search

mordred-descriptor / mordred

a molecular descriptor calculator
http://mordred-descriptor.github.io/documentation/master/
BSD 3-Clause "New" or "Revised" License
355 stars 95 forks source link

RuntimeWarning: overflow encountered in reduce #77

Open frozenarm opened 5 years ago

frozenarm commented 5 years ago

Molecular descriptors not calculated

python3.6/site-packages/numpy/core/fromnumeric.py:86: RuntimeWarning: overflow encountered in reducepython3.6/site-packages/numpy/core/fromnumeric.py:86:

Please fill me if possible.

environment

OS/distribution

rhel7

conda or pip

pip3

python version

Please execute the command and paste result.

python3.6

library version

Please execute the command and paste result.

philopon commented 5 years ago

Thank you for your reporting.

Could you please give me more detail information? (e.g. input molecule if you possible, the value you were supposed and the actual calculated value.)

Cheers,

frozenarm commented 5 years ago

I am afraid I can't share structures or values. I am trying to identify which descriptors give nans and though that you should be aware of the issue. Is there a way to control time-out ?

philopon commented 5 years ago

I am afraid I can't share structures or values.

Don't worry about it.

I am trying to identify which descriptors give nans and though that you should be aware of the issue. Is there a way to control time-out?

mordred is calculating very long time, then overflow is reported? DetourMatrix descriptor may such behavior (it is NP-hard problem). In version 1.2.0, it is limited about < 60seconds.

Could you tell me the mordred version?

python -c 'import mordred; print(mordred.__version__)'

Thanks,

frozenarm commented 5 years ago

The version that I am using is: 1.2.0 I found two 2d descriptors that generate nans: VR1_A and VR2_A if that helps.

philopon commented 5 years ago

Thank you for your information.

It seems bit strange result. These two descriptors are numerically stable relatively.

Would you tell me as much as you can say about approximate size (<50 heavy atoms, Mw > 1000 etc.) of the molecule?

Thanks,

ardejani commented 4 years ago

I am facing the same issue. This happens for certain descriptors but for all of my widely different molecules:

(638, 175) MAXsLi \ 0 max() arg is an empty sequence (MAXsLi)
1 max() arg is an empty sequence (MAXsLi)
2 max() arg is an empty sequence (MAXsLi)
3 max() arg is an empty sequence (MAXsLi)
4 max() arg is an empty sequence (MAXsLi)
.. ...
633 max() arg is an empty sequence (MAXsLi)
634 max() arg is an empty sequence (MAXsLi)
635 max() arg is an empty sequence (MAXsLi)
636 max() arg is an empty sequence (MAXsLi)
637 max() arg is an empty sequence (MAXsLi) 

                                  MAXssBe  \

0 max() arg is an empty sequence (MAXssBe)
1 max() arg is an empty sequence (MAXssBe)
2 max() arg is an empty sequence (MAXssBe)
3 max() arg is an empty sequence (MAXssBe)
4 max() arg is an empty sequence (MAXssBe)
.. ...
633 max() arg is an empty sequence (MAXssBe)
634 max() arg is an empty sequence (MAXssBe)
635 max() arg is an empty sequence (MAXssBe)
636 max() arg is an empty sequence (MAXssBe)
637 max() arg is an empty sequence (MAXssBe) 

                                  MAXssssBe  \

0 max() arg is an empty sequence (MAXssssBe)
1 max() arg is an empty sequence (MAXssssBe)
2 max() arg is an empty sequence (MAXssssBe)
3 max() arg is an empty sequence (MAXssssBe)
4 max() arg is an empty sequence (MAXssssBe)
.. ...
633 max() arg is an empty sequence (MAXssssBe)
634 max() arg is an empty sequence (MAXssssBe)
635 max() arg is an empty sequence (MAXssssBe)
636 max() arg is an empty sequence (MAXssssBe)
637 max() arg is an empty sequence (MAXssssBe) 

                                  MAXssBH  \

0 max() arg is an empty sequence (MAXssBH)
1 max() arg is an empty sequence (MAXssBH)
2 max() arg is an empty sequence (MAXssBH)
3 max() arg is an empty sequence (MAXssBH)
4 max() arg is an empty sequence (MAXssBH)
.. ...
633 max() arg is an empty sequence (MAXssBH)
634 max() arg is an empty sequence (MAXssBH)
635 max() arg is an empty sequence (MAXssBH)
636 max() arg is an empty sequence (MAXssBH)
637 max() arg is an empty sequence (MAXssBH) 

                                  MAXsssB  \

0 max() arg is an empty sequence (MAXsssB)
1 max() arg is an empty sequence (MAXsssB)
2 max() arg is an empty sequence (MAXsssB)
3 max() arg is an empty sequence (MAXsssB)
4 max() arg is an empty sequence (MAXsssB)
.. ...
633 max() arg is an empty sequence (MAXsssB)
634 max() arg is an empty sequence (MAXsssB)
635 max() arg is an empty sequence (MAXsssB)
636 max() arg is an empty sequence (MAXsssB)
637 max() arg is an empty sequence (MAXsssB) 

                                  MAXssssB  \

0 max() arg is an empty sequence (MAXssssB)
1 max() arg is an empty sequence (MAXssssB)
2 max() arg is an empty sequence (MAXssssB)
3 max() arg is an empty sequence (MAXssssB)
4 max() arg is an empty sequence (MAXssssB)
.. ...
633 max() arg is an empty sequence (MAXssssB)
634 max() arg is an empty sequence (MAXssssB)
635 max() arg is an empty sequence (MAXssssB)
636 max() arg is an empty sequence (MAXssssB)
637 max() arg is an empty sequence (MAXssssB) 

