python - sklearn's PLSRegression: "ValueError: array must not contain infs or NaNs" -

-

when using sklearn.cross_decomposition.plsregression:

import numpy np import sklearn.cross_decomposition  pls2 = sklearn.cross_decomposition.plsregression() xx = np.random.random((5,5)) yy = np.zeros((5,5) )   yy[0,:] = [0,1,0,0,0] yy[1,:] = [0,0,0,1,0] yy[2,:] = [0,0,0,0,1] #yy[3,:] = [1,0,0,0,0] # uncommenting line solves issue  pls2.fit(xx, yy)

i get:

c:\anaconda\lib\site-packages\sklearn\cross_decomposition\pls_.py:44: runtimewarning: invalid value encountered in divide   x_weights = np.dot(x.t, y_score) / np.dot(y_score.t, y_score) c:\anaconda\lib\site-packages\sklearn\cross_decomposition\pls_.py:64: runtimewarning: invalid value encountered in less   if np.dot(x_weights_diff.t, x_weights_diff) < tol or y.shape[1] == 1: c:\anaconda\lib\site-packages\sklearn\cross_decomposition\pls_.py:67: userwarning: maximum number of iterations reached   warnings.warn('maximum number of iterations reached') c:\anaconda\lib\site-packages\sklearn\cross_decomposition\pls_.py:297: runtimewarning: invalid value encountered in less   if np.dot(x_scores.t, x_scores) < np.finfo(np.double).eps: c:\anaconda\lib\site-packages\sklearn\cross_decomposition\pls_.py:275: runtimewarning: invalid value encountered in less   if np.all(np.dot(yk.t, yk) < np.finfo(np.double).eps): traceback (most recent call last):   file "c:\svn\hw4\code\test_plsr2.py", line 8, in <module>     pls2.fit(xx, yy)   file "c:\anaconda\lib\site-packages\sklearn\cross_decomposition\pls_.py", line 335, in fit     linalg.pinv(np.dot(self.x_loadings_.t, self.x_weights_)))   file "c:\anaconda\lib\site-packages\scipy\linalg\basic.py", line 889, in pinv     = _asarray_validated(a, check_finite=check_finite)   file "c:\anaconda\lib\site-packages\scipy\_lib\_util.py", line 135, in _asarray_validated     = np.asarray_chkfinite(a)   file "c:\anaconda\lib\site-packages\numpy\lib\function_base.py", line 613, in asarray_chkfinite     "array must not contain infs or nans") valueerror: array must not contain infs or nans

what issue?

i aware of scikit-learn github issue #2089, since use scikit-learn 0.16.1 (with python 2.7.10 x64) problem should solved (the code snippets mentioned in github issue work fine).

please check if of values being passed in nan or inf:

np.isnan(xx).any() np.isnan(yy).any()  np.isinf(xx).any() np.isinf(yy).any()

if of yields true. remove nan entries or inf entries. e.g. can set them 0 with:

xx = np.nan_to_num(xx) yy = np.nan_to_num(yy)

it's possible numpy fed such large positive , negative , zeroed values, equations deep down in library producing zeros, nan's or inf's. 1 workaround, oddly enough, send in smaller numbers (say representative numbers between -1 , 1. 1 way standardization, see: https://stackoverflow.com/a/36390482/445131

if none of solves problem, may dealing low level bug in library using, or sort of singularity in data. create sscce , post stackoverflow or create new bug report on library maintaining software.


Comments

Post a Comment

Popular posts from this blog

java - Static nested class instance -

-
this question has answer here: java inner class , static nested class 23 answers i came across in java, know static nested class why create instance of 'static'. isn't whole idea of 'static' use without instance? understand concept of inner classes, why (and frankly how possible to) create 'instance' of 'static' member @ ? why this: 1- outerclass.staticnestedclass nestedobject = new outerclass.staticnestedclass(); 2- nestedobject.anestedclassmethod(); and not this: 1- outerclass outerinstance=new outerclass(); 2- outerinstance.staticnestedclass.anestedclassmethod(); use on inner classes, keyword static indicates can access inner class without instance of outer class. for example: public class outer { public static class inner { /* code here. */ } } with construction, can create instance of inn
Read more

c# - Bluetooth LE CanUpdate Characteristic property -

-
i building xamarin.forms cross platform mobile app, uses monkey.robotics bluetoth low energy functionality. connecting mbed based implimentation of custom gatt service . in xamarin c#, triggers characteristic. canupdate property in monkey.robotics? this standard example c# based on: if (characteristic.canupdate) { characteristic.valueupdated += (s, e) => { debug.writeline("characteristic.valueupdated"); device.begininvokeonmainthread( () => { updatedisplay(characteristic); }); isbusy = false; // spin until first result received }; isbusy = true; characteristic.startupdates(); } this has been working, since changed own custom gatt service connecting to, canupdate property false. property , how triggered? me debugging gatt service code. thanks monkey.robotics open sour
Read more

JavaScript - Replace variable from string in all occurrences -

-
ok, know if have character '-' , want remove in places in string javascript, ... someword = someword.replace(/-/g, ''); but, when applying array of characters, s not working ... const badchars = ('\/:*?"<>|').split(''); let filename = title.replace(/ /g, '-').tolocalelowercase(); (let item = 0; item < badchars.length; item++) { // below not work global '/ /g' filename = filename.replace(/badchars[item]/g, ''); } any ideas? /badchars[item]/g looks badchars , literally, followed i , t , e , or m . if you're trying use character badchars[item] , you'll need use regexp constructor, , you'll need escape regex-specific characters. escaping regular expression has been well-covered . using that: filename = filename.replace(new regexp(regexp.quote(badchars[item]), 'g'), ''); but , don't want that. want character class: let filename = title.replac
Read more