Chapter 7. Data and Feature Management

Running your end-to-end pipeline

$ cd ~/tmp/Machine-Learning-Engineering-with-MLflow/Chapter07/psystock-data-features-main/

$ conda env create -f conda.yaml
$ conda activate pystock-data-features
$ mlflow run . --experiment-name=psystock_data_pipelines
$ mlflow ui

http://localhost:5000

// Пришлось сделать
$ pip uninstall mlflow
$ pip install mlflow

[???] Using a feature store

Issues:

https://github.com/PacktPublishing/Machine-Learning-Engineering-with-MLflow/issues/8

$ cd ~/tmp/Machine-Learning-Engineering-with-MLflow/Chapter07/psystock_feature_store

$ pip install feast==0.10 protobuf==3.20.*
$ protobuf package to 3.20.x
$ feast init

????

$ feast apply