dianna
dianna copied to clipboard
dianna-ai
Deep Insight And Neural Network Analysis
I came across this library recently: https://github.com/nbQA-dev/nbQA With it, you can lint notebooks with the same tools you use to lint regular python files. Adding this snippet to `.pre-commit-config.yaml` enables...
The current dashboard output for the reset model looks like the attached screenshot. This output is generated using the Dianna visualisation. This is different from the visualisation in the readme....
The example notebook for rise for time is missing a multivariate example.
Looks good! One thing is the XAI method's parameters. Those are method specific and I suggest we therefore move them in the method's (output) section like in my mockups above....
It is getting a bit convoluted what the input types of specific functions are. We now have options: numpy array, xarray, string, list of strings. I think it would help...
LimeSegment assume convenient input including tests See #369 and #366 for background.
might be fixed automatically after fixing #435
Personally, I find the colour coding in the explanation figure, especially for text, quite confusing, mainly because to me the colour red always has some negative meaning. This to me...