Jacob Schreiber

icon indicating a website link https://jmschrei.github.io/ icon indicating an email [email protected]

icon location Research Institute of Molecular Pathology icon location I am a Guest Scientist at the Research Institute of Molecular Pathology in Vienna, studying large scale machine learning and computational biology

Results 259 comments of Jacob Schreiber

DeepLift / DeepLiftShap incorrect with common ReLU usage

Hi @NarineK. That makes sense to me. Would it be possible to make that explanation more explicit in the documentation for DeepLift / DeepLiftShap? Given that it doesn't raise an...

DeepLift RevealCancel Rule Support

Just checking in on this. It'd be great to have the RevealCancel rule implemented for people in genomics settings.

Distributing training (or running model.summarize in parallel

Thank you for opening an issue. pomegranate has recently been rewritten from the ground up to use PyTorch instead of Cython (v1.0.0), and so all issues are being closed as...

BayesianNetwork: n_jobs > 1 brings no speed up, and arguments being passed incorrectly

Thank you for opening an issue. pomegranate has recently been rewritten from the ground up to use PyTorch instead of Cython (v1.0.0), and so all issues are being closed as...

Added evidence-based Gibbs sampling to Bayesian Networks

Thank you for your contribution. However, pomegranate has recently been rewritten from the ground up to use PyTorch instead of Cython, and so this PR is unfortunately out of date...

Add linux aarch64 wheel build support

Thank you for your contribution. However, pomegranate has recently been rewritten from the ground up to use PyTorch instead of Cython, and so this PR is unfortunately out of date...

matplotlib breaks `HiddenMarkovModel.fit` if the states are modelled as `MultivariateGaussianDistribution`

Thank you for opening an issue. pomegranate has recently been rewritten from the ground up to use PyTorch instead of Cython (v1.0.0), and so all issues are being closed as...

Support for Apple Silicon?

Thank you for opening an issue. pomegranate has recently been rewritten from the ground up to use PyTorch instead of Cython (v1.0.0), and so all issues are being closed as...

can't fit network using constraints graph with one variable in one of it's nodes

@dolevamir @arainboldt my recollection is that each of the entries has to be a tuple, even if there's only a single element in it. So, try ```python a = networkx.DiGraph()...

can't fit network using constraints graph with one variable in one of it's nodes

Thank you for opening an issue. pomegranate has recently been rewritten from the ground up to use PyTorch instead of Cython (v1.0.0), and so all issues are being closed as...