seismometer
seismometer copied to clipboard
AI model evaluation with a focus on healthcare
We do a check if the scores are all between some value and 100, and if so we divide by 100 so that we get percent values. We should consider...
## Is your feature request related to a problem? Please describe When describing model outcomes, we expect a threshold to come from the user, but its not always clear what...
## Is your feature request related to a problem? Please describe Understanding how a site's patient population differs from the model training or local population is important during pre-procurement validation...
## Is your feature request related to a problem? Not a problem, but it would be nice to support Dark Mode. ## Describe the solution you'd like General Dark Mode...
## Is your feature request related to a problem? Please describe Tracking of ipynb's are inherently error prone as diffs of output cells can be very tough. On the other...
## Is your feature request related to a problem? Please describe Default score threshold values are read from `metadata.json` which is loaded from the `config.yml`'s. When loaded, we look at...
## Problem Summary Many of the tests' setup, particularly around data for loaders, is scoped more narrowly than necessary. ## Impact Running pytest suite creates more temp files than needed....
## Problem Summary On build of sphinx documentation, the example notebook is rerun. It is possible for the notebook to raise an error (such as SSLCertVerification) and still have the...
## Problem Summary Our manifest ini tries to include example notebooks, with two problems: Most importantly, pip install doesn't put things like this in discoverable places (and the template builder...
## Problem Summary With the ability to set up data from preloaded frames instead of going through a full load pipeline, there are gaps in expectations. Specifically, when loading events...