kwikteam / klusta Goto Github PK

Hi, I had a problem with using klusta
As you guided in 'Installation Issue #33', I installed klusta with the 'klustaviewa-setup.exe' file.
And I get an error like below

C:\Program Files\KlustaViewa\WinPython-64bit-2.7.6.4\python-2.7.6.amd64\Scripts>klusta.exe 0526.prm
11:59:55 [I] creator:446 Using kwik file C:\Program Files\KlustaViewa\WinPython-64bit-2.7.6.4\python-2.7.6.amd64\Scripts\0526.kwik.
11:59:55 [I] launch:206 Starting spike detection.
11:59:55 [I] launch:53 SpikeDetekt parameters:
11:59:55 [I] launch:55 threshold_strong_std_factor = 4.5
11:59:55 [I] launch:55 use_single_threshold = 1
11:59:55 [I] launch:55 filter_low = 500.0
11:59:55 [I] launch:55 dtype = int16
11:59:55 [I] launch:55 threshold_weak_std_factor = 2.0
11:59:55 [I] launch:55 n_channels = 16
11:59:55 [I] launch:55 filter_lfp_low = 0
11:59:55 [I] launch:55 raw_data_files = 0526.dat
11:59:55 [I] launch:55 chunk_size_seconds = 1
11:59:55 [I] launch:55 filter_high_factor = 0.475
11:59:55 [I] launch:55 prb_file = 0526.prb
11:59:55 [I] launch:55 voltage_gain = 10.0
11:59:55 [I] launch:55 extract_s_after = 16
11:59:55 [I] launch:55 chunk_overlap_seconds = 0.015
11:59:55 [I] launch:55 n_excerpts = 50
11:59:55 [I] launch:55 filter_butter_order = 3
11:59:55 [I] launch:55 weight_power = 2
11:59:55 [I] launch:55 excerpt_size_seconds = 1
11:59:55 [I] launch:55 connected_component_join_size = 1
11:59:55 [I] launch:55 extract_s_before = 16
11:59:55 [I] launch:55 experiment_name = 0526.kwik
11:59:55 [I] launch:55 sample_rate = 30000
11:59:55 [I] launch:55 n_features_per_channel = 3
11:59:55 [I] launch:55 detect_spikes = negative
11:59:55 [I] launch:55 pca_n_waveforms_max = 10000
11:59:55 [I] launch:55 filter_lfp_high = 300
11:59:55 [I] spikedetekt:212 Found 4 live channels in group 0.
11:59:55 [I] spikedetekt:212 Found 4 live channels in group 1.
11:59:55 [I] spikedetekt:212 Found 4 live channels in group 2.
11:59:55 [I] spikedetekt:212 Found 4 live channels in group 3.
11:59:55 [I] spikedetekt:285 Finding the thresholds...
Detecting spikes : 100%|##########| 204/204 [00:43<00:00, 4.71it/s]
Extracting waveforms : 100%|##########| 204/204 [00:11<00:00, 17.35it/s]
Performing PCA : 100%|##########| 4/4 [00:00<00:00, 114.29it/s]
Computing features : 100%|##########| 204/204 [00:18<00:00, 8.73it/s]
12:01:12 [I] launch:218 Spike detection done! 1123 spikes found in shank 0, 1016 spikes found in shank 1, 12448 spikes found in shank 2, 2127 spikes found in shank 3
12:01:12 [I] launch:249 Starting clustering on shank 0/4.
12:01:12 [I] launch:252 Clustering group 0 (1123 spikes).
12:01:12 [I] launch:123 Starting KK...
ImportError: No module named klustakwik2

C:\Program Files\KlustaViewa\WinPython-64bit-2.7.6.4\python-2.7.6.amd64\Scripts>

I use Window 10, 64bit.
Can you check this problem?

Hey guys,

Not sure why I am getting this error, but I followed the instructions on this page: http://klusta.readthedocs.io/en/latest/

This is what my terminal looks like:

Any assistance would be much appreciated!

Hello,
I have this issue with kulsta, channels on clusterview doesn't match channels on waveformview. As you can see in the picture, I selected id 3 which is channel one, and on waveformview you can see it shows me channel 19. Can someone help me with that please?
Thanks,
Migle

Hi, I had some issue with klustaviewa. So I just tried to reinstall the whole klusta.
I uninstalled miniconda and reinstall it and tried to install klusta
But I bumped into an error below.

