luispedro/milk

MILK: Machine Learning Toolkit

Luis Pedro Coelho

• My personal website: https://luispedro.org/
• My newsletter: https://luispedro.substack.com/
• My Twitter profile: @luispedrocoelho
• My Google Scholar: https://scholar.google.com/citations

milk's Issues

Installation errors using Python 2.7.3

Hey @luispedro,

I have Python 2.7.3 installed through homebrew, and I'm seeing the following error when trying to install through pip:

Downloading/unpacking milk
  Running setup.py egg_info for package milk
    build_src
    building extension "milk.supervised._perceptron" sources
    building extension "milk.supervised._lasso" sources
    building extension "milk.unsupervised._kmeans" sources
    building extension "milk.supervised._tree" sources
    building extension "milk.unsupervised._som" sources
    building extension "milk.supervised._svm" sources
    build_src: building npy-pkg config files

Installing collected packages: milk
  Running setup.py install for milk
    unifing config_cc, config, build_clib, build_ext, build commands --compiler options
    unifing config_fc, config, build_clib, build_ext, build commands --fcompiler options
    build_src
    building extension "milk.supervised._perceptron" sources
    building extension "milk.supervised._lasso" sources
    building extension "milk.unsupervised._kmeans" sources
    building extension "milk.supervised._tree" sources
    building extension "milk.unsupervised._som" sources
    building extension "milk.supervised._svm" sources
    build_src: building npy-pkg config files
    customize UnixCCompiler
    customize UnixCCompiler using build_ext
    customize UnixCCompiler
    customize UnixCCompiler using build_ext
    building 'milk.supervised._lasso' extension
    compiling C++ sources
    C compiler: c++ -fno-strict-aliasing -fno-common -dynamic -I/usr/local/include -DNDEBUG -g -O3 -Wall

    compile options: '-I/usr/local/lib/python2.7/site-packages/numpy/core/include -I/usr/local/Cellar/python/2.7.3/Frameworks/Python.framework/Versions/2.7/include/python2.7 -c'
    extra options: '--std=c++0x'
    c++: milk/supervised/_lasso.cpp
    milk/supervised/_lasso.cpp:7:10: fatal error: 'random' file not found
    #include <random>
             ^
    1 error generated.
    milk/supervised/_lasso.cpp:7:10: fatal error: 'random' file not found
    #include <random>
             ^
    1 error generated.
    error: Command "c++ -fno-strict-aliasing -fno-common -dynamic -I/usr/local/include -DNDEBUG -g -O3 -Wall -I/usr/local/lib/python2.7/site-packages/numpy/core/include -I/usr/local/Cellar/python/2.7.3/Frameworks/Python.framework/Versions/2.7/include/python2.7 -c milk/supervised/_lasso.cpp -o build/temp.macosx-10.8-x86_64-2.7/milk/supervised/_lasso.o --std=c++0x" failed with exit status 1
    Complete output from command /usr/local/Cellar/python/2.7.3/Frameworks/Python.framework/Versions/2.7/Resources/Python.app/Contents/MacOS/Python -c "import setuptools;__file__='/var/folders/cj/5794yqw14v3381qs0br04hf80000gn/T/pip-build/milk/setup.py';exec(compile(open(__file__).read().replace('\r\n', '\n'), __file__, 'exec'))" install --record /var/folders/cj/5794yqw14v3381qs0br04hf80000gn/T/pip-RpL1TH-record/install-record.txt --single-version-externally-managed:
    running install

running build

running config_cc

unifing config_cc, config, build_clib, build_ext, build commands --compiler options

running config_fc

unifing config_fc, config, build_clib, build_ext, build commands --fcompiler options

running build_src

build_src

building extension "milk.supervised._perceptron" sources

building extension "milk.supervised._lasso" sources

building extension "milk.unsupervised._kmeans" sources

building extension "milk.supervised._tree" sources

building extension "milk.unsupervised._som" sources

building extension "milk.supervised._svm" sources

build_src: building npy-pkg config files

running build_py

running build_ext

customize UnixCCompiler

customize UnixCCompiler using build_ext

customize UnixCCompiler

customize UnixCCompiler using build_ext

building 'milk.supervised._lasso' extension

compiling C++ sources

C compiler: c++ -fno-strict-aliasing -fno-common -dynamic -I/usr/local/include -DNDEBUG -g -O3 -Wall



compile options: '-I/usr/local/lib/python2.7/site-packages/numpy/core/include -I/usr/local/Cellar/python/2.7.3/Frameworks/Python.framework/Versions/2.7/include/python2.7 -c'

extra options: '--std=c++0x'

c++: milk/supervised/_lasso.cpp

milk/supervised/_lasso.cpp:7:10: fatal error: 'random' file not found

#include <random>

         ^

1 error generated.

