feat(docker): add per-engine Docker dev environments (#649)

Replace the single legacy Dockerfile (rapidocr_api only) with 7
per-engine development images covering all supported inference backends:

- onnxruntime-cpu: ONNX Runtime on CPU (python:3.10-slim-bookworm)
- onnxruntime-gpu: ONNX Runtime with CUDA (nvidia/cuda:12.4.1)
- tensorrt: NVIDIA TensorRT (nvcr.io/nvidia/deepstream:7.0-gc-triton-devel)
- paddle: PaddlePaddle CPU
- openvino: Intel OpenVINO
- pytorch: PyTorch CPU
- mnn: Alibaba MNN

Infrastructure:
- Dockerfile.base: shared foundation for CPU variants
- docker-compose.yaml: all 7 services with shared model volume
- Makefile: make build-*/test-*/shell-* convenience commands
- .dockerignore: excludes .git, models, build artifacts

Key design decisions:
- PYTHONPATH-based imports (setup.py incompatible with pip editable install)
- Named volume (rapidocr-models) for persistent model cache across rebuilds
- Source bind-mounted at runtime for live editing
- tensorrt pinned to 8.6.x + cuda-python 12.x (matching DeepStream 7.0)

All 7 images built and tested with full OCR pipeline inference.

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

Author: LocNgoXuan23

Makefile

@@ -0,0 +1,37 @@
+# RapidOCR Docker Dev Environments
+COMPOSE_FILE := docker/docker-compose.yaml
+ENGINES := onnxruntime-cpu onnxruntime-gpu tensorrt paddle openvino pytorch mnn
+
+.PHONY: help build-all clean $(foreach e,$(ENGINES),build-$(e) test-$(e) shell-$(e))
+
+help:
+	@echo "RapidOCR Docker Dev Environments"
+	@echo ""
+	@echo "Usage:"
+	@echo "  make build-{engine}    Build a dev image"
+	@echo "  make test-{engine}     Run tests in container"
+	@echo "  make shell-{engine}    Open bash shell in container"
+	@echo "  make build-all         Build all images"
+	@echo "  make clean             Remove all containers and images"
+	@echo ""
+	@echo "Engines: $(ENGINES)"
+	@echo ""
+	@echo "Examples:"
+	@echo "  make build-tensorrt"
+	@echo "  make test-onnxruntime-cpu"
+	@echo "  make shell-pytorch"
+
+build-%:
+	docker compose -f $(COMPOSE_FILE) build $*
+
+test-%:
+	docker compose -f $(COMPOSE_FILE) run --rm $* pytest tests/ -v
+
+shell-%:
+	docker compose -f $(COMPOSE_FILE) run --rm $* bash
+
+build-all:
+	docker compose -f $(COMPOSE_FILE) build
+
+clean:
+	docker compose -f $(COMPOSE_FILE) down --rmi local --volumes