C:\Users\MNS>conda env create -n klusta -f environment.yml

Fetching package metadata ...........
Solving package specifications: .
mkl-2017.0.1-0 100% |###############################| Time: 0:00:27 4.86 MB/s
jpeg-8d-vc14_2 100% |###############################| Time: 0:00:00 4.10 MB/s
python-3.5.2-0 100% |###############################| Time: 0:00:08 3.77 MB/s
tk-8.5.18-vc14 100% |###############################| Time: 0:00:00 5.17 MB/s
zlib-1.2.8-vc1 100% |###############################| Time: 0:00:00 4.89 MB/s
colorama-0.3.7 100% |###############################| Time: 0:00:00 2.44 MB/s
cython-0.25.2- 100% |###############################| Time: 0:00:00 4.07 MB/s
decorator-4.0. 100% |###############################| Time: 0:00:00 2.82 MB/s
hdf5-1.8.15.1- 100% |###############################| Time: 0:00:00 4.32 MB/s
ipython_genuti 100% |###############################| Time: 0:00:00 3.52 MB/s
libpng-1.6.27- 100% |###############################| Time: 0:00:00 4.50 MB/s
libtiff-4.0.6- 100% |###############################| Time: 0:00:00 5.85 MB/s
numpy-1.11.3-p 100% |###############################| Time: 0:00:00 6.75 MB/s
path.py-10.1-p 100% |###############################| Time: 0:00:00 7.49 MB/s
pygments-2.1.3 100% |###############################| Time: 0:00:00 6.65 MB/s
pyparsing-2.1. 100% |###############################| Time: 0:00:00 7.73 MB/s
pytz-2016.10-p 100% |###############################| Time: 0:00:00 8.45 MB/s
setuptools-27. 100% |###############################| Time: 0:00:00 5.88 MB/s
simplegeneric- 100% |###############################| Time: 0:00:00 1.96 MB/s
sip-4.18-py35_ 100% |###############################| Time: 0:00:00 7.59 MB/s
six-1.10.0-py3 100% |###############################| Time: 0:00:00 2.19 MB/s
wcwidth-0.1.7- 100% |###############################| Time: 0:00:00 1.55 MB/s
wheel-0.29.0-p 100% |###############################| Time: 0:00:00 5.97 MB/s
win_unicode_co 100% |###############################| Time: 0:00:00 3.51 MB/s
cycler-0.10.0- 100% |###############################| Time: 0:00:00 1.82 MB/s
h5py-2.6.0-np1 100% |###############################| Time: 0:00:00 7.92 MB/s
pickleshare-0. 100% |###############################| Time: 0:00:00 1.96 MB/s
pip-9.0.1-py35 100% |###############################| Time: 0:00:00 6.64 MB/s
prompt_toolkit 100% |###############################| Time: 0:00:00 7.63 MB/s
python-dateuti 100% |###############################| Time: 0:00:00 8.27 MB/s
qt-4.8.7-vc14_ 100% |###############################| Time: 0:00:10 5.20 MB/s
scipy-0.18.1-n 100% |###############################| Time: 0:00:04 2.77 MB/s
traitlets-4.3. 100% |###############################| Time: 0:00:00 1.55 MB/s
ipython-5.2.2- 100% |###############################| Time: 0:00:00 3.93 MB/s
pyqt-4.11.4-py 100% |###############################| Time: 0:00:00 5.91 MB/s
matplotlib-1.5 100% |###############################| Time: 0:00:01 5.59 MB/s
ERROR conda.core.link:_execute_actions(330): An error occurred while installing package 'defaults::qt-4.8.7-vc14_9'.
UnicodeDecodeError('utf-8', b' 1\xb0\xb3 \xc6\xc4\xc0\xcf\xc0\xcc \xba\xb9\xbb\xe7\xb5\xc7\xbe\xfa\xbd\xc0\xb4\xcf\xb4\xd9.\r\n', 9, 10, 'invalid start byte')
Attempting to roll back.

