Searchable Keras layers
under review
P
PerceptiLabs
This would add a searchbar to the Workspace which lets you search for any Keras operation and turn it into a PerceptiLabs component.
This provides a huge amount of extra flexibility to the custom modelling part of the tool (without having to create custom components).
Julian Moore
I upvoted but didn't justify the vote, so here goes: signature information is readily available and the PL structure seems to be standardised so it would appear to be moderately low effort approach that would significantly increase what can be done in PL at a stroke - optional/kwarg arguments could either have UI elements for the Settings area generated automatically (nicer!), or the user could simply specify them in custom code (easier)
One caveat however: there are other TF packages such as TensorFlow Probability, which I would very much like to use. However, simply adding TFP layers such as densevariational does require additional effort since the output value (prediction) is no longer a single value but needs to be accessed via .mean(), std_dev() [or similar, IDK exactly)
That said, I think TFP statistical info will also contribute greatly to the value of exploitable models (other that e.g. classifications with softmax outputs where label probabilities are standard) where some idea of the confidence in the output is desirable.
R
Robert Lundberg
Julian Moore: Great point with TFP!
For visualizations it shouldn't be an issue, but there might some blocker somewhere else depending on how TF handles it.
Is there any specific part/component of TFP you are interested in? Then we can add a feature request for it as a start.
Julian Moore
Robert Lundberg: Right now only dense variational so that I can get at mean, SD, but whatever might have to be done to enable that would hopefully enable related things.
Unfortunately, there might be a need to have e.g. distribution/distributionlambdas earlier in the graph for other TFP such as densevariational to work, and that would need checking too. But, enormously useful for when regressions (especially in decision-making) come along so that that the confidence in the output can be assessed.
R
Robert Lundberg
under review