ray/release/golden_notebook_tests/modin_xgboost_app_config.yaml
Amog Kamsetty 9ec5793bea
[Release] Fix XGBoost Golden Notebook Tests (#23996)
Xgboost released a new version a few days ago. Due to caching of the Anyscale cluster env, this resulted in the server having an outdated xgboost version while the client has the most recent version causing the test to fail.

Instead, we reinstall xgboost-ray and xgboost in the post build commands so that these dependencies are not being cached in the cluster env.
2022-04-18 21:44:47 -07:00

26 lines
1 KiB
YAML
Executable file

base_image: "anyscale/ray:nightly-py37"
env_vars: { }
debian_packages:
- curl
python:
pip_packages:
- pandas>=1.3.0 # otherwise, a version mismatch between local and remote will cause an exception
- git+https://github.com/ray-project/xgboost_ray.git#egg=xgboost_ray
- modin>=0.11.0 # ray.services has been removed
- s3fs
- fastapi
- uvicorn
- tblib
conda_packages: [ ]
post_build_cmds:
- pip uninstall -y ray || true
- pip install -U {{ env["RAY_WHEELS"] | default("ray") }}
- {{ env["RAY_WHEELS_SANITY_CHECK"] | default("echo No Ray wheels sanity check") }}
# Upgrade the XGBoost-Ray version in post build commands, otherwise it will be cached in the Anyscale Docker image.
- echo {{ env["TIMESTAMP"] }}
- pip install -U xgboost # Upgrade to latest xgboost version so cached version is not used.
- pip uninstall xgboost_ray -y # Uninstall first so pip does a reinstall.
- pip install -U --no-cache-dir git+https://github.com/ray-project/xgboost_ray.git#egg=xgboost_ray