UnicodeDecodeError('utf-8', b' 1\xb0\xb3 \xc6\xc4\xc0\xcf\xc0\xcc \xba\xb9\xbb\xe7\xb5\xc7\xbe\xfa\xbd\xc0\xb4\xcf\xb4\xd9.\r\n', 9, 10, 'invalid start byte')

C:\Users\MNS>

I use Window 10, 64bit.
And I installed python 3.5 miniconda for 64bit.
Do you have any idea about this error?
Thanks,

Kibong

Hi

I have installed klusta from the web page :

https://github.com/kwikteam/klusta/#quick-install-guide

and tried to open klustaviewa but I get the following error message:

klustaviewa
Traceback (most recent call last):
File "/home/mint/miniconda3/envs/klusta/bin/klustaviewa", line 7, in
from klustaviewa.scripts.runklustaviewa import main
File "/home/mint/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/init.py", line 44, in
from kwiklib.utils import logger as log
File "/home/mint/miniconda3/envs/klusta/lib/python2.7/site-packages/kwiklib/init.py", line 9, in
import dataio
File "/home/mint/miniconda3/envs/klusta/lib/python2.7/site-packages/kwiklib/dataio/init.py", line 4, in
from kwikloader import *
File "/home/mint/miniconda3/envs/klusta/lib/python2.7/site-packages/kwiklib/dataio/kwikloader.py", line 31, in
from .experiment import Experiment
File "/home/mint/miniconda3/envs/klusta/lib/python2.7/site-packages/kwiklib/dataio/experiment.py", line 16, in
from klusta.traces.waveform import WaveformLoader, SpikeLoader
File "/home/mint/miniconda3/envs/klusta/lib/python2.7/site-packages/klusta/traces/init.py", line 9, in
from .spikedetekt import SpikeDetekt
File "/home/mint/miniconda3/envs/klusta/lib/python2.7/site-packages/klusta/traces/spikedetekt.py", line 13, in
from tqdm import tqdm
File "/home/mint/miniconda3/envs/klusta/lib/python2.7/site-packages/tqdm/init.py", line 6, in
from ._main import main
File "/home/mint/miniconda3/envs/klusta/lib/python2.7/site-packages/tqdm/_main.py", line 3, in
from docopt import docopt
ImportError: No module named docopt

Any idea?

Thanks,

Rodrigo.

When extracting spike_times from a .kwik file using KwikModel, the model always displays spike_times for the first channel_group accessed. Setting a new channel_group updates the other measurements (i.e. spike_samples) appropriately, it seems there is no functionality in spike_times(self) to force recomputation after changing channel group.

Our quick fix: removing the if statement forces recomputation on each call of spike_times:

def spike_times(self):
    """Spike times from the current channel_group.

    This is a NumPy array containing `float64` values (in seconds).

    The spike times of all recordings are concatenated. There is no gap
    between consecutive recordings, currently.
    """
    self._spike_times = (self.spike_samples.astype(np.float64) /
                             self.sample_rate)
    return self._spike_times

hi, everyone,
I use the plexon system to recording the neural signal. sorting is very big problem to me. I'm very interesting in klusta. who help me how to convert plexon data to klusta data format.Thanks.

xuefeng

Extracting waveforms : 77%|#######7 | 3398/4404 [22:34<07:05, 2.37it/s]/home/smgxsk1/miniconda3/envs/klusta/lib/python2.7/site-packages/klusta/traces/waveform.py:134: RuntimeWarning: i
nvalid value encountered in double_scalars
s_aligned = np.sum(wave_n_p * u) / np.sum(wave_n_p) + s_min

I've encountered this new bug.

I have been tring to use klusta in a recording with only one channel, how should I set the probe file in this case. I try with
channel_groups = { 0: { 'channels': list(range(1)), 'graph': [ (0,0,), ], etc.

Is it ok?
I have fixed a few issues that I had with the parameters and It looks that it works, do you have a doc with all the parameters explained?

To handle flat binary files with a header.

Command to update metadata in a Kwik file (useful when changing KK2 parameters after spike detection).

$ klusta file.kwik --metadata 
key1=val1
key2=val2
...
$ klusta file.kwik --metadata key2
val2
$ klusta file.kwik --metadata key2=val2
key2=val2
Done.

Does Phy work opening the file remotely?
Before spending time trying to work this out, I was wondering if this is an inherent limitation of phy.

I am currently using Ubuntu 16.04, and I get the following error:
phy: cannot connect to X server localhost:10.0

I guess I can decide the threshold for spike detection in .prm file.
But I don't know what value should I change.
Can you help me for this?
In specific, I want to lower threshold for detecting spike as many as possible.

I'm using KlustaKwik (Masked EM) and Phy GUI to sort data from a 16-channel multi-electrode array. The recordings are saved in 60 second data files and we have 80 minutes of data (80 data files). The 80 files are listed in the PRM config for raw_data_files and the SpikeDetekt and Masked EM algorithms work like a charm.

