0.3.3

Improvements

• MD images can now be uploaded and processed with TM
• Boost of performance in case many features are selected. Reduced size of test sets

Bug fixes

• Added default channel value to 1. This is general, not only for MD images but valid for all microscope type.

0.3.2

Improvements

• expose created_at and updated_at columns of tasks table via REST API
• separate model training and prediction for classifier tools to reduce memory usage and allow parallelization of label prediction
• speed up insertion of tool results into database by targeting individual shards directly
• no longer serialize type of workflow step arguments and enforce types rather than performing type checks
• implement ORM interface for distributed tables to allow modification of distributed tables via ExperimentSession (this further abstracts table distribution; users can now handle distributed tables the same way they would handle any other table)

0.3.1

Limit number of submitted jobs based on the number of available cores.

0.3.0

Speed improvements:

• change partition strategy for distributed tables (new column partition_key to distribute objects by site, well or plate)
• improve performance for bulk ingestion of segmented objects and feature data by using COPY rather than INSERT
• remove table site_intersection for better normalisation (the corresponding columns have been moved to sites table and renamed for clarity)

General Improvements:

• Improve handling of arguments memory and cores for job submission
• give tool jobs more resources by default (they get an entire node allocated)
• Optimization of default values for pipeline
• Small improvements to metaextract
• No longer compute aggregates in jterator collect (reversing decision from 0.2.2)
• Automatically create Slurm account via tm_add script
• Distribute sites randomly across jterator jobs

Bug Fixes:

• Fix stitching bugs in pyramid building (and added some unit tests to this)
• Fix bug related to metadata config for cellvoyager
• Improve setting threshold levels for object visualization

0.2.2

Extending jterator collect step such that it computes aggregates for different mapobject types
Adding method for spatial filtering of mapobjects
Fixing detection of plate layout for high number of plates and very low numbers of plates
Allowing selection of subset of features in spatial query
Fixing status bug in the interface
Adding metamorph to supported microscope types
Rounding feature values to single precision (6 decimals)
Simplifying query for retrieving mapobjects per site

0.2.1

Updates to dependencies and bugfixes

0.2.0

Performance improvements through support for a distributed database.

Distribution of tables channel_layer_tiles, mapobjects, mapobject_segmentations, feature_values and label_values.

Store feature values in HSTORE column rather than using EAV model.

Removal of Spark support for data analysis tools.

Representation of task hierarchy in database tables.

Sequential processing of corilla and align steps in "multiplexing" workflows.

Simplified implementation of custom microscope reader classes.

0.1.1

Release notes

Bug fixes and improvements related to handling of image metadata in the metaconfig workflow step.

Changes are fully backwards compatible.

Acknowledgements

I would like to thank all users for their efforts in testing and bug reporting, in particular
@alecrimi
@SaadiaIftikhar
@lalil0u
@jluethi

0.1.0 (Rigi)

Release notes

Initial beta release.