cyhsm / deepinsight Goto Github PK

Hello, I tried running the ephys example notebook you provided but am getting an installation error. I'm getting a similar error when I try to install it locally via anaconda.
input: !pip install -e git+https://github.com/CYHSM/DeepInsight.git

Output:
Obtaining DeepInsight from git+https://github.com/CYHSM/DeepInsight.git#egg=DeepInsight
Cloning https://github.com/CYHSM/DeepInsight.git to ./src/deepinsight
Running command git clone --filter=blob:none --quiet https://github.com/CYHSM/DeepInsight.git /content/src/deepinsight
Resolved https://github.com/CYHSM/DeepInsight.git to commit e5a66be
Preparing metadata (setup.py) ... done
Collecting tensorflow-gpu (from DeepInsight)
Downloading tensorflow-gpu-2.12.0.tar.gz (2.6 kB)
error: subprocess-exited-with-error

× python setup.py egg_info did not run successfully.
│ exit code: 1
╰─> See above for output.

note: This error originates from a subprocess, and is likely not a problem with pip.
Preparing metadata (setup.py) ... error
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

Hi there,

I'm facing an error when the training starts (deepinsight.train.run_from_path)

....

error:
OSError: Unable to open file (unable to lock file, errno = 35, error message = 'Resource temporarily unavailable')

I did change the tensorflow version in the "requirements.txt" when installing it
tensorflow==1.13.0-rc1
I'm not sure if that could be the issue.

Any insights?
Thanks in advance

Hi Markus,
I really liked this paper, and want to give it a try with other datasets.
I have two questions:

1. Can it decode time-compressed representations, like replay in the hippocampus? assuming a constant compression factor (but maybe unknown value). If not, do you think the model can be extended to handle it?
2. How sensitive is it to (slow) drifts in the neural activity? i.e. slow changes in the spikes amplitude.
   Thanks a lot! Tamir :)

Hi Markus,
Would it be possible to get the sample .nwb file or some explanation about the data structure expected by DeepInsight (for the channels, timestamps, position, etc)? I have data from a different recording system (Neuralynx) and would be very interested in trying to use it.

Thanks and happy holidays,
Elhanan

When I try to run the available example codes in Google Colab i get the following errors:
UnimplementedError: 2 root error(s) found.
(0) UNIMPLEMENTED: DNN library is not found.
[[{{node time_distributed/conv_tsr0/Conv2D}}]]
[[loss/AddN/_519]]
(1) UNIMPLEMENTED: DNN library is not found.
[[{{node time_distributed/conv_tsr0/Conv2D}}]]
0 successful operations.
0 derived errors ignored.

Meanwhile it is running fine on my personal laptop. Do you know any solution for this problem?

Thank you for this interesting paper and work!
I had a question regarding the decoder here coded:
Is it correct that your model takes as input wavelet powers for a temporal window "T=64 (corresponding to 2.13s)" (which are in fact down-sampled M=1000 times over the temporal dimension from the original wavelet powers matrix), and output one value for each behavioral variable?
Should it not decode 64*(nb of behavioral variable) values, one for each time-step, since over 2.13s the variable can really change a lot (for example, for the head-direction speed of a mouse, an order of magnitude is 40deg/s).

Thank you for your help!

Hi Dr. Frey,
Thank you for your article. As your results are amazing I wanted to give Deepinsight a try on my lab data. However, I encouter an issue concerning the preprocessing of the input data.
After running the deepinsight.preprocess.preprocess_input( ) function with our electrophysiology data (which have a 30.000Hz sampling rate), the result of the wavelet transform is unexpected. The frequency bands used to compute the wavelet transform return as follows:

deepinsight.preprocess.preprocess_input(fp_deepinsight, input_data, sampling_rate=sampling_rate, channels=channels) 
hdf5_file = h5py.File(fp_deepinsight, mode='r')
frequencies = np.round(hdf5_file['inputs/fourier_frequencies'], 3)
print(list(frequencies))

It returns:
[inf, inf, inf, inf, inf, inf, inf, inf, inf, inf, inf, inf, inf, inf, inf, inf, 58.6, 41.44, 29.3, 20.72, 14.65, 10.36, 7.324, 5.18, 3.662, 2.59]

I assume that the wavelet transform functions automatically determines the best frequency bands to perform the transform, so I do not understand the origin of these "inf" values.
Do you have any idea about what is wrong, or on how to constrain the frequency bands ?

Thank you in advance,
Allan Muller

Hi,
The file in the jupyter notebook:
!wget https://ndownloader.figshare.com/files/20150468 -O ./example_data/processed_R2478.h5
is not available