milk/supervised/_lasso.cpp:7:10: fatal error: 'random' file not found

#include <random>

         ^

1 error generated.

error: Command "c++ -fno-strict-aliasing -fno-common -dynamic -I/usr/local/include -DNDEBUG -g -O3 -Wall -I/usr/local/lib/python2.7/site-packages/numpy/core/include -I/usr/local/Cellar/python/2.7.3/Frameworks/Python.framework/Versions/2.7/include/python2.7 -c milk/supervised/_lasso.cpp -o build/temp.macosx-10.8-x86_64-2.7/milk/supervised/_lasso.o --std=c++0x" failed with exit status 1

When I clone the repo and run python setup.py install, I get this error:

compile options: '-I/usr/local/lib/python2.7/site-packages/numpy/core/include -I/usr/local/Cellar/python/2.7.3/Frameworks/Python.framework/Versions/2.7/include/python2.7 -c'
extra options: '-std=c++0x -stdlib=libc++'
c++: milk/supervised/_lasso.cpp
milk/supervised/_lasso.cpp:10:10: fatal error: 'eigen3/Eigen/Dense' file not found
#include <eigen3/Eigen/Dense>
         ^
1 error generated.
milk/supervised/_lasso.cpp:10:10: fatal error: 'eigen3/Eigen/Dense' file not found
#include <eigen3/Eigen/Dense>
         ^
1 error generated.

Thanks for any help!

Python 3 version?

I'm interested in doing some work with self organising maps in python 3. I figure it'd make more sense to use existing code, rather than write my own bugs. Is there a plan to convert milk to python 3? If not, and I attempt it, and would you be willing to pull a 3.x branch, or would it be better to focus on getting some SOM code into a bigger project, like scikit-learn?

milk build fails saying 'skipping incompatible.....' on windows 7 64 bit anaconda(spyder 2.2.5) Python 2.7.5 64bits, Qt 4.8.4, PySide 1.2.1

I've been trying to install milk on my windows machine [windows 7 64 bit anaconda(spyder 2.2.5) Python 2.7.5 64bits, Qt 4.8.4, PySide 1.2.1]. Spent whole day trying to figure out what is causing this build fail but failed.
Below is a trail of the >>>pip install milk log...

at the end it skips many lines of code saying 'skipping incompatible...'. Can someone please help me with a way to get around this to work?

copying milk\unsupervised\gaussianmixture.py -> build\lib.win-amd64-2.7\milk\unsupervised
copying milk\unsupervised\kmeans.py -> build\lib.win-amd64-2.7\milk\unsupervised
copying milk\unsupervised\normalise.py -> build\lib.win-amd64-2.7\milk\unsupervised
copying milk\unsupervised\parzen.py -> build\lib.win-amd64-2.7\milk\unsupervised
copying milk\unsupervised\pca.py -> build\lib.win-amd64-2.7\milk\unsupervised
copying milk\unsupervised\pdist.py -> build\lib.win-amd64-2.7\milk\unsupervised
copying milk\unsupervised\som.py -> build\lib.win-amd64-2.7\milk\unsupervised
copying milk\unsupervised\__init__.py -> build\lib.win-amd64-2.7\milk\unsupervised
creating build\lib.win-amd64-2.7\milk\utils
copying milk\utils\parallel.py -> build\lib.win-amd64-2.7\milk\utils
copying milk\utils\utils.py -> build\lib.win-amd64-2.7\milk\utils
copying milk\utils\__init__.py -> build\lib.win-amd64-2.7\milk\utils
creating build\lib.win-amd64-2.7\milk\wrapper
copying milk\wrapper\wraplibsvm.py -> build\lib.win-amd64-2.7\milk\wrapper
copying milk\wrapper\__init__.py -> build\lib.win-amd64-2.7\milk\wrapper

creating build\lib.win-amd64-2.7\milk\tests\data

copying milk\tests\data\jugparallel_jugfile.py -> build\lib.win-amd64-2.7\milk\tests\data
copying milk\tests\data\jugparallel_kmeans_jugfile.py -> build\lib.win-amd64-2.7\milk\tests\data
copying milk\tests\data\__init__.py -> build\lib.win-amd64-2.7\milk\tests\data

