search

tridesclous / tridesclous

offline/online spike sorting. Un logiciel de gauche qui ne fait pas d'amalgame entre la classe et la coquetterie.
http://tridesclous.readthedocs.io
MIT License
52 stars 29 forks source link

Neuralynx file #109

Open beastgit opened 5 years ago

beastgit commented 5 years ago

I am new to python and to spike sorting. I am trying to teach myself spike sorting. I am unable to load a Neuralynx file onto tridesclou. Its in my local desktop (and is named LHu1.ncs). Ould anybody please help me with this basic step. I was going to proceed after that by following one of the example in your documentation.

thanks

samuelgarcia commented 5 years ago

Could you send me the file please. So I could test if it works.

If you want to test tridesclous there are open dataset builtin. See quick start in the doc.

Best

Le jeu. 28 févr. 2019 03:54, beastgit notifications@github.com a écrit :

I am new to python and to spike sorting. I am trying to teach myself spike sorting. I am unable to load a Neuralynx file onto tridesclou. Its in my local desktop (and is named LHu1.ncs). Ould anybody please help me with this basic step. I was going to proceed after that by following one of the example in your documentation.

thanks

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/tridesclous/tridesclous/issues/109, or mute the thread https://github.com/notifications/unsubscribe-auth/AAxyC_qZc_QYMQTK6mhIMm-RmKCcMnXaks5vR0TwgaJpZM4bV5dJ .

beastgit commented 5 years ago

Hi Samuel,

Thanks a lot from your quick reply. you can download from https://www.dropbox.com/sh/8pitvkmoa7hnymx/AADO_OEh3FNn3Q0JfDPa_FNXa?dl=0

Haris

On Thu, Feb 28, 2019 at 1:41 AM Garcia Samuel notifications@github.com wrote:

Could you send me the file please. So I could test if it works.

If you want to test tridesclous there are open dataset builtin. See quick start in the doc.

Best

Le jeu. 28 févr. 2019 03:54, beastgit notifications@github.com a écrit :

I am new to python and to spike sorting. I am trying to teach myself spike sorting. I am unable to load a Neuralynx file onto tridesclou. Its in my local desktop (and is named LHu1.ncs). Ould anybody please help me with this basic step. I was going to proceed after that by following one of the example in your documentation.

thanks

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/tridesclous/tridesclous/issues/109, or mute the thread < https://github.com/notifications/unsubscribe-auth/AAxyC_qZc_QYMQTK6mhIMm-RmKCcMnXaks5vR0TwgaJpZM4bV5dJ

.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/tridesclous/tridesclous/issues/109#issuecomment-468156747, or mute the thread https://github.com/notifications/unsubscribe-auth/AodajBaw-Ys6DeXnbZ6dmKEeNItkKCJoks5vR3n9gaJpZM4bV5dJ .

samuelgarcia commented 5 years ago

Thank you for this file.

The neuralynx file is not form cheetah but pegasus. It was not handle in python-neo. So I made a patch.

You should clone this branch to test it: https://github.com/NeuralEnsemble/python-neo/pull/655

Note that neuralynx is not file base but dirrectory based. This mean that the file you send me is only one channel. In tridesclous you have to select the directory that inlcude that file ncs, file.

Another problem that is not solve at the moment is that for neuralynx system you can have packet loss and so gaps in the signals recording. In neo, theses gaps are detected and each time there is a gap, then a new segment is construct. In present file, there are 300 gaps. I don't known if is true or a bug.

Could you open an issue here https://github.com/NeuralEnsemble/python-neo And send this kind of file but a smaller one that could be included in test suite.

Please mention this issue in the other issue.

beastgit commented 5 years ago

Hi Samuel, thanks for the prompt reply. I will check your suggestions and get back to you. best

On Fri, Mar 1, 2019 at 4:05 AM Garcia Samuel notifications@github.com wrote:

Thank you for this file.

The neuralynx file is not form cheetah but pegasus. It was not handle in python-neo. So I made a patch.

You should clone this branch to test it: NeuralEnsemble/python-neo#655 https://github.com/NeuralEnsemble/python-neo/pull/655

Note that neuralynx is not file base but dirrectory based. This mean that the file you send me is only one channel. In tridesclous you have to select the directory that inlcude that file ncs, file.

Another problem that is not solve at the moment is that for neuralynx system you can have packet loss and so gaps in the signals recording. In neo, theses gaps are detected and each time there is a gap, then a new segment is construct. In present file, there are 300 gaps. I don't known if is true or a bug.

Could you open an issue here https://github.com/NeuralEnsemble/python-neo And send this kind of file but a smaller one that could be included in test suite.

Please mention this issue in the other issue.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/tridesclous/tridesclous/issues/109#issuecomment-468594939, or mute the thread https://github.com/notifications/unsubscribe-auth/AodajCOjyEMaHtf17XruBlV3ZmUxHGZDks5vSO1IgaJpZM4bV5dJ .

samuelgarcia commented 5 years ago

Hi @beastgit : This is fixed but to close this issue we need a test file that could be public and small (<10Mo). See https://github.com/NeuralEnsemble/python-neo/pull/655

best