Skip to content

Updated README.md

Updated README.md #13

name: model-training-evaluate
on: [push]
jobs:
run:
runs-on: [ubuntu-latest]
container: docker://dvcorg/cml-py3:latest
steps:
- uses: actions/checkout@v2
- name: 'Train and Evaluate model'
shell: bash
env:
repo_token: ${{ secrets.GITHUB_TOKEN }}
AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }}
AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
run: |
# Install requirements
pip install -r requirements.txt
# Pull data & run-cache from S3 and reproduce pipeline
dvc pull --run-cache
dvc repro
# Report metrics
echo "## Metrics" >> report.md
git fetch --prune
dvc metrics diff master --show-md >> report.md
# Publish confusion matrix diff
echo -e "## Plots\n### ROC Curve" >> report.md
cml-publish ./results/roc_curve.png --md >> report.md
echo -e "\n### Precision and Recall Curve" >> report.md
cml-publish ./results/precision_recall_curve.png --md >> report.md
cml-send-comment report.md