creating build\lib.win-amd64-2.7\milk\unsupervised\nnmf
copying milk\unsupervised\nnmf\hoyer.py -> build\lib.win-amd64-2.7\milk\unsupervised\nnmf
copying milk\unsupervised\nnmf\lee_seung.py -> build\lib.win-amd64-2.7\milk\unsupervised\nnmf
copying milk\unsupervised\nnmf\__init__.py -> build\lib.win-amd64-2.7\milk\unsupervised\nnmf
copying milk\tests\data\regression-2-Dec-2009.pp.gz -> build\lib.win-amd64-2.7\milk\tests\data
copying milk\tests\data\__init__.pyc -> build\lib.win-amd64-2.7\milk\tests\data

running build_ext

Looking for python27.dll

customize Mingw32CCompiler

customize Mingw32CCompiler using build_ext

Looking for python27.dll

customize Mingw32CCompiler

customize Mingw32CCompiler using build_ext

building 'milk.supervised._perceptron' extension

compiling C++ sources

C compiler: g++ -g -DDEBUG -DMS_WIN64 -O0 -Wall



creating build\temp.win-amd64-2.7

creating build\temp.win-amd64-2.7\Release

creating build\temp.win-amd64-2.7\Release\milk

creating build\temp.win-amd64-2.7\Release\milk\supervised

compile options: '-DNPY_MINGW_USE_CUSTOM_MSVCR -D__MSVCRT_VERSION__=0x0900 -Ie:\
programs\Anaconda\lib\site-packages\numpy\core\include -Ie:\programs\Anaconda\in
clude -Ie:\programs\Anaconda\PC -c'

extra options: '-std=c++0x'

g++ -g -DDEBUG -DMS_WIN64 -O0 -Wall -DNPY_MINGW_USE_CUSTOM_MSVCR -D__MSVCRT_VERS
ION__=0x0900 -Ie:\programs\Anaconda\lib\site-packages\numpy\core\include -Ie:\pr
ograms\Anaconda\include -Ie:\programs\Anaconda\PC -c milk/supervised/_perceptron
.cpp -o build\temp.win-amd64-2.7\Release\milk\supervised\_perceptron.o -std=c++0
x

Found executable e:\programs\Rtools\gcc-4.6.3\bin\g++.exe

g++ -g -shared build\temp.win-amd64-2.7\Release\milk\supervised\_perceptron.o -L
e:\programs\Anaconda\libs -Le:\programs\Anaconda\PCbuild\amd64 -lpython27 -lmsvc
r90 -o build\lib.win-amd64-2.7\milk\supervised\_perceptron.pyd

e:/programs/rtools/gcc-4.6.3/bin/../lib/gcc/i686-w64-mingw32/4.6.3/../../../../i
686-w64-mingw32/bin/ld.exe: skipping incompatible e:\programs\Anaconda\libs/libp
ython27.a when searching for -lpython27

e:/programs/rtools/gcc-4.6.3/bin/../lib/gcc/i686-w64-mingw32/4.6.3/../../../../i
686-w64-mingw32/bin/ld.exe: skipping incompatible e:\programs\Anaconda\libs/pyth
on27.lib when searching for -lpython27

e:/programs/rtools/gcc-4.6.3/bin/../lib/gcc/i686-w64-mingw32/4.6.3/../../../../i
686-w64-mingw32/bin/ld.exe: skipping incompatible e:\programs\Anaconda\libs\libp
ython27.a when searching for -lpython27

e:/programs/rtools/gcc-4.6.3/bin/../lib/gcc/i686-w64-mingw32/4.6.3/../../../../i
686-w64-mingw32/bin/ld.exe: skipping incompatible e:\programs\Anaconda\libs/libp
ython27.a when searching for -lpython27

e:/programs/rtools/gcc-4.6.3/bin/../lib/gcc/i686-w64-mingw32/4.6.3/../../../../i
686-w64-mingw32/bin/ld.exe: skipping incompatible e:\programs\Anaconda\libs/pyth
on27.lib when searching for -lpython27

e:/programs/rtools/gcc-4.6.3/bin/../lib/gcc/i686-w64-mingw32/4.6.3/../../../../i
686-w64-mingw32/bin/ld.exe: skipping incompatible e:\programs\Anaconda\libs\pyth
on27.lib when searching for -lpython27

e:/programs/rtools/gcc-4.6.3/bin/../lib/gcc/i686-w64-mingw32/4.6.3/../../../../i
686-w64-mingw32/bin/ld.exe: cannot find -lpython27

