4

I'm trying to follow the example of brain observatory ipython notebook.

However, I became stuck loading the nwb file like below.

from allensdk.core.brain_observatory_cache import BrainObservatoryCache
boc = BrainObservatoryCache(manifest_file='boc/manifest.json')

data_set = boc.get_ophys_experiment_data(501940850)  # problem here

So, I opened the nwb file by HDFview.

All of the brain observatory nwb files were not opened except for 502376461.nwb.

It threw the following error:

IOError: Unable to open file (Truncated file: eof = 82280448, sblock->base_addr = 0, stored_eoa = 204046519)

When I tried to open the 502376461.nwb in the ipython notebook example from allen, it worked!! But the others (501940850, 503820068...) failed like above.

Robin De Schepper
  • 4,942
  • 4
  • 35
  • 56
Hyunsu Lee
  • 41
  • 2

1 Answers1

3

Summarizing the thread from github:

https://github.com/AllenInstitute/AllenSDK/issues/22

The files were partially downloaded or corrupted somehow. No exceptions were reported during the download, so urllib must not have noticed a problem.

AllenSDK developers are investigating some sort of file consistency check and/or a different HTTP library.

https://github.com/AllenInstitute/AllenSDK/issues/28

If others run into this, you can delete the bad file and re-run the download function (BrainObservatoryCache.get_ophys_experiment_data). Files are downloaded into a subdirectory of the BrainObservatoryCache manifest file, which defaults to the current working directory if unspecified.

davidf
  • 313
  • 1
  • 6