search

CRIMAC-WP4-Machine-learning / CRIMAC-annotationtools

This repository contatin tools to read, convert and write annotatons from fisheries acoustics
MIT License
1 stars 2 forks source link

issue occured when running step 3 #5

Open YassirAnas opened 1 year ago

YassirAnas commented 1 year ago

Thank you so much, Tomasz, and to the other contributors for developing this package and making it available for us to preprocess the challenging acoustic data. Using this package to process a sample of data generated by LSSS, I am able to run the two first steps ( generating the parquet file and Sv gridded data ); however, I am still struggling with the third step. From my perspective, I thought it might be a problem in LSSS annotations (work files). The command of running step 3 is rendering the following issue for each work file, and here is an example:

2017401-D20170509-T073827.work

/dataout/S2020842_sv.zarr Reading:/workin/2017401-D20170509-T073827.work --- labels from zarr --- /workin/2017401-D20170509-T073827.work /dataout/S2020842_sv.zarr /dataout/S2020842_sv.zarr /dataout/S2020842_ping_time-raw_file.parquet Exception type: <class 'AttributeError'> File name: /usr/local/lib/python3.8/site-packages/annotationtools/crimactools/parseworkfiles.py Line number: 115

Any kind of help from your side will be much appreciated. Thank you so much

tomaszfurmanek commented 11 months ago

We are currently working with a new version that will use LSSS to convert the workfiles to the newest version. step 3 is only tested for some surveys that use schoolboxes for annotation Exception type: <class 'AttributeError'> indicates that your work file does not have the correct attribute (XML tag) compared to the work files we use in our script. This may be solved when we make a new version that will use LSSS to read the work files and convert to the latest version