How to Log and Inspect XGBoost Model Training Metadata

Connecting to Neptune and creating a run

Uploading the source files to Neptune

Data versioning the data stored in the S3 bucket

Downloading the data from the S3 bucket

For download do we need the read or the read/write access to the S3 bucket?

Can we use Neptune to track our raw data processing in the S3?

Uploading the sample data frame to Neptune

Creating a callback to keep track of the XGBoost training

What metadata does the callback keep track of in the XGBoost training?

Downloading the model from Neptune

What will the results look like in the Neptune UI?