Theano and Sklearn standard tests methods disabled from l_python3_bu_2018.0.008.tgz

[sergio_r_]

Hello folks,

Not sure how this helps to trust the Intel Python distribution,
but the following tests suite were disabled from the Intel Python Distribution for Linux  
[Python 3.5.3 |Intel Corporation| (default, May 18 2017, 22:17:21)]

$ theano-nose --theano  -v

----------------------------------------------------------------------
Ran 0 tests in 0.001s

OK

$ nosetests -v sklearn

----------------------------------------------------------------------
Ran 0 tests in 0.002s

OK

Sergio
Enhance your #MachineLearning and #BigData skills via #Python #SciPy
1) https://www.packtpub.com/big-data-and-business-intelligence/numerical-and-scientific-computing-scipy-video
2) https://www.packtpub.com/big-data-and-business-intelligence/learning-scipy-numerical-and-scientific-computing-second-edition

3) https://www.researchgate.net/publication/301293668_Aprendiendo_a_Programar_en_Python_con_mi_Computador_Primeros_pasos_rumbo_a_computos_de_gran_escala_en_las_Ciencias_e_Ingenierias

[Rohit_J_Intel]

Sergio,

We test scikit-learn and theano using following commands respectively:
1. <Intel-Distribution-for-Python_installation>/bin/python -c "import nose; nose.main()" -v sklearn
2. <Intel-Distribution-for-Python_installation>/bin/python -c "import theano; theano.test()"

Thanks,
Rohit

[sergio_r_]

Rohit J. (Intel) wrote:

Sergio,

We test scikit-learn and theano using following commands respectively:
1. <Intel-Distribution-for-Python_installation>/bin/python -c "import nose; nose.main()" -v sklearn
2. <Intel-Distribution-for-Python_installation>/bin/python -c "import theano; theano.test()"

Thanks,
Rohit

Hi Rohit,

Thanks for your reply.

The first (testing sklearn) command did not work:

$ python -c "import nose; nose.main()" -v sklearn

----------------------------------------------------------------------
Ran 0 tests in 0.002s

OK

The second one (testing theano) ended with and error. The output is below. Hope someone knows how to overcome it.

Regards,

Sergio
Enhance your #MachineLearning and #BigData skills via #Python #SciPy
1) https://www.packtpub.com/big-data-and-business-intelligence/numerical-and-scientific-computing-scipy-video
2) https://www.packtpub.com/big-data-and-business-intelligence/learning-scipy-numerical-and-scientific-computing-second-edition
3) FREE (beginner) PYTHON book in Spanish:
   https://www.researchgate.net/publication/301293668_Aprendiendo_a_Programar_en_Python_con_mi_Computador_Primeros_pasos_rumbo_a_computos_de_gran_escala_en_las_Ciencias_e_Ingenierias

$ python -c "import theano; theano.test()"

icpc version 18.0.0 Beta (gcc version 4.8.0 compatibility)
Theano version 0.9.0dev2.dev-00216eff7fa927ba26c0629697e22fbd2d708e5c
theano is installed in /home/intel2018py/intelpython3/lib/python3.5/site-
packages/theano
NumPy version 1.12.1
NumPy relaxed strides checking option: True
NumPy is installed in /home/intel2018py/intelpython3/lib/python3.5/site-p
ackages/numpy
Python version 3.5.3 |Intel Corporation| (default, May 18 2017, 22:17:21) [GCC 4
.8.2 20140120 (Red Hat 4.8.2-15)]
nose version 1.3.7
/home/intel2018py/intelpython3/lib/python3.5/site-packages/theano/misc/py
cuda_init.py:35: UserWarning: PyCUDA import failed in theano.misc.pycuda_init
  warnings.warn("PyCUDA import failed in theano.misc.pycuda_init")
/home/intel2018py/intelpython3/lib/python3.5/site-packages/theano/sandbox
/gpuarray/__init__.py:10: UserWarning: theano.sandbox.gpuarray has been moved to
 theano.gpuarray. Please update your code and pickles. If the warning persists,
clear theano's cache ('$theano/bin/theano-cache clear').
...
...
...
............./home/intel2018py/intelpython3/lib/python3.5/site-packages/scipy/sparse/compressed.py:774: SparseEfficiencyWarning: Changing the sparsity structure of a csc_matrix is expensive. lil_matrix is more efficient.
  SparseEfficiencyWarning)
/home/intel2018py/intelpython3/lib/python3.5/site-packages/scipy/sparse/compressed.py:774: SparseEfficiencyWarning: Changing the sparsity structure of a csr_matrix is expensive. lil_matrix is more efficient.
  SparseEfficiencyWarning)
...S....................................Error in file [/home/.theano/compiledir_Linux-3.13--generic-x86_64-with-debian-jessie-sid-x86_64-3.5.3-64/tmp4hdl47fr/mod.cpp:750], err code (-1)

[Christophe_H_Intel2]

Sergio,

Can you try running the theano tests with the following command and post the output?

theano-nose --theano -v --exe

The tests require the "nose_parameterized" module as well

conda install nose-parameterized

Thanks,

Chris