e:/programs/rtools/gcc-4.6.3/bin/../lib/gcc/i686-w64-mingw32/4.6.3/../../../../i
686-w64-mingw32/bin/ld.exe: skipping incompatible e:\programs\Anaconda\libs/libm
svcr90.a when searching for -lmsvcr90

e:/programs/rtools/gcc-4.6.3/bin/../lib/gcc/i686-w64-mingw32/4.6.3/../../../../i
686-w64-mingw32/bin/ld.exe: skipping incompatible e:\programs\Anaconda\libs\libm
svcr90.a when searching for -lmsvcr90

collect2: ld returned 1 exit status

error: Command "g++ -g -shared build\temp.win-amd64-2.7\Release\milk\supervised\
_perceptron.o -Le:\programs\Anaconda\libs -Le:\programs\Anaconda\PCbuild\amd64 -
lpython27 -lmsvcr90 -o build\lib.win-amd64-2.7\milk\supervised\_perceptron.pyd"
failed with exit status 1

----------------------------------------
Cleaning up...
Command e:\programs\Anaconda\python.exe -c "import setuptools, tokenize;__file__
='c:\\users\\ramesh\\appdata\\local\\temp\\pip_build_Ramesh\\milk\\setup.py';exe
c(compile(getattr(tokenize, 'open', open)(__file__).read().replace('\r\n', '\n')
, __file__, 'exec'))" install --record c:\users\ramesh\appdata\local\temp\pip-6g
mee8-record\install-record.txt --single-version-externally-managed --compile fai
led with error code 1 in c:\users\ramesh\appdata\local\temp\pip_build_Ramesh\mil
k
Storing debug log for failure in C:\Users\Ramesh\pip\pip.log

E:\Programs\Anaconda>

New release on PyPI?

Version 0.5.2 from PyPI fails to install on Wakari due to the C++11 requirement, whereas installation from the Git source succeeds presumably due to PR #9. I'm not sure what your release schedule/policy is usually, but can you push a new release to PyPI? Wakari users will thank you 😃

RuntimeError when running SOM

Hi,

Thanks for the great work!

I found an issue when running SOM. My python shell is under Win7 64bit using Anaconda.
Here is my code.

    from sklearn import datasets
    import milk
    n_points = 1000
    X, color = datasets.samples_generator.make_s_curve(n_points, random_state=0)
    g=(1000,)
    grid = milk.unsupervised.som(data=X, shape=g, iterations=1000, L=.2, radius=4, R=None)

Here is the error message.

    RuntimeError                              Traceback (most recent call last)
    <ipython-input-7-dbad0be19950> in <module>()
    ----> 1 grid = milk.unsupervised.som(data=X, shape=g, iterations=1000, L=.2, radius=4, R=None)

    C:\Anaconda\lib\site-packages\milk\unsupervised\som.pyc in som(data, shape, iterations, L, radius, R)
    109         data = data.astype(np.float32)
    110     grid = np.array(R.sample(list(data), np.product(shape))).reshape(shape + (d,))
    --> 111     putpoints(grid, data, L=L, radius=radius, iterations=iterations, shuffle=True, R=R)
    112     return grid

    C:\Anaconda\lib\site-packages\milk\unsupervised\som.pyc in putpoints(grid, points, L, radius, iterations, shuffle, R)
     49         if shuffle:
     50             random.shuffle(points)
    ---> 51         _som.putpoints(grid, points, L, radius)
     52
     53 def closest(grid, f):

    RuntimeError: Arguments to putpoints don't conform to expectation. Are you calling this directly? This is an internal function!

Did I make a mistake? I have checked the _som.pyd file is in the same folder with SOM.py file, so it seems no issue in SOM.py. However, since I don't know how to open _som.pyd file, I am not sure whether I make a mistake in using codes.

Thanks!

