Scott Lundberg

Just noting here that there is now a `matplotlib=True` option to force_plot that was kindly added by @jverre. There is more to be done on it, but it is pretty...

@graciofilipe You need the code you pasted (inside a valid HTML document) and also the JS code from `shap.initjs()`.

@noreallyimfine I am not familiar with that conversion process so I can't say how it might go.

These are great questions! Your description of the time bounds is right, except that when `feature_dependence="interventional"` the runtime is O(TLN) for each test sample you want to explain, where T...

Yes. GradientExplainer and DeepExplainer both directly explain tensors of any order. TreeExplainer and KernelExplainer would require you to reshape each sample into a single row.

@shamitb yes, but until your varmax model is in tensorflow or Pytorch then you would have to use one of the model agnostic explainers (like KernelExplainer). Since VARMAX is linear...

It is just a matplotlib plot, so you if you pass `show=False` you can keep manipulating the figure: ```python shap.summary_plot(shap_values, X, show=False) import matplotlib.pyplot as pl pl.savefig("summary_plot.pdf") ```

@vivitar123 they are just a numpy array so `numpy.savetxt` will work. @chaupmcs Perhaps you have defined inline display for matplotlib already in your notebook somewhere? If you did that then...

Looks good, great work!! :) Only one main question: > regex(r"\d*") + "7" > Current guidance is allowed to emit an EOS after any sequence of digits ending in "7"...

Another note: > Current guidance allows grammars that are in an "accepting state" to sample a token without constraints. If the sampled token is not accepted by the grammar, then...