README.md

@@ -71,6 +71,22 @@ print(result)
 result.vis("vis_result.jpg")
 ```
 
+### 🐳 Docker
+
+Docker development environments are available for all supported inference engines:
+
+```bash
+# Build and test with ONNX Runtime (CPU)
+make build-onnxruntime-cpu
+make test-onnxruntime-cpu
+
+# Or use any engine: onnxruntime-gpu, tensorrt, paddle, openvino, pytorch, mnn
+make build-tensorrt
+make shell-tensorrt
+```
+
+See [docker/README.md](./docker/README.md) for full setup instructions, GPU configuration, and troubleshooting.
+
 ### 📚 Documentation
 
 Full documentation can be found on [docs](https://rapidai.github.io/RapidOCRDocs/), in Chinese.

docker/.dockerignore

@@ -0,0 +1,20 @@
+.git
+__pycache__
+*.pyc
+*.pyo
+*.egg-info
+dist/
+build/
+.tox/
+.pytest_cache/
+.mypy_cache/
+python/rapidocr/models/
+python/benchmark_*
+python/pred.txt
+python/merged_benchmark_report_*
+python/tensorrt_build_logs_*
+python/validate_benchmark_*
+python/generate_merged_report.py
+python/mnn_benchmark_reference.md
+*.engine
+.mcp.json

docker/Dockerfile.base

@@ -0,0 +1,35 @@
+FROM python:3.10-slim-bookworm
+
+ENV DEBIAN_FRONTEND=noninteractive
+ENV PYTHONDONTWRITEBYTECODE=1
+ENV PYTHONUNBUFFERED=1
+ENV PYTHONPATH=/app
+
+WORKDIR /app
+
+# System dependencies for opencv, pyclipper, shapely C extensions
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    libgl1 \
+    libglib2.0-0 \
+    libsm6 \
+    libxext6 \
+    libxrender1 \
+    gcc \
+    g++ \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy only requirements first for layer caching
+COPY python/requirements.txt /tmp/requirements.txt
+
+# Install core deps with opencv-headless (no GUI needed in container)
+RUN pip install --no-cache-dir --upgrade pip && \
+    pip install --no-cache-dir -r /tmp/requirements.txt && \
+    pip uninstall -y opencv-python 2>/dev/null; \
+    pip install --no-cache-dir opencv-python-headless && \
+    pip install --no-cache-dir pytest && \
+    rm /tmp/requirements.txt
+
+# Copy source (bind-mounted at runtime for live editing via docker-compose)
+COPY python/ /app/
+
+CMD ["bash"]

docker/Dockerfile.mnn

@@ -0,0 +1,5 @@
+FROM rapidocr-base:latest
+
+RUN pip install --no-cache-dir MNN
+
+CMD ["bash"]

docker/Dockerfile.onnxruntime-cpu

@@ -0,0 +1,5 @@
+FROM rapidocr-base:latest
+
+RUN pip install --no-cache-dir onnxruntime
+
+CMD ["bash"]

docker/Dockerfile.onnxruntime-gpu

@@ -0,0 +1,49 @@
+FROM nvidia/cuda:12.4.1-cudnn-runtime-ubuntu22.04
+
+ENV DEBIAN_FRONTEND=noninteractive
+ENV PYTHONDONTWRITEBYTECODE=1
+ENV PYTHONUNBUFFERED=1
+ENV PYTHONPATH=/app
+
+WORKDIR /app
+
+# Install Python 3.10 and system deps
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    software-properties-common \
+    && add-apt-repository ppa:deadsnakes/ppa \
+    && apt-get update \
+    && apt-get install -y --no-install-recommends \
+    python3.10 \
+    python3.10-venv \
+    python3.10-dev \
+    python3-pip \
+    libgl1 \
+    libglib2.0-0 \
+    libsm6 \
+    libxext6 \
+    libxrender1 \
+    gcc \
+    g++ \
+    && rm -rf /var/lib/apt/lists/*
+
+# Make python3.10 the default
+RUN update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.10 1 && \
+    update-alternatives --install /usr/bin/python python /usr/bin/python3.10 1
+
+# Copy and install deps
+COPY python/requirements.txt /tmp/requirements.txt
+
+RUN python -m pip install --no-cache-dir --upgrade pip && \
+    pip install --no-cache-dir -r /tmp/requirements.txt && \
+    pip uninstall -y opencv-python 2>/dev/null; \
+    pip install --no-cache-dir opencv-python-headless && \
+    pip install --no-cache-dir pytest && \
+    rm /tmp/requirements.txt
+
+# Install ONNX Runtime GPU
+RUN pip install --no-cache-dir onnxruntime-gpu
+
+# Copy source (bind-mounted at runtime for live editing via docker-compose)
+COPY python/ /app/
+
+CMD ["bash"]

docker/Dockerfile.openvino

@@ -0,0 +1,5 @@
+FROM rapidocr-base:latest
+
+RUN pip install --no-cache-dir openvino==2023.3.0
+
+CMD ["bash"]

docker/Dockerfile.paddle

@@ -0,0 +1,7 @@
+FROM rapidocr-base:latest
+
+# PaddlePaddle CPU uses a special index URL
+RUN pip install --no-cache-dir paddlepaddle==3.1.1 \
+    -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
+
+CMD ["bash"]

docker/Dockerfile.pytorch

@@ -0,0 +1,7 @@
+FROM rapidocr-base:latest
+
+# PyTorch CPU from official index
+RUN pip install --no-cache-dir torch torchvision torchaudio \
+    --index-url https://download.pytorch.org/whl/cpu
+
+CMD ["bash"]