The issue arises when I use phy kwik-gui to visualize the clusters and waveforms. It appears that the FeatureView scatter plots are misaligning the data points. Namely, it appears that the index of the data files as listed in the KWIK file is:

/recordings/0
/recordings/1
/recordings/11
/recordings/12
/recordings/13
...
/recordings/19
/recordings/2
/recordings/20
/recordings/21
/recordings/22
...

Have others had this issue? Am I missing a parameter/setting/update?

I noticed in spikedetekt that there was parameter called 'save_high' available, which allowed the user to save the high pass filtered data to disk. I would like to use this feature in 'klusta' so that when I run it multiple times, it doesn't have to filter the data again and again. But it seems like this parameter is not accepted in 'klusta' because it throws an error when I try to use. Is there anyway to save the high pass filtered data so it doesn't have to redo it when klusta is run multiple times?

This might seem like a very nube question but i am unsure how to get a DAT file intially to be able to run the program and extract the spikes and cluster them. is there a very detailed tutorial or video or someone willing to explain to me how to get the inital .dat file containing the data (i am using lab chart), as well as the other files that are needed for the analysis to be run.

thanks in advance

Hi,

I receive this error message when opening klustaviewa:

(klusta)milodon-2:session01 rsalazar$ klustaviewa

/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/IPython/qt.py:13: ShimWarning: The IPython.qt package has been deprecated. You should import from qtconsole instead.
"You should import from qtconsole instead.", ShimWarning)
<klustaviewa.gui.mainwindow.MainWindow object at 0x10d4989d0>

Any idea?

After the error message, klustaviewa opens but I cannot load any .kwik file. If I run the following command:

klustaviewa 2016030301.kwik

I get the same error message:

Plus the folowing message in a separate window after klustaviewa is opened:

An error occurred: Traceback (most recent call last):
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/gui/threads.py", line 33, in open
loader.open(path)
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/kwiklib/dataio/kwikloader.py", line 120, in open
self.experiment = Experiment(basename, dir=dir, mode='a')
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/kwiklib/dataio/experiment.py", line 403, in init
self.channel_groups = self._gen_children('channel_groups', ChannelGroup)
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/kwiklib/dataio/experiment.py", line 165, in _gen_children
for child in container
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/kwiklib/dataio/experiment.py", line 440, in init
self.clusters = ClustersNode(self._files, self._node.clusters, root=self._root)
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/tables/group.py", line 813, in getattr
return self._f_get_child(name)
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/tables/group.py", line 683, in _f_get_child
self._g_check_has_child(childname)
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/tables/group.py", line 407, in _g_check_has_child
% (self._v_pathname, name))
NoSuchNodeError: group /channel_groups/1 does not have a child named clusters

Thanks,

Rodrigo.

Hi guys,

I'm new to klusta and python. I successfully installed it. But when I try 'klusta hybrid_10sec.prm', it says 'klusta is not recognized as an internal or exteral command, operable program or batch file.' I already activated klusta. How can I deal with this? Many thanks.

windows 7

E. Quin

Hello!

Is there a way to disable filtering of data during spike detection? I have data already filtered, so filtering it again for spike detection makes no sense :)

Hi,

I just installed klusta on a Linux machine following your instructions. It raised a few issues (I didn't keep track of them).

I tried running klusta on the sample data set. "hybrid_10sec". I got the following error:
ImportError: No module named 'klustakwik2'

I tried to reinstall klustakwik2 using: pip install klustakwik2. I'm attaching the error

I get the following error message at the end:

Failed building wheel for klustakwik2
Command "/home/amanadmin/miniconda3/envs/klusta/bin/python -u -c "import setuptools, tokenize;file='/tmp/pip-build-t3gq66n8/klustakwik2/setup.py';f=getattr(tokenize, 'open', open)(file);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, file, 'exec'))" install --record /tmp/pip-f7pbfsjx-record/install-record.txt --single-version-externally-managed --compile" failed with error code 1 in /tmp/pip-build-t3gq66n8/klustakwik2/

The other messages it pops out are in the attached file:
errormsg.txt

Hi,
i have been using your code on a different comp and it worked very well.
but now that i upgraded my comp and reinstalled everything i get this error:

14:51:16 [I] launch:214 Spike detection done!
14:51:16 [I] launch:239 Starting clustering on shank 0/1.
14:51:16 [I] launch:242 Clustering group 0 (7758 spikes).
14:51:16 [I] launch:122 Starting KK...
ImportError: DLL load failed: The specified module could not be found.

all steps before work really well, but no clustering is made and i cannot open the kwik in the "phy kwik-gui" afterwards.

please help.
thanks,
elor

i was wondering if there is any plan to parallelize klusta in the future?

