These are chat archives for FreeCodeCamp/DataScience
discussion on how we can use statistical methods to measure and improve the efficacy of http://freeCodeCamp.com
goldbergdata sends brownie points to @erictleung :sparkles: :thumbsup: :sparkles:
Compared to human visual assessment, the labels in the ChestXray14 dataset are inaccurate, unclear, and often describe medically unimportant findings.
I would be sad if I screwed up and people lost money or whatever, but healthcare is much higher stakes :worried:
Oh yeah, that can be a big issue. However, the responsibility should always be in the hands of the physician. I don't believe that computers should have the end all say all on medical decisions. So maybe you won't have to worry as much.