Open aaron-cognitiv opened 6 years ago
Thanks @aaron-cognitiv for using our package and reporting your issues! I am happy that the BioformatsReader works for your Olympus files. Concerning your points
I have made a PR to update to 5.4.1 (#266) however this gives issues with different file types (nd2). I silently hope that this issue will go away in a future (or current?) version of Bioformats. I will check up on that.
Memory issues can occur using the java interface, because it defines a default heap size of 512 MB. The easiest option is increasing the heap size using BioformatsReader(<filename>, java_memory='1024m')
or more.
Thanks
Perhaps 5.6 has solved the issue.
Unfortunately, the memory exception is a red-herring. Increasing the heap size does not help, and I can successfully open much larger files with current heap size. The problem seems to be that the bioformats reader does work when opening .oir sub-files. I'm communicating on the bioformats forum in hopes of getting this resolved.
Tagging https://github.com/openmicroscopy/bioformats/issues/2955
1) I'm interested in loading .oir files using your BioformatsReader. OME Bioformats added support for Olympus's .oir files with version 5.5. By upgrading the loci_tools.jar file that Pims uses, I was able to successfully read most .oir files. Therefore, I wonder if you can officially upgrade the bioformats version that pims downloads/uses?
2) When .oir files reach a size larger than 1Gb, Olympus automatically begins splitting the file into subfiles (i.e. 'datafile.oir' will contain time points 0 through 399, then datafile_00001 will contain time points 400 through 799, etc...). However, when I attempt to use the BioformatsReader to load a subfile, I receive an Exception:
This exception appears to relate to the use of the ChannelSeparator object to parse the subfile, since the python-bioformat's ImageReader object can successfully load the subfiles.
P.S. Pims is a great package. Thank you.