When first starting out, I grabbed an example .prm file from here, which didn't include the use_single_threshold field. That means that it always defaulted to True, as shown in default_settings.py. I think it would be better to have it default to False.

Some channels are noisier than others. Some are completely faulty. If a noisy channel is included in spike detection, by default it currently does two undesirable things: it unnecessarily increases the detection threshold for all other channels, and it causes lots of false positive detections on the noisy channel. In one particular example, I have a channel whose noise level is maybe 2X that of all the others. With use_single_threshold=True, this particular channel was constantly being triggered off of, greatly slowing down detection and clustering. It was quite extreme. For this 20 min recording I ended up with about 1 million detected spikes on the faulty channel, and only a few 10s of thousands on the others. Setting use_single_threshold=False eliminated the problem, without requiring me to exclude the channel from spike detection.

I've heard that there was a similar setting in the Klusters spike sorting pipeline, and that it defaulted to using a different threshold for each channel.

Thoughts?

21:22:41 [I] launch:214 Spike detection done!
21:22:42 [I] launch:239 Starting clustering on shank 0/1.
21:22:42 [I] launch:242 Clustering group 0 (354282 spikes).
21:22:45 [W] launch:112 Unable to resume KK from /scratch/scratch/smgxsk1/kilonips/20150601/.klustakwik2/spike_clusters.txt, there are 804147 values instead of 354282
21:22:45 [I] launch:122 Starting KK...

There is only one .klustakwik2 folder produced. This is not produced inside a subfolder that is specific to the running job. If klusta is attempted on two files in a single directory - the last clustering is therefore lost.

when using a wide interval i received this warning.

17:59:09 [W] spikedetekt:95 Loading all traces in memory: this will require 15.367 GB of RAM!
17:59:09 [W] spikedetekt:96 To avoid this, do not specify --interval. This bug will be fixed later.

to http://klusta.readthedocs.org/en/latest/

Just the res and clu would be sufficient for now.

/anaconda3/envs/klusta/lib/python2.7/site-packages/IPython/qt.py:13: ShimWarning: The IPython.qt package has been deprecated. You should import from qtconsole instead.
"You should import from qtconsole instead.", ShimWarning)
Cannot mix incompatible Qt library (version 0x40801) with this library (version 0x40805)
Aborted (core dumped):

Can this be b'coz of ubuntu 12.04 version when launching klusta?

Hi,

I have installed 'klusta' using the following instructions (https://github.com/kwikteam/klusta/#quick-install-guide) on a windows and a mac machine.When I run the command:

klusta params.prm --debug

on one of your examples ('https://github.com/klusta-team/example') I keep getting comparable errors messages from both machines:

File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/click/core.py", line 716, in call
return self.main(_args, *_kwargs)
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/click/core.py", line 696, in main
rv = self.invoke(ctx)
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/click/core.py", line 889, in invoke
return ctx.invoke(self.callback, *_ctx.params)
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/click/core.py", line 534, in invoke
return callback(_args, *_kwargs)
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/klusta/launch.py", line 334, in main
return klusta(_args, **kwargs)
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/klusta/launch.py", line 182, in klusta
overwrite=overwrite,
File "/Users/rsalazar/miniconda3/envs/klusta/lib/python2.7/site-packages/klusta/kwik/creator.py", line 395, in create_kwik
assert 'spikedetekt' in prm

Thanks for your help,

Rodrigo.

https://github.com/kwikteam/klusta/blob/master/klusta/traces/waveform.py#L94

If you feed in a data file consisting of zeros, this function returns errors. Probably best to detect (somewhere) that tw != ts and exit with a clean error

The "show all shortcuts" option as well as F1 do not work for me, on my Windows 10 desktop or on my MacBook pro. I reinstalled both. I remember working with an older version of this where there were a ton of shortcuts and keyboard tools -- have these changed? It's been about a year since I was working with klusta.

Since I can't access the list of new shortcuts, can someone send me what they are so I can test them? Thanks!

