Collect cached results from s3 bucket

[katxiao]

• SDGym version: 0.3.0

Description

In issue #79 we added a way to collect results from a collection of intermediate cached results to a single scores CSV file.

It should also be possible to read the intermediate cached results from an S3 bucket, and to store the results csv to a S3 bucket.

Python:

sdgym.collect.collect_results(input_path='s3://my-bucket/path/to/results', output_file='s3://my-bucket/path/to/my/results.csv')

CLI

sdgym collect ... -i s3://my-bucket/path/to/results -o s3://my-bucket/path/to/my/results.csv

If the bucket is private, the AWS key and secret introduce in PR #74 should be used.

[csala]

Closed via #84