                                  MAXsCH3  \

0 1.92183
1 1.92548
2 1.94464
3 max() arg is an empty sequence (MAXsCH3)
4 1.92668
.. ...
633 1.94655
634 1.9251
635 2.153
636 max() arg is an empty sequence (MAXsCH3)
637 max() arg is an empty sequence (MAXsCH3) 

                                  MAXdCH2  \

0 max() arg is an empty sequence (MAXdCH2)
1 max() arg is an empty sequence (MAXdCH2)
2 max() arg is an empty sequence (MAXdCH2)
3 max() arg is an empty sequence (MAXdCH2)
4 max() arg is an empty sequence (MAXdCH2)
.. ...
633 max() arg is an empty sequence (MAXdCH2)
634 max() arg is an empty sequence (MAXdCH2)
635 max() arg is an empty sequence (MAXdCH2)
636 max() arg is an empty sequence (MAXdCH2)
637 max() arg is an empty sequence (MAXdCH2) 

                                  MAXssCH2  \

0 max() arg is an empty sequence (MAXssCH2)
1 0.90711
2 0.687273
3 0.627547
4 max() arg is an empty sequence (MAXssCH2)
.. ...
633 0.494152
634 0.446215
635 1.13061
636 max() arg is an empty sequence (MAXssCH2)
637 max() arg is an empty sequence (MAXssCH2) 

                                  MAXtCH  ...  \

0 max() arg is an empty sequence (MAXtCH) ...
1 max() arg is an empty sequence (MAXtCH) ...
2 max() arg is an empty sequence (MAXtCH) ...
3 max() arg is an empty sequence (MAXtCH) ...
4 max() arg is an empty sequence (MAXtCH) ...
.. ... ...
633 max() arg is an empty sequence (MAXtCH) ...
634 max() arg is an empty sequence (MAXtCH) ...
635 max() arg is an empty sequence (MAXtCH) ...
636 max() arg is an empty sequence (MAXtCH) ...
637 max() arg is an empty sequence (MAXtCH) ... 

                          MDEO-11                           MDEO-12  \

0 float division by zero (MDEO-11) float division by zero (MDEO-12)
1 float division by zero (MDEO-11) float division by zero (MDEO-12)
2 float division by zero (MDEO-11) 0.2
3 float division by zero (MDEO-11) 0.125
4 0.333333 float division by zero (MDEO-12)
.. ... ...
633 0.5 0.333333
634 0.5 0.5
635 0.45428 0.685007
636 0.421716 0.60822
637 1.48512 float division by zero (MDEO-12) 

                          MDEO-22                           MDEN-11  \

0 float division by zero (MDEO-22) float division by zero (MDEN-11)
1 float division by zero (MDEO-22) float division by zero (MDEN-11)
2 float division by zero (MDEO-22) float division by zero (MDEN-11)
3 float division by zero (MDEO-22) float division by zero (MDEN-11)
4 float division by zero (MDEO-22) float division by zero (MDEN-11)
.. ... ...
633 float division by zero (MDEO-22) float division by zero (MDEN-11)
634 float division by zero (MDEO-22) float division by zero (MDEN-11)
635 float division by zero (MDEO-22) float division by zero (MDEN-11)
636 float division by zero (MDEO-22) float division by zero (MDEN-11)
637 float division by zero (MDEO-22) float division by zero (MDEN-11) 

                          MDEN-12                           MDEN-13  \

0 float division by zero (MDEN-12) float division by zero (MDEN-13)
1 float division by zero (MDEN-12) float division by zero (MDEN-13)
2 float division by zero (MDEN-12) float division by zero (MDEN-13)
3 0.44814 0.166667
4 float division by zero (MDEN-12) float division by zero (MDEN-13)
.. ... ...
633 float division by zero (MDEN-12) float division by zero (MDEN-13)
634 float division by zero (MDEN-12) float division by zero (MDEN-13)
635 float division by zero (MDEN-12) float division by zero (MDEN-13)
636 float division by zero (MDEN-12) float division by zero (MDEN-13)
637 float division by zero (MDEN-12) float division by zero (MDEN-13) 

                          MDEN-22                           MDEN-23  \

0 2.81887 1.15194
1 2.37629 1.64603
2 1.21227 0.491859
3 0.819828 0.629961
4 0.819828 float division by zero (MDEN-23)
.. ... ...
633 0.25 1.1547
634 0.25 1.1547
635 float division by zero (MDEN-22) float division by zero (MDEN-23)
636 float division by zero (MDEN-22) float division by zero (MDEN-23)
637 float division by zero (MDEN-22) float division by zero (MDEN-23) 

                          MDEN-33     Vabc

0 float division by zero (MDEN-33) 364.741
1 0.0769231 357.748
2 float division by zero (MDEN-33) 379.142
3 float division by zero (MDEN-33) 359.761
4 float division by zero (MDEN-33) 296.241
.. ... ...
633 float division by zero (MDEN-33) 286.171
634 float division by zero (MDEN-33) 284.086
635 float division by zero (MDEN-33) 357.015
636 float division by zero (MDEN-33) 241.473
637 float division by zero (MDEN-33) 244.343