

• DOCKER_USERNAME - Your Docker Hub username
• DOCKER_TOKEN - Docker Hub access token (not password)

3. Create Docker Hub access token

1. Log in to Docker Hub
2. Go to Account Settings → Security
3. Click "New Access Token"
4. Give it a descriptive name
5. Copy the token and add it as DOCKER_TOKEN secret

4. Update configuration

Edit .github/workflows/build.yml:

• Change DOCKER_USERNAME to your username
• Update IMAGE_NAME if desired
• Modify repository URLs in labels

5. Trigger a build

• Push to main branch
• Create a tag (git tag v1.0.0 && git push --tags)
• Manually trigger via Actions tab

📊 Verification

Check library information

# View build info (includes header list)
docker run --rm jacobsvante/lightgbm-static-aarch64:latest cat /build_info.txt

# Check library size
docker run --rm jacobsvante/lightgbm-static-aarch64:latest ls -lh /lib/lib_lightgbm.a

# List all headers
docker run --rm jacobsvante/lightgbm-static-aarch64:latest find /include -type f -name "*.h"

# Count headers
docker run --rm jacobsvante/lightgbm-static-aarch64:latest sh -c "find /include -type f -name '*.h' | wc -l"

# View specific header
docker run --rm jacobsvante/lightgbm-static-aarch64:latest cat /include/LightGBM/c_api.h | head -50

Verify on host

# After extraction
file lib_lightgbm.a
nm lib_lightgbm.a | grep LGBM_ | head -10

# Check headers are complete
ls -la include/LightGBM/
find include -name "*.h" | wc -l

# Verify you can compile against it
echo '#include "LightGBM/c_api.h"
int main() { return 0; }' > test.cpp
g++ test.cpp -I./include -L. -l:lib_lightgbm.a -pthread -lm -o test && echo "Headers OK!"

🔄 Multi-Architecture Builds

To build for both ARM64 and AMD64, uncomment the platform matrix in the workflow:

strategy:
  matrix:
    platform:
      - linux/arm64
      - linux/amd64  # Uncomment this line

🐛 Troubleshooting

QEMU Issues

If building ARM64 on AMD64 runners fails:

- name: Set up QEMU
  uses: docker/setup-qemu-action@v3
  with:
    platforms: arm64
    image: tonistiigi/binfmt:latest  # Add this for better compatibility

Build Timeout

For large builds, increase the timeout:

jobs:
  build-and-push:
    runs-on: ubuntu-latest
    timeout-minutes: 120  # Increase from default 60

📄 License

This Docker image build process is MIT licensed. LightGBM itself is licensed under the MIT License.

🤝 Contributing

1. Fork the repository
2. Create your feature branch
3. Commit your changes
4. Push to the branch
5. Open a Pull Request

📚 Resources

• LightGBM Documentation
• LightGBM C API
• Docker Hub Repository
• GitHub Repository