[#860] Adding Spurious Correlation feature

[allincowell]

Summary

🎯 Purpose: Adding Spurious Correlation feature for Image datasets.

📜 Example Usage: Finds correlation score between one of the image properties like dark score, blurry score, information score, size, aspect_ratio, etc. and the class labels using certain metrics like baseline accuracy and held-out accuracy by fitting a univariate model.

[ ✏️ Write your summary here. ]

1. Added spurious_correlation.py module in cleanlab/datalab/internal location.
2. Added a private instance method _spurious_correlation in Datalab class that uses an instance of SpuriousCorrelations class.

Links to Relevant Issues or Conversations

Issue Link: #860
Early PR attempted: #872

[elisno]

For now, let's omit this argument in Datalab._spurious_correlation().

Remember to remove the parameter in the docstring as well.

[elisno]

Please add a validation step that ensures that the issues dataframe has all the relevant (image-specific) scores.
If it doesn't an error with a helpful message should be raised.

[allincowell]

Added a validation step here to cjeck all vision/image issues are present in the correlations dataframe

[elisno]

To be clear, the issues dataframe should be validated, not the correlations_df.

[codecov]

Codecov Report

Attention: Patch coverage is 39.62264% with 32 lines in your changes missing coverage. Please review.

Project coverage is 95.75%. Comparing base (18dfb0d) to head (454f980).

❗ Current head 454f980 differs from pull request most recent head 2b01057

Please upload reports for the commit 2b01057 to get more accurate results.

Files	Patch %	Lines
cleanlab/datalab/internal/spurious_correlation.py	45.23%	23 Missing ⚠️
cleanlab/datalab/datalab.py	18.18%	9 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1140      +/-   ##
==========================================
+ Coverage   94.34%   95.75%   +1.41%     
==========================================
  Files          80       81       +1     
  Lines        6100     6153      +53     
  Branches     1079     1019      -60     
==========================================
+ Hits         5755     5892     +137     
+ Misses        261      168      -93     
- Partials       84       93       +9     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.