MLops_POC

ML Data, Model Management and Pipelining

Tasks

1. Experiment Tracking
2. Experiment tracking
3. Artifact/model tracking
4. Code versioning
5. Data versioning
6. Data quality
7. Feature stores
8. ML Orchestration
9. Model serving
10. Model monitoring
11. Model explainability

Environment setup:

run conda env create -f environment.yml

Description:

1. Datasets: Datasets are stored in GCP Cloud Storage.
2. property_train: Code to build and train models on property
3. inference: Code for predicting new data using final models