ValueError: too many boolean indices

I'm getting this error:

try_milk/1.9.3 $ python bug.py 
Traceback (most recent call last):
  File "bug.py", line 14, in <module>
    learner.train(features, labels)
  File "/Users/b/.pythonbrew/pythons/Python-2.7/lib/python2.7/site-packages/milk/supervised/multi.py", line 120, in train
    model = self.base.train(features[idxs], (labels[idxs]==i).astype(int), **child_kwargs)
  File "/Users/b/.pythonbrew/pythons/Python-2.7/lib/python2.7/site-packages/milk/supervised/adaboost.py", line 86, in train
    H,A = _adaboost(features, labels, self.base, self.max_iters)
  File "/Users/b/.pythonbrew/pythons/Python-2.7/lib/python2.7/site-packages/milk/supervised/adaboost.py", line 39, in _adaboost
    train_out = names[train_out]
ValueError: too many boolean indices

When I run this code from the examples:

#!/usr/bin/env python

import milk.supervised.randomforest
import milk.supervised.adaboost
import milksets.wine
from milk.supervised.multi import one_against_one

weak = milk.supervised.randomforest.rf_learner()
learner = milk.supervised.adaboost.boost_learner(weak)
learner = one_against_one(learner)

features, labels = milksets.wine.load()

learner.train(features, labels)

system information

try_milk/1.9.3 $ python --version
Python 2.7

try_milk/1.9.3 $ uname -a
Darwin username.local 12.0.0 Darwin Kernel Version 12.0.0: Sun Jun 24 23:00:16 PDT 2012; root:xnu-2050.7.9~1/RELEASE_X86_64 x86_64

Virtualenv:

try_milk/1.9.3 $ yolk -l
Pygments        - 1.5          - active 
Python          - 2.7          - active development (/Users/b/.pythonbrew/pythons/Python-2.7/lib/python2.7/lib-dynload)
bpython         - 0.11         - active 
distribute      - 0.6.26       - active 
ipython         - 0.12.1       - active 
irckit          - 0.1.1        - active 
milk            - 0.4.2        - active 
milksets        - 0.1.3        - active 
numpy           - 1.6.2        - active 
pandas          - 0.8.1        - active 
pip             - 1.1          - active 
python-dateutil - 1.5          - active 
pytz            - 2012c        - active 
setuptools      - 0.6c11       - active 
virtualenv-clone - 0.2.4        - active 
virtualenv      - 1.7.2        - active 
virtualenvwrapper - 3.5          - active 
wsgiref         - 0.1.2        - active development (/Users/b/.pythonbrew/pythons/Python-2.7/lib/python2.7)
yolk            - 0.4.3        - active

pip version 0.6 is missing utils/utils.h

The 0.6 version in

https://pypi.python.org/pypi/milk/

is missing file utils/utils.h.

Import milk shows : SystemError: initialization of _kmeans failed without raising an exception

I installed milk and imported in Ipyhton by import milk. It says the below system error.

In [1]: import milk
---------------------------------------------------------------------------
SystemError                               Traceback (most recent call last)
<ipython-input-1-2c1431d8c8f7> in <module>()
----> 1 import milk

/Users/elancheliyan/anaconda/lib/python3.5/site-packages/milk/__init__.py in <module>()
     55
     56 try:
---> 57     from .nfoldcrossvalidation import nfoldcrossvalidation
     58     from .supervised.defaultclassifier import defaultclassifier
     59     from .supervised.defaultlearner import defaultlearner

/Users/elancheliyan/anaconda/lib/python3.5/site-packages/milk/nfoldcrossvalidation.py in <module>()
----> 1 from .measures.nfoldcrossvalidation import foldgenerator, getfold, nfoldcrossvalidation

/Users/elancheliyan/anaconda/lib/python3.5/site-packages/milk/measures/nfoldcrossvalidation.py in <module>()
      5
      6 from __future__ import division
----> 7 from ..supervised.classifier import normaliselabels
      8 import numpy as np
      9 from functools import reduce

/Users/elancheliyan/anaconda/lib/python3.5/site-packages/milk/supervised/__init__.py in <module>()
     51
     52 from .defaultclassifier import defaultclassifier, svm_simple
---> 53 from .classifier import normaliselabels
     54 from .gridsearch import gridsearch
     55 from .tree import tree_learner

/Users/elancheliyan/anaconda/lib/python3.5/site-packages/milk/supervised/classifier.py in <module>()
     23 from __future__ import division
     24 import numpy as np
---> 25 from .normalise import normaliselabels
     26 from .base import supervised_model
     27

/Users/elancheliyan/anaconda/lib/python3.5/site-packages/milk/supervised/normalise.py in <module>()
      8 import numpy as np
      9 from .base import supervised_model
---> 10 from ..unsupervised.normalise import zscore
     11
     12 __all__ = [

/Users/elancheliyan/anaconda/lib/python3.5/site-packages/milk/unsupervised/__init__.py in <module>()
     16 '''
     17
---> 18 from .kmeans import kmeans,repeated_kmeans, select_best_kmeans
     19 from .gaussianmixture import *
     20 from .pca import pca, mds, mds_dists

/Users/elancheliyan/anaconda/lib/python3.5/site-packages/milk/unsupervised/kmeans.py in <module>()
     25 from numpy import linalg
     26
---> 27 from . import _kmeans
     28 from ..utils import get_pyrandom
     29 from .normalise import zscore

SystemError: initialization of _kmeans failed without raising an exception

I found milk will be helpful in case to do K means clustering with mahalanobis distance instead of euclidean distance. can you please tell me how to rectify this error.

