New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add no StatsInfo option to importer #3580
Add no StatsInfo option to importer #3580
Conversation
In cases where a vast amount of data is being imported minimum and maximum (StatsInfo) calculation can be prohibitive. This commit allows for disabling this calculation and population if the Bio-Formats reader does not already have the metadata available to it.
To import images: use |
both insight and web work fine. Removing breaking |
The only case with the |
Tested on a large SPIM dataset with |
Thanks @chris-allan. Merging this and I will open a follow-up PR for additional pixel checksum disabling. |
Add no StatsInfo option to importer
--no-rebase |
Is there (or is it worth) a card somewhere for being able to cause calculation of this stuff in the background for already imported images? |
There is no card, the only please at the moment we use the stats info is at the rendering engine level. |
There is no card, the only place at the moment we use the stats info is at the rendering engine level. |
@mtbc: this point was partially discussed in ome/omero-documentation#1152 while adding the limitation. I share @jburel's thoughts, we need to drive this only if there is a concrete well-defined need from a particular community. |
This PR adds a
--no_stats_info
option to OMERO.importer to allow for the disabling of minima and maxima calculation when not present as part of the Bio-Formats reader metadata. This is of particular benefit where the data volume being imported is extremely high (10s of GB) and the person importing the data is aware of the tradeoffs that apply. Under such conditions calculating global minima and maxima can take hours. We have already applied a similar strategy to images of the digital pathology domain.The significant bits populated as part of the import process are now being used to inform the default dynamic range of the rendering engine if such metadata is available. Furthermore, certain areas of the code that did not populate the bit width when creating a
PixelsType
have been fixed. At least one area where the pixels type name was being compared incorrectly has also been fixed.Unit and integration tests cover the semantic implementation outlined here. Functional testing will have to be discussed.
/cc @emilroz