Skip to content

fix recopying step for forest loss driver #154

fix recopying step for forest loss driver

fix recopying step for forest loss driver #154

name: Forest-Loss-Driver-Prediction
on:
workflow_dispatch:
push:
branches:
- henryh/favyen/forest-loss-20240917
env:
REGISTRY: ghcr.io
IMAGE_NAME: ${{ github.repository }}
SERVICE_NAME: "rslearn_projects"
concurrency:
group: ${{ github.workflow }}-${{ github.ref }}
cancel-in-progress: true
jobs:
build:
runs-on: ubuntu-latest
permissions:
contents: read
packages: write
outputs:
ghcr_docker_image: ${{ steps.image-names.outputs.ghcr_docker_image }}
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Log in to the Container registry
uses: docker/login-action@v3
with:
registry: ${{ env.REGISTRY }}
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
- name: Extract metadata (tags, labels) for Docker
id: meta
uses: docker/metadata-action@v5
with:
images: |
${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
tags: |
type=sha,format=long
type=sha,format=short
type=raw,value=latest,enable={{is_default_branch}}
- name: Cleanup disk space
run: |
sudo docker rmi $(docker image ls -aq) >/dev/null 2>&1 || true
sudo docker image prune --all --force >/dev/null 2>&1 || true
sudo rm -rf /usr/share/dotnet
sudo rm -rf /opt/ghc
sudo rm -rf /usr/local/share/boost
- name: Build and push Docker image
id: build-push
uses: docker/build-push-action@v6
with:
context: .
push: true
tags: ${{ steps.meta.outputs.tags }}
labels: ${{ steps.meta.outputs.labels }}
build-args: |
GIT_USERNAME=${{ secrets.GIT_USERNAME }}
GIT_TOKEN=${{ secrets.GIT_TOKEN }}
- name: Store Image Names
id: image-names
run: |-
GHCR_IMAGE="${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}@${{ steps.build-push.outputs.digest }}"
GHCR_IMAGE=`echo ${GHCR_IMAGE} | tr '[:upper:]' '[:lower:]'` # docker requires that all image names be lowercase
echo "ghcr_docker_image=\"${GHCR_IMAGE}\"" >> $GITHUB_OUTPUT
predict:
needs: build
runs-on: ubuntu-latest-m
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Log in to the Container registry
uses: docker/login-action@v3
with:
registry: ${{ env.REGISTRY }}
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
- name: Cleanup disk space
run: |
sudo docker rmi $(docker image ls -aq) >/dev/null 2>&1 || true
sudo docker image prune --all --force >/dev/null 2>&1 || true
sudo rm -rf /usr/share/dotnet
sudo rm -rf /opt/ghc
sudo rm -rf /usr/local/share/boost
- name: Authenticate into gcp
uses: "google-github-actions/auth@v2"
with:
credentials_json: ${{ secrets.GCP_VM_DEPLOYER_CREDENTIALS }}
- name: Run Extract Dataset Job on VM and Launch Prediction Job on Beaker
run: |
export PIPELINE_INFERENCE_CONFIG_PATH="rslp/forest_loss_driver/inference/config/forest_loss_driver_predict_pipeline_config.yaml" && \
export PRED_PIPELINE_CONFIG_ARG="--pred_pipeline_config $PIPELINE_INFERENCE_CONFIG_PATH" && \
# NOTE: The Index cahce dir will be copied to the VM and mounted as a volume in the docker container
export INDEX_CACHE_DIR=${{ secrets.RSLP_PREFIX }}/datasets/forest_loss_driver/index_cache_dir && \
export TILE_STORE_ROOT_DIR=${{ secrets.RSLP_PREFIX }}/datasets/forest_loss_driver/tile_store_root_dir && \
export DATASET_EXTRACT_COMMAND="python -m rslp.main forest_loss_driver extract_dataset $PRED_PIPELINE_CONFIG_ARG" && \
export RSLP_PROJECT="forest_loss_driver" && \
bash .github/workflows/deploy_image_on_vm.sh \
--project-id ${{ secrets.GCP_PROJECT_ID }} \
--zone "us-west1-b" \
--machine-type "n2-standard-128" \
--docker-image ${{ needs.build.outputs.ghcr_docker_image }} \
--command "$DATASET_EXTRACT_COMMAND" \
--user ${{ secrets.GCP_USER }} \
--ghcr-user allenai \
--service-account ${{ secrets.FOREST_LOSS_DRIVER_INFERENCE_SERVICE_ACCOUNT }} \
--delete no \
--beaker-token ${{ secrets.BEAKER_TOKEN }} \
--beaker-addr "https://beaker.org" \
--beaker-username ${{ secrets.BEAKER_USERNAME }} \
--rslp-project $RSLP_PROJECT \
--rslp-prefix ${{ secrets.RSLP_PREFIX }} \
--gpu-count 1 \
--shared-memory "64Gib" \
--cluster ${{ secrets.BEAKER_CLUSTER_INFERENCE }} \
--priority "normal" \
--task-name "${RSLP_PROJECT}_inference_$(uuidgen | cut -c1-8)" \
--budget ${{ secrets.BEAKER_BUDGET }} \
--workspace ${{ secrets.BEAKER_WORKSPACE }} \
--extra_args_model_predict "$PRED_PIPELINE_CONFIG_ARG" && \
echo "Inference job launched!"