WIP test out new dockerfile with more nvidia tools

.github/workflows/beta.yml

@@ -0,0 +1,36 @@
+name: beta-docker-images
+
+on:
+ workflow_dispatch:
+ pull_request:
+
+jobs:
+ build-axolotl-beta:
+ if: ${{ ! contains(github.event.commits[0].message, '[skip docker]]') && github.repository_owner == 'OpenAccess-AI-Collective' }}
+ strategy:
+ fail-fast: false
+ runs-on: axolotl-gpu-runner
+ steps:
+ - name: Checkout
+ uses: actions/checkout@v4
+ - name: Docker metadata
+ id: metadata
+ uses: docker/metadata-action@v5
+ with:
+ images: winglian/axolotl-beta
+ - name: Set up Docker Buildx
+ uses: docker/setup-buildx-action@v3
+ - name: Login to Docker Hub
+ uses: docker/login-action@v3
+ with:
+ username: ${{ secrets.DOCKERHUB_USERNAME }}
+ password: ${{ secrets.DOCKERHUB_TOKEN }}
+ # guidance for testing before pushing: https://docs.docker.com/build/ci/github-actions/test-before-push/
+ - name: Build and export to Docker
+ uses: docker/build-push-action@v5
+ with:
+ context: .
+ file: ./docker/Dockerfile-beta
+ tags: |
+ ${{ steps.metadata.outputs.tags }}
+ labels: ${{ steps.metadata.outputs.labels }}

docker/Dockerfile-base

@@ -35,3 +35,9 @@ RUN git lfs install --skip-repo && \
  pip3 install awscli && \
  # The base image ships with `pydantic==1.8.2` which is not working
  pip3 install -U --no-cache-dir pydantic==1.10.10
+
+WORKDIR /workspace
+
+RUN git clone --depth=1 https://github.com/OpenAccess-AI-Collective/axolotl.git
+
+WORKDIR /workspace/axolotl

docker/Dockerfile-beta

@@ -0,0 +1,40 @@
+FROM nvcr.io/nvidia/pytorch:24.03-py3
+
+RUN python3 -m pip install --upgrade pip
+
+RUN groupadd axolotl && useradd -m -g axolotl -s /bin/bash axolotl
+
+USER axolotl
+
+RUN mkdir -p /home/axolotl/venv
+
+RUN python -m venv /home/axolotl/venv/axolotl
+
+ENV PATH="/home/axolotl/venv/axolotl/bin:$PATH"
+
+RUN echo "source /home/axolotl/venv/axolotl/bin/activate" >> /home/axolotl/.bashrc
+
+RUN git lfs install --skip-repo && \
+ pip3 install awscli
+
+RUN pip install causal_conv1d && \
+ pip install -e .[deepspeed,flash-attn,mamba-ssm,galore]
+
+# So we can test the Docker image
+RUN pip install pytest
+
+# fix so that git fetch/pull from remote works
+RUN git config remote.origin.fetch "+refs/heads/*:refs/remotes/origin/*" && \
+ git config --get remote.origin.fetch
+
+# helper for huggingface-login cli
+RUN git config --global credential.helper store
+
+
+ENV HF_DATASETS_CACHE="/workspace/data/huggingface-cache/datasets"
+ENV HUGGINGFACE_HUB_CACHE="/workspace/data/huggingface-cache/hub"
+ENV TRANSFORMERS_CACHE="/workspace/data/huggingface-cache/hub"
+ENV HF_HOME="/workspace/data/huggingface-cache/hub"
+ENV HF_HUB_ENABLE_HF_TRANSFER="1"
+
+CMD ["sleep", "infinity"]