Poisson naive Bayes classifier (PoissonNB) #3708
Conversation
| X = check_array(X) | ||
|
|
||
| joint_log_likelihood = [] | ||
| for i in range(np.size(self.classes_)): |
There was a problem hiding this comment.
np.size(self.classes_) -> len(self.classes_)
|
looks clean you'll need to add proper tests and some narrative doc to explain for what type of data this model is useful. |
|
By narrative doc, do you mean expand the class docstring or somewhere else? |
|
Narrative as distinct from docstrings, examples and tutorials: stuff in On 27 September 2014 23:36, BrianLondon notifications@github.com wrote:
|
| >>> clf.fit(X, y) | ||
| PoissonNB() | ||
| >>> print(clf.predict(X)) | ||
| [1 1 1 2 2 2] |
There was a problem hiding this comment.
Can you add here a reference paper which includes the derivation?
| @@ -0,0 +1,91 @@ | |||
| """ | |||
There was a problem hiding this comment.
why is this file here?
same for out.txt
looks like something went wrong in this PR.
There was a problem hiding this comment.
Yeah, I checked in four working files by mistake, then only pulled two back out.
There was a problem hiding this comment.
you'll need to squash commits to remove these files from history
Poisson classifiers have important applications in systems neuroscience. See: Ma et Al. Nat. Neuroscience 9:1432 (2006). Added some tests specifically for the PoissonNB class. Implemented some of the TODOs in test_naive_bayes.py while I was in there. Added documentation of PoissonNB to naive_bayes.rst and some details to the docstring in naive_bayes.py. Adding documentation for PoissonNB class.
|
Added previous round (including PEP8) fixes and rebased. |
| for i in range(np.size(self.classes_)): | ||
|
|
||
| joint_log_likelihood = np.zeros((np.shape(X)[0], | ||
| np.size(self.classes_))) |
|
Closing stale PR. I'll reopen if I get time to work on this again. |
Added PoissonNB class to naive_bayes.py to implement a Poisson naieve Bayes classifier. Poisson classifiers have important applications in systems neuroscience. See: Ma et Al. Bayesian inference with probabilistic population codes Nat. Neuroscience 9:1432 (2006).