Dear All,
I'm having a problem that I don't know how to understand,
I'm running klusta on a macbook pro with OSX 10.11.4 (15E65), I followed the instructions to install given on the quick install and tryed with the data and .prm in the documentation. Everything runs, but on the GUI I don't get access to the merge or split options, they are grayed out as seen on the screenshot.
Where did I go wrong?

Many thanks!

it looks like the --interval parameter only accepts an initial value of 0. i get the following error when i put something else.

NotImplementedError: Need to add start to the spike samples

Hello -
I downloaded and created the klusta environment per directions, but when I try to test with the sample data ($ klusta hybrid_10sec.prm) it raises an "ImportError" that reads:

ImportError: dlopen(/Applications/anaconda/envs/klusta/lib/python3.5/site-packages/scipy/sparse/_sparsetools.cpython-35m-darwin.so, 2): no suitable image found.
Did find:
/Applications/anaconda/envs/klusta/lib/python3.5/site-packages/scipy/sparse/_sparsetools.cpython-35m-darwin.so: truncated mach-o error: segment __TEXT extends to 3825664 which is past end of file 3080192""

I am relatively new to the programming world, my experience being limited to simple data extraction/manipulation with pandas/python. I have tried re-installing klusta but get the same error. Any advice on additional steps I could take?

THANK YOU!

I just installed klusta (Ubuntu 15.04) following the quickstart in the docs, and after downloading the test dataset, running klusta on it, I tried to open up klustaviewa but threw this error:

Traceback (most recent call last):
File "/home/---/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/gui/mainwindow.py", line 887, in open_done
self.taskgraph.update_cluster_view()
File "/home/---/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/gui/taskgraph.py", line 75, in
return lambda _args, *kwargs: self.run(('' + name, args, kwargs))
File "/home/---/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/gui/taskgraph.py", line 63, in run
outputs = self.run_single(action)
File "/home/---/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/gui/taskgraph.py", line 52, in run_single
return getattr(self, method)(_args, *_kwargs)
File "/home/---/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/gui/taskgraph.py", line 395, in _update_cluster_view
self.get_view('ClusterView').set_data(**data)
File "/home/---/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/views/clusterview.py", line 554, in set_data
self.model = ClusterViewModel(**kwargs)
File "/home/---/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/views/clusterview.py", line 106, in init
self.load(**kwargs)
File "/home/---/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/views/clusterview.py", line 146, in load
parent=self.get_group(select(cluster_groups, clusteridx)))
File "/home/---/miniconda3/envs/klusta/lib/python2.7/site-packages/klustaviewa/views/clusterview.py", line 455, in get_group
(group.groupidx() == groupidx)][0]
File "/home/---/miniconda3/envs/klusta/lib/python2.7/site-packages/pandas/core/series.py", line 225, in wrapper
if len(self) != len(other):
TypeError: len() of unsized object
None

Any thoughts?
Thanks,

M

I've gotten as far as playing with the sample data provided. I can open the clustered data in the KwikGUI and can click on any single cluster to view the data. However, if I then try to click on a different cluster or change the view in any way the software crashes. A windows message appears stating that "Python has stopped working". I'm running windows 7 and just installed the latest version of Miniconda3.

• Backport the waveform-from-raw-data extractor from phy
• Try to backport the pure Python CCG option from phy to avoid relying on Cython
• New release

hi,

In order to run klustakwik i need to support a feature file with "fet" ending.
I use the sort3D klustakwik interface to do it. There is a button "only write feature file" that does the work.
Is there any matlab script or any script/ automated way to create this file from neuralynx files?

thanks in advanced!

Hi,

following the installation instructions (https://github.com/kwikteam/klusta/#quick-install-guide), I clicked on the link supposed to allow downloading the environment file (https://raw.githubusercontent.com/kwikteam/klusta/master/installer/environment.yml), but all I got was a page with a list of dependencies for klusta - text copied below:

name: klusta
dependencies:

• python
• pip
• numpy
• scipy
• h5py
• six
• cython
• ipython
• matplotlib
• pyqt=4
• pip:

  • pytest

  • requests

  • responses

  • tqdm

  • click

  • traitlets

  • joblib

  • pyopengl

  • vispy

  • klustakwik2

  • klusta

  • phy

  • phycontrib

Am I missing something obvious?

Thanks,
Giovanni

Here is the screenshot. THANKS!