use features factory in notebooks #4269
Conversation
| " \n", | ||
| " gmm_est=GMM(num_components)\n", | ||
| " gmm_est.set_features(features)\n", | ||
| " gmm_est.set_features(features(X.T))\n", |
There was a problem hiding this comment.
actually we can't, GMM is a CDistrubition, where 'features' is not registered
| "source": [ | ||
| "def estimate_gmm(X, num_components):\n", | ||
| " # bring data into shogun representation (note that Shogun data is in column vector form, so transpose)\n", | ||
| " features=RealFeatures(X.T)\n", |
There was a problem hiding this comment.
for the sake of minimal changes, i would not delete this and do inline below, but rather replace the line with the factory
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "metadata": { |
There was a problem hiding this comment.
in order to avoid all these unrelated changes in the diff, could you first send a PR where do dont change anything, but rather just save the notebook, and then do a second Pr with your changes.
But only do it next time, this one is OK
| " * proper mixture model interface with classical distribution and EM interface (E,M methods)\n", | ||
| " \n", | ||
| " * set_coef dimensiosn check. DImension check in general\n", | ||
| " * put('m_coefficients', _) dimensiosn check. DImension check in general\n", |
| @@ -1,1078 +1,1041 @@ | |||
| { | |||
| "cells": [ | |||
There was a problem hiding this comment.
as you can see, this diff here is everything. Can you send a separate PR that just stores the nb in the new format and then update this one here to just have your changes? thx
vinx13
force-pushed
the
doc/update-notebook
branch
from
65c2c25 to
3d9d608
Compare
|
Great! |
* Update notebook GMM * Update KMeans notebook * Update KernelDensity notebook * Update Tapkee notebook * Update Classification notebook * Update SVM notebook * Update Regression notebook * Update PCA notebook * Update ica notebooks
2 participants
[WIP]