Phy GUI Error - No Features Displayed in the FeatureView Window

350 views
Skip to first unread message

Bapun Giri

unread,
Nov 4, 2016, 12:57:13 PM11/4/16
to phy-users
Hi Phy Users,

I am experiencing a strange error where some of the clusters in the Phy-GUI do not show the features in the feautureview window. When clicking on that cluster the terminal prints out the following error:

11:47:12 [E] cli:39               An error has occurred (IndexError): index 111569 is out of bounds for axis 1 with size 111549

11:47:13 [E] cli:39               An error has occurred (IndexError): index 111918 is out of bounds for axis 1 with size 111549

and for some clusters

11:53:36 [E] cli:39               An error has occurred (AssertionError): 

11:53:37 [E] cli:39               An error has occurred (AssertionError): 

11:53:37 [E] cli:39               An error has occurred (AssertionError): 

11:53:38 [E] cli:39               An error has occurred (AssertionError): 

11:53:39 [E] cli:39               An error has occurred (AssertionError): 

11:53:39 [E] cli:39               An error has occurred (AssertionError): 

11:53:40 [E] cli:39               An error has occurred (AssertionError): 

11:53:41 [E] cli:39               An error has occurred (AssertionError): 

11:53:42 [E] cli:39               An error has occurred (AssertionError): 

11:53:42 [E] cli:39               An error has occurred (AssertionError): 

11:53:43 [E] cli:39               An error has occurred (AssertionError): 


I have attached the kwik files for you guys to have a look.

Thanks



tedmaze1-sh03.kwik
tedmaze1-sh03.prm

Cyrille Rossant

unread,
Nov 7, 2016, 4:59:47 AM11/7/16
to Bapun Giri, phy-users
Hi, could you send me the kwx file too?
Thanks
Cyrille

--
You received this message because you are subscribed to the Google Groups "phy-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to phy-users+unsubscribe@googlegroups.com.
To post to this group, send email to phy-...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/phy-users/c96deb47-b272-4fb3-a542-45465b91d0db%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Mostafa Safaie

unread,
Nov 7, 2016, 11:01:16 AM11/7/16
to phy-users
I have the same problem. My phy, phycontrib, and klusta are the latest version (I don't think I had this error before recent updates). When I run phy kwik-gui, most of the clusters raise the following error:
15:53:45 [D] gui_component:393    Select clusters: 13.
15:53:45 [E] cli:39               An error has occurred (IndexError): index 1516749 is out of bounds for axis 1 with size 1516330
15:53:45 [D] cli:42               Traceback (most recent call last):
  File "/home/david/anaconda3/envs/phy/lib/python3.5/site-packages/phy/gui/widgets.py", line 196, in _emit_from_js
    self.emit(text_type(name), json.loads(text_type(arg_json)))
  File "/home/david/anaconda3/envs/phy/lib/python3.5/site-packages/phy/gui/widgets.py", line 115, in emit
    return self._event.emit(*args, **kwargs)
  File "/home/david/anaconda3/envs/phy/lib/python3.5/site-packages/phy/utils/event.py", line 121, in emit
    res.append(callback(*args, **kwargs))
  File "/home/david/anaconda3/envs/phy/lib/python3.5/site-packages/phy/cluster/manual/gui_component.py", line 334, in on_select
    self._emit_select(cluster_ids)
  File "/home/david/anaconda3/envs/phy/lib/python3.5/site-packages/phy/cluster/manual/gui_component.py", line 395, in _emit_select
    self.gui.emit('select', cluster_ids)
  File "/home/david/anaconda3/envs/phy/lib/python3.5/site-packages/phy/gui/gui.py", line 228, in emit
    return self._event.emit(*args, **kwargs)
  File "/home/david/anaconda3/envs/phy/lib/python3.5/site-packages/phy/utils/event.py", line 121, in emit
    res.append(callback(*args, **kwargs))
  File "/home/david/anaconda3/envs/phy/lib/python3.5/site-packages/phy/cluster/manual/views.py", line 1214, in on_select
    x = self._get_feature(x_dim[i, j], spike_ids, f)
  File "/home/david/anaconda3/envs/phy/lib/python3.5/site-packages/phy/cluster/manual/views.py", line 1055, in _get_feature
    return -1. + (2. / self.duration) * self.spike_times[spike_ids]
IndexError: index 1516749 is out of bounds for axis 1 with size 1516330

If you found a solution, please let me know as well.

