add plot_npc and plot_active_latent_units #9
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
leander: check z / zmean defaul handling for VAEs |
Done. Now waiting for code cleanup form martins side. |
le-ander
approved these changes
Nov 6, 2020
|
@davidsebfischer I would like to merge this branch, however it says "Merging is blocked" above. Does this have to do with the branch protection you enabled? |
|
Thanks for the note, this is because of the required signed commits, this wouldn't be an issue if we merged via a dev branch as previously imagined, I am deactivating this for now again so that merging should be possible again. |
davidsebfischer
added a commit
that referenced
this pull request
Dec 10, 2020
* add plot_npc and plot_active_latent_units (#9) * add plot_npc and plot_active_latent_units * make sure handling of z and z_mean is consistent for VAE embeddings * clean up and documentation * formatting Co-authored-by: Martin König <martin.koenig@ymail.com> Co-authored-by: le-ander <20015434+le-ander@users.noreply.github.com> * added data loader for interactive workflows with unprocessed data * made cell type loading optional in dataset .load() * enabled usage of type estimator on data without labels in prediction mode * recursively search custom model repo for weights files * sort model lookuptable alphabetically before writing it * make sure mode_path is set correctly in model_lookuptable when recursive weights loading is used * fix os.path.join usage in dataloaders * replace path handling through string concatenations with os.paths.join and f-strings * fix bug in lookup table writing * add mdoel file path to lookup table * reset index in model lookuptable before saving * add method to user interface for pushing local model weights to zenodo * fix bug in user interface * fix bux in summaries.py * use absolute model paths when model_lookuptable is used * fix bug in pretrained weights loading * fix bug in pretrained weights loading * automatically create an InteractiveDataset when loading data through the UI * fix bug inUI data loading * Explicitly cast indices and indptr of final backed file to int64. (#17) For the background on this: scverse/anndata#453 * update human lung dataset doi * align mouse organ names with human organ names * fix typo in trachea organ naming in mouse * rename mouse ovary organ to femalegonad * rename mouse ovary organ to femalegonad * sort by model type in classwise f1 heatmap plot * another hacky solution to ensure a summary tab can be created when both vae and other models are loaded at once * allow custom metadata in zenodo submission * do not return doi but deposit url after depositing to zenodo sandbox as dois don't wrk on sandbox * updated model zoo description * recognise all .h5 and .data-0000... files as sfaira weights when constructing lookuptable * Update README.rst * Add selu activation and lecun_normal weight_init scheme for human VAEVAMP. (#19) * update sfaira erpo url and handle .h5 extension in model lookuptable id * add meta_data download information to all human dataloaders * updated docs * updated reference to README in docs * updated index * included reference to svensson et al data base in docs * fixed typo in docs * fixed typos in docs * restructured docs * fixed bug in reference roadmap in docs * updated data and model zoo description * added summary picture into index of docs * fixed typo in docs * updated summary panel * add badges to readme and docs index * updated summary panel (#20) * Doc updates (#21) * updated summary panel * fixed concept figure references * Doc updates (#22) * updated zoo panels * move from `import sfaira.api as sfaira` to `import sfaira` and from `import sfaira_extension.api as sfairae` to `import sfaira_extension` * add custom genomes to sfaira_extension * fix loading of custom topology versions from sfaira_extension * fix circular imports between sfaira_extension and sfaira * fix dataloader * fix celltype versioning through sfaira_extension * fix celltype versioning through sfaira_extension * formatting * Doc updates (#25) * added mention of download scripts into docs Co-authored-by: mk017 <martin.koenig@tum.de> Co-authored-by: Martin König <martin.koenig@ymail.com> Co-authored-by: le-ander <20015434+le-ander@users.noreply.github.com> Co-authored-by: Abdul Moeed <abdulmoeed444@gmail.com>
davidsebfischer
added a commit
that referenced
this pull request
Dec 10, 2020
* add plot_npc and plot_active_latent_units (#9) * add plot_npc and plot_active_latent_units * make sure handling of z and z_mean is consistent for VAE embeddings * clean up and documentation * formatting Co-authored-by: Martin König <martin.koenig@ymail.com> Co-authored-by: le-ander <20015434+le-ander@users.noreply.github.com> * added data loader for interactive workflows with unprocessed data * made cell type loading optional in dataset .load() * enabled usage of type estimator on data without labels in prediction mode * recursively search custom model repo for weights files * sort model lookuptable alphabetically before writing it * make sure mode_path is set correctly in model_lookuptable when recursive weights loading is used * fix os.path.join usage in dataloaders * replace path handling through string concatenations with os.paths.join and f-strings * fix bug in lookup table writing * add mdoel file path to lookup table * reset index in model lookuptable before saving * add method to user interface for pushing local model weights to zenodo * fix bug in user interface * fix bux in summaries.py * use absolute model paths when model_lookuptable is used * fix bug in pretrained weights loading * fix bug in pretrained weights loading * automatically create an InteractiveDataset when loading data through the UI * fix bug inUI data loading * Explicitly cast indices and indptr of final backed file to int64. (#17) For the background on this: scverse/anndata#453 * update human lung dataset doi * align mouse organ names with human organ names * fix typo in trachea organ naming in mouse * rename mouse ovary organ to femalegonad * rename mouse ovary organ to femalegonad * sort by model type in classwise f1 heatmap plot * another hacky solution to ensure a summary tab can be created when both vae and other models are loaded at once * allow custom metadata in zenodo submission * do not return doi but deposit url after depositing to zenodo sandbox as dois don't wrk on sandbox * updated model zoo description * recognise all .h5 and .data-0000... files as sfaira weights when constructing lookuptable * Update README.rst * Add selu activation and lecun_normal weight_init scheme for human VAEVAMP. (#19) * update sfaira erpo url and handle .h5 extension in model lookuptable id * add meta_data download information to all human dataloaders * updated docs * updated reference to README in docs * updated index * included reference to svensson et al data base in docs * fixed typo in docs * fixed typos in docs * restructured docs * fixed bug in reference roadmap in docs * updated data and model zoo description * added summary picture into index of docs * fixed typo in docs * updated summary panel * add badges to readme and docs index * updated summary panel (#20) * Doc updates (#21) * updated summary panel * fixed concept figure references * Doc updates (#22) * updated zoo panels * move from `import sfaira.api as sfaira` to `import sfaira` and from `import sfaira_extension.api as sfairae` to `import sfaira_extension` * add custom genomes to sfaira_extension * fix loading of custom topology versions from sfaira_extension * fix circular imports between sfaira_extension and sfaira * fix dataloader * fix celltype versioning through sfaira_extension * fix celltype versioning through sfaira_extension * formatting * Doc updates (#25) * added mention of download scripts into docs Co-authored-by: mk017 <martin.koenig@tum.de> Co-authored-by: Martin König <martin.koenig@ymail.com> Co-authored-by: le-ander <20015434+le-ander@users.noreply.github.com> Co-authored-by: Abdul Moeed <abdulmoeed444@gmail.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.