updated update data

removing models that are known to have training data contaminated with evaluations

updated with new hugging face results

updated pipeline and init

removed commented code

added a test

shortened file name

shortened file name

using URL as file name

WIP. Updated download file. Can now download all files. Need to integrate that code to loop through all files to download or combine files first into a single dataframe and then save that

added new test for a file that currently can be downloaded

Replicating 404 error with a test so I can troubleshoot

Updated download_file method

Build URL from file path is working

moved methods to better match flow

removed most commented out code from details processor

Find files is working as expected

WIP commit. Finding files can be identical as the method in results_data_processor.

added mostly hardcoded generate url method and test

Added download file method and test

Added basic structure of details data processing and testing. For downloading huggingface details dataset files

added todo for test

added a TODO

changed to save and load in a directory

updated gitignore

Updated regression test

comparing current code to the saved file from the last commit

script to save dataframe to a file only if there are no uncommitted files

Added a first regression test attempt. It currently fails and values are hardcoded

fixed test_streamlit_app_runs

Fixed type error

WIP commit. Currently have nlargest error

Added test to test the specific method that is currently producting an error

Added failing integration test. Currently fails because of the addition of the organization to the dataframe

Added organization to dataframe

added failing test for new behavior of organization column. Updated test for rows for the newly added rows

removed code to print the number of outliers. could add it back later as logging potentially

renamed file for clarity

Added .gitignore file

Added new results from hugging face

MC1 column had 8 rows with a value of 1. It didn't make sense given the next highest value was 0.47 . Assuming they were data errors, they were removed

truthfulqa data added to dataframe

Refactor to make later code changes easier

Added test for removal of undesired columns. fixed code error in column removal

Added initial tests. test_columns and test_rows will be updated or removed later as they test for the exact number of columns and rows. the number of rows will change as more models are added

Updated the last updated date to 18Aug

updated with new hugging face results

Updated description with more models

updated results

fixed error