Thanks.
Mostafa
To unsubscribe from this group and stop receiving emails from it, send an email to phy-users+...@googlegroups.com.

Bapun Giri

unread,
Nov 7, 2016, 12:38:52 PM11/7/16
to phy-users
Hi Cyrille,

Please find the attached .kwx file.




Thanks,

Bapun

To unsubscribe from this group and stop receiving emails from it, send an email to phy-users+...@googlegroups.com.

Cyrille Rossant

unread,
Nov 8, 2016, 6:27:52 AM11/8/16
to phy-users
Hi Mostafa and Bapun
I cannot reproduce the error on both of your datasets. It works fine on my computer with the latest development versions of klusta, phy, and phycontrib. Can you update the packages, delete the .phy subfolders, and try again?
Best,
Cyrille

Mostafa Safaie

unread,
Nov 8, 2016, 8:55:30 AM11/8/16
to Cyrille Rossant, phy-users
Hi Cyrrile,

I did the following:

source activate phy
git clone https://github.com/kwikteam/
(
p
hy
/phy-contrib/klusta)
cd (phy/phy-contrib/klusta)
python setup.py develop 
for each one of the three packages. After that using phy kwik-gui abcd.kwik unfortunately I still get the same error for most of the clusters.


Mostafa



--
You received this message because you are subscribed to the Google Groups "phy-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to phy-users+unsubscribe@googlegroups.com.

To post to this group, send email to phy-...@googlegroups.com.

Bapun Giri

unread,
Nov 8, 2016, 2:45:23 PM11/8/16
to phy-users
Hi Cyrille,

Deleting the .phy folder did solve the issue. However, the error resurfaces once I move from one kwik file to the other, but deleting the .phy subfolder again resolves the issue.

Thanks for the help,

Bapun

Bapun Giri

unread,
Nov 8, 2016, 2:46:34 PM11/8/16
to phy-users
Hi Mostafa,

Have you tried deleting the .phy subfolder ?

Best,

Bapun


On Tuesday, November 8, 2016 at 7:55:30 AM UTC-6, Mostafa Safaie wrote:
Hi Cyrrile,

I did the following:

source activate phy
git clone https://github.com/kwikteam/
(
p
hy
/phy-contrib/klusta)
cd (phy/phy-contrib/klusta)
python setup.py develop 
for each one of the three packages. After that using phy kwik-gui abcd.kwik unfortunately I still get the same error for most of the clusters.


Mostafa


On Tue, Nov 8, 2016 at 11:27 AM, Cyrille Rossant <cyrille...@gmail.com> wrote:
Hi Mostafa and Bapun
I cannot reproduce the error on both of your datasets. It works fine on my computer with the latest development versions of klusta, phy, and phycontrib. Can you update the packages, delete the .phy subfolders, and try again?
Best,
Cyrille

--
You received this message because you are subscribed to the Google Groups "phy-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to phy-users+...@googlegroups.com.

To post to this group, send email to phy-...@googlegroups.com.

Mostafa Safaie

unread,
Nov 8, 2016, 2:50:33 PM11/8/16
to Bapun Giri, phy-users
Hi Bapun,

Yes, I did and it worked.


To unsubscribe from this group and stop receiving emails from it, send an email to phy-users+unsubscribe@googlegroups.com.

To post to this group, send email to phy-...@googlegroups.com.

Cyrille Rossant

unread,
Nov 10, 2016, 5:10:26 AM11/10/16
to Bapun Giri, phy-users
This problem is now fixed in the development version of phycontrib. You no longer have to delete the .phy subfolder when you sort different kwik files in the same directory. 

Boleslaw Osinski

unread,
Jun 15, 2017, 5:50:45 PM6/15/17
to phy-users
Hi all,

I'm running what I think are the latest versions of phy (1.0.9), phy-contrib (1.0.15), and klusta (3.0.16), but I'm still getting this error. Deleting the .phy subfolder fixes it as it did before. The error only occurs after I change a parameter and rerun klusta in the same folder. Just letting you know in case other people are still having it.

Thanks,
Bo
Reply all
Reply to author
Forward
0 new messages