v0.9.1

What's Changed

• Update changes link by @thorstenwagner in #98
• Update installation.rst by @YousefMetwally in #99
• Single forward run for embedding of anchor, negative and positive volumes. by @YousefMetwally in #103
• Update installation.rst by @YousefMetwally in #104
• Update torchtrainer.py by @YousefMetwally in #105
• Update cuml and pandas by @thorstenwagner in #111
• change default neighbors to 64 by @thorstenwagner in #112

New Contributors

• @YousefMetwally made their first contribution in #99

Full Changelog: v0.9.0...v0.9.1

v0.9.0

With TomoTwin v0.9.0 we introduce a new target manager for napari. The documentation of the clustering workflow has been updated accordingly. Also, locate is now parallelized using dask and is now much faster :-)

What's Changed

• Update rapids to 23.10 by @thorstenwagner in #48
• Update pytorch-metric-learning by @thorstenwagner in #87
• Locate is now using dask by @thorstenwagner in #90
• Remove tomotwin from env file by @thorstenwagner in #91
• Switch to SPDX-License-Identifiers by @kephale in #96
• Fast locate, new rapids and updated tutorial by @thorstenwagner in #89

New Contributors

• @kephale made their first contribution in #96

Full Changelog: v0.8.0...v0.9.0

v0.8.0

What's Changed

This release mainly improves the user experience when working with the clustering workflow. The changes you will notice are:

• BREAKING CHANGE: When calculating a umap, no label mask is created anymore.Instead, this map is now computed in the background within napari and is transparent to the user. This also means that you don't have to select a label layer anymore when loading a umap (PR #82 MPI-Dortmund/napari-tomotwin#22). However, this is a breaking change, so you need to recalculate your umaps to take advantage of it.
• Napari no longer freezes when loading umaps and shows a progress bar while loading (MPI-Dortmund/napari-tomotwin#20)
• When selecting clusters in the umap, there is a time delay between the selection and the actual highlighting. Now this process is 2x-3x faster (PR BiAPoL/napari-clusters-plotter#283 BiAPoL/napari-clusters-plotter#282)
• When saving cluster targets, the appropriate embedding file is automatically selected if possible (PR MPI-Dortmund/napari-tomotwin#26). This means that the user does not have to remember which embedding belongs to which umap.
• Saving cluster targets (medoids) is now 5x faster (MPI-Dortmund/napari-tomotwin#23).

Minor changes

• Add median filter strategy by @GavinR1 in #76 and #79
• Stop making tomotwin_tools crash on non gpu machines. by @thorstenwagner in #78
• Add short training tutorial by @thorstenwagner in #81

Full Changelog: 0.7.0...v0.8.0

v0.8.0b1

What's Changed

Breaking changes:

• Write embeddings metadata + positions into umap attributes @thorstenwagner in #82
• Remove calculation of label mask by thorstenwagner in #82

The above change have the effect, that already calculated umaps to not work anymore. However, you can easily recreate them with tomotwin_tools.py umap. With argument -m you can even provide the old fitted model. In that way you will get the same umap as before.

Minor changes

• Add median filter strategy by @GavinR1 in #76 and #79
• Stop making tomotwin_tools crash on non gpu machines. by @thorstenwagner in #78
• Add short training tutorial by @thorstenwagner in #81

Full Changelog: 0.7.0...v0.8.0b1

v0.7.0

What's Changed

With this release we add new tools for advanced strategies that describe how to speed up the calculation of embeddings, potentially improve their umaps, and refine references/targets!

Improve reference/targets with umaps (Strategy 1)

Add new umap and filter options by @thorstenwagner in #64

Faster embeddings with median filtering (Strategy 2)

Add fast median mask estimation by @thorstenwagner in #66

It can dramatically speed up the embedding step compared to using no masking, and it also outperforms the previous method based on intensity heuristics:

Other

• increase timeout of the dataloader to 180 seconds. by @thorstenwagner in #57
• --save_after_improvement saves models after each epoch by @thorstenwagner in #60
• Check & correct for duplicate embedded subvolumes by @thorstenwagner in #61
• Fix padding when creating heatmaps during locate with --write_heatmaps by @thorstenwagner in #65

Documentation

• Add strategies pages to the documentation and link them by @thorstenwagner in #67 #68 #69 #70 #71

Full Changelog: v0.6.1...0.7.0

v0.6.1

What's Changed

• Use random port for distributed processing by @thorstenwagner in #54
• Use stable pytorch 2.1 by @thorstenwagner in #52
• Fix read the docs building by @thorstenwagner in #55
• Fix 'too many open files' problem by @thorstenwagner in #56

Full Changelog: v0.6.0.3...v0.6.1

v0.6.0.1

What's Changed

• tomotwim_embed.py is now 1.6x faster and linearly scales across multiple GPUs by @thorstenwagner in #40
  • Exploiting new the compile option of the latest pytorch 2.1 nightly build.
  • Internally DistributedDataParallel is used instead of DataParallel
• Unpin pytorch version by @thorstenwagner in #46

Full Changelog: v0.5.1...v0.6.0.1

v0.5.0

What's Changed

TomoTwin

• Add option to specify a mask during embedding by @thorstenwagner in #27
• Accept key error when evaluation is running and use more flexibel label idenetification by @thorstenwagner in #32
• Fix bug when extracting label from filename by @thorstenwagner in #33

Napari tools

• napari-tomotwin has been updated to version 0.2 to add medoid cluster target calculation by @thorstenwagner . Learn more at https://github.com/MPI-Dortmund/napari-tomotwin/releases/tag/v0.2.0
• napari-boxmanager has been updated to version 0.4 by @thorstenwagner. Learn more at https://github.com/MPI-Dortmund/napari-boxmanager/releases/tag/v0.4.0

Documentation

• Added citation file (citation.cff) to nature methods by @thorstenwagner in #22
• Added some developer information @thorstenwagner in #26
• Added documentation for masking by @thorstenwagner in #28
• Updated changelist for version 0.5 by @thorstenwagner in #34
• Various updates by @GavinR1 and @thorstenwagner in #23 #30 #31 #29

Full Changelog: v0.4.3...v0.5.0

v0.4.3

What's Changed

• Add the citation.cff by @thorstenwagner in #18
• Automatic release by GitHub actions on pypi by @thorstenwagner in #19
• Updated dependencies and new installation instructions by @thorstenwagner in #21

Full Changelog: v0.4.2.2...v0.4.3

TomoTwin v0.4.2

This is the official clustering workflow release :-) We updated the tutorial and added all the necessary tools. Reinstall to get all the necessary tools!

What's Changed

• Make pytest and coverage running by @thorstenwagner in #8
• Refactor locate and add unit tests by @thorstenwagner in #9
• Added more unit tests, documentation and did some refactoring by @thorstenwagner in #10
• Clustering workflow release by @thorstenwagner in #11
• add changes for 0.4.0 by @thorstenwagner in #12
• update installation instructions by @thorstenwagner in #13

Full Changelog: v0.3.0...v0.4.2