mirror of
https://github.com/ruvnet/RuVector.git
synced 2026-05-29 19:33:34 +00:00
3dc7753473
- Rewrite run_calibration.py to use gguf Python package + llama-cpp-python prebuilt wheels instead of compiling llama.cpp from source - Simplify Dockerfile: single-stage, pip install only, no CUDA compilation (build time: ~5min vs 20+min) - Update ADR-129 with tooling decision section explaining ruvllm-native choice - Remove llama-imatrix and llama-quantize binary dependencies Co-Authored-By: claude-flow <ruv@ruv.net>
40 lines
1.2 KiB
Docker
40 lines
1.2 KiB
Docker
# RuvLTRA Training Pipeline
|
|
# Supports: LoRA SFT, DPO, imatrix calibration, GGUF conversion
|
|
# Target: Cloud Run Jobs with L4 GPU or Vertex AI
|
|
|
|
FROM nvidia/cuda:12.4.1-runtime-ubuntu22.04
|
|
|
|
ENV DEBIAN_FRONTEND=noninteractive
|
|
ENV PYTHONUNBUFFERED=1
|
|
|
|
RUN apt-get update && apt-get install -y --no-install-recommends \
|
|
python3.11 python3.11-venv python3.11-dev python3-pip git curl \
|
|
&& rm -rf /var/lib/apt/lists/* \
|
|
&& update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.11 1 \
|
|
&& update-alternatives --install /usr/bin/python python /usr/bin/python3.11 1
|
|
|
|
# Install Python dependencies (prebuilt wheels, no CUDA compilation needed)
|
|
RUN pip install --no-cache-dir \
|
|
torch==2.3.1 \
|
|
transformers>=4.44.0 \
|
|
peft>=0.12.0 \
|
|
trl>=0.9.0 \
|
|
datasets>=2.20.0 \
|
|
huggingface_hub>=0.24.0 \
|
|
accelerate>=0.33.0 \
|
|
bitsandbytes>=0.43.0 \
|
|
sentencepiece \
|
|
protobuf \
|
|
safetensors
|
|
|
|
# Install llama-cpp-python with CUDA support (prebuilt wheel)
|
|
RUN CMAKE_ARGS="-DGGML_CUDA=on" pip install --no-cache-dir llama-cpp-python>=0.2.80
|
|
|
|
# Install llama.cpp tools via pip (no source build needed)
|
|
RUN pip install --no-cache-dir gguf
|
|
|
|
WORKDIR /app
|
|
COPY . /app/
|
|
|
|
ENTRYPOINT ["python", "-u"]
|
|
CMD ["run_calibration.py"]
|