Unreproducable: accuracy for porn class in nsfw_data_scrapper dataset

I tried to test the model against all the data from https://github.com/alexkimxyz/nsfw_data_scraper. It turns out the the model accuracy was pretty low (0.84) for porn classes. I checked several false negative images, should belong to porn. Any suggestions? Thanks

Below are detailed results for porn images:

porn 0.8378566785677277
sexy 0.09817904345614349
neutral 0.028440081768767722
hentai 0.033282149350465834
drawings 0.0022420468568952363

Start by reproducing the reported evaluation against the nsfw_data_scrapper dataset and compare the stated porn-class results with the false-negative images described in the issue. Done means determining whether the low accuracy is reproducible and documenting the cause or a concrete next step; no source file or test is named in the payload.

Written by the indexing model from the issue text.