Would it be possible to make it Python3 compatible ?

no py3 wheel on http://www.lfd.uci.edu/~gohlke/pythonlibs/#milk

Errors new in numpy 1.9.0r1

======================================================================
FAIL: milk.tests.test_pdist.test_pdist
----------------------------------------------------------------------
Traceback (most recent call last):
  File "X:\Python27-x64\lib\site-packages\nose\case.py", line 197, in runTest
    self.test(*self.arg)
  File "X:\Python27-x64\lib\site-packages\milk\tests\test_pdist.py", line 11, in test_pdist
    assert np.allclose(Dxx[i,j], np.sum((X[i]-X[j])**2))
AssertionError

======================================================================
FAIL: milk.tests.test_pdist.test_plike
----------------------------------------------------------------------
Traceback (most recent call last):
  File "X:\Python27-x64\lib\site-packages\nose\case.py", line 197, in runTest
    self.test(*self.arg)
  File "X:\Python27-x64\lib\site-packages\milk\tests\test_pdist.py", line 27, in test_plike
    assert Lxx[0,0] == Lxx2[0,0]
AssertionError

Not sure what is going on here.

Documentation does not explain affinity propagation

Hi. I notice that your documentation claims that milk supports affinity propagation, but it does not explain nor provide an explicit example. There are also no affinity propagation benchmarks listed.

Does milk in fact support affinity propagation? If so, how would one use it, and do you have any performance benchmarks against scikit.learn? Further, would this implementation support a precomputed matrix of similarities a la scikit?

Please let me know. Thanks.

Installing milk using virtualenv

I am having problems installing milk using virtualenv. I am writing a script in Jenkins that regularly builds milk since some of the older projects we use have dependencies. I am seeing this error on every build

(virtual_environment)jenkins@developers:/usr0/home/jenkins/workspace/milk$ python setup.py install
running install
running bdist_egg
running egg_info
running build_src
build_src
building extension "milk.supervised._perceptron" sources
building extension "milk.supervised._lasso" sources
building extension "milk.unsupervised._kmeans" sources
building extension "milk.supervised._tree" sources
building extension "milk.unsupervised._som" sources
building extension "milk.supervised._svm" sources
build_src: building npy-pkg config files
writing milk.egg-info/PKG-INFO
writing top-level names to milk.egg-info/top_level.txt
writing dependency_links to milk.egg-info/dependency_links.txt
reading manifest file 'milk.egg-info/SOURCES.txt'
reading manifest template 'MANIFEST.in'
warning: no files found matching '*' under directory 'milk/supervised/eigen3'
writing manifest file 'milk.egg-info/SOURCES.txt'
installing library code to build/bdist.linux-x86_64/egg
running install_lib
running build_py
running build_ext
customize UnixCCompiler
customize UnixCCompiler using build_ext
customize UnixCCompiler
customize UnixCCompiler using build_ext
building 'milk.supervised._lasso' extension
compiling C++ sources
C compiler: g++ -pthread -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall -fPIC

compile options: '-I/usr0/home/jenkins/workspace/milk/virtual_environment/local/lib/python2.7/site-packages/numpy/core/include -I/usr/include/python2.7 -c'
extra options: '-std=c++0x'
g++: milk/supervised/_lasso.cpp
milk/supervised/_lasso.cpp:10: fatal error: eigen3/Eigen/Dense: No such file or directory
compilation terminated.
milk/supervised/_lasso.cpp:10: fatal error: eigen3/Eigen/Dense: No such file or directory
compilation terminated.
error: Command "g++ -pthread -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall -fPIC -I/usr0/home/jenkins/workspace/milk/virtual_environment/local/lib/python2.7/site-packages/numpy/core/include -I/usr/include/python2.7 -c milk/supervised/_lasso.cpp -o build/temp.linux-x86_64-2.7/milk/supervised/_lasso.o -std=c++0x" failed with exit status 1