Skip to content
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

version 0.2.22 #62

Merged
merged 114 commits into from
Apr 14, 2024
Merged

version 0.2.22 #62

merged 114 commits into from
Apr 14, 2024

Conversation

matthias-k
Copy link
Owner

No description provided.

matthias-k and others added 30 commits May 30, 2022 23:09
Signed-off-by: Matthias Kümmmerer <[email protected]>
Signed-off-by: Matthias Kümmmerer <[email protected]>
Signed-off-by: Matthias Kümmerer <[email protected]>
Signed-off-by: Matthias Kümmerer <[email protected]>
…t in FixationTrains instances

Signed-off-by: Matthias Kümmmerer <[email protected]>
Signed-off-by: Matthias Kümmmerer <[email protected]>
Signed-off-by: Matthias Kümmmerer <[email protected]>
Signed-off-by: Matthias Kümmmerer <[email protected]>
Signed-off-by: Matthias Kümmmerer <[email protected]>
Signed-off-by: Matthias Kümmmerer <[email protected]>
Signed-off-by: Matthias Kümmmerer <[email protected]>
matthias-k and others added 29 commits November 9, 2023 23:58
#34)

* Bugfix: NUSEF fixation locations often not correctly scaled to image coordinates
* NUSEF: Don't include fixation data for segmention-mask only images, remove empty images
* keep source file
Also:
* add pyproject.toml and adapt test workflow
---------

Signed-off-by: Matthias Kümmmerer <[email protected]>
Add new filter functions: filter_stimuli_by_attribute, filter_scanpaths_by_attribute, filter_fixations_by_attribute, filter_scanpaths_by_length
Signed-off-by: Matthias Kümmmerer <[email protected]>
* Make UniformModel more efficient

Signed-off-by: Matthias Kümmmerer <[email protected]>

* Verbosity parameter in CrossvalMultipleRegularizations

Signed-off-by: Matthias Kümmmerer <[email protected]>

---------

Signed-off-by: Matthias Kümmmerer <[email protected]>
Signed-off-by: Matthias Kümmmerer <[email protected]>
The functionally is very simple and should, IMO, not be automatic.

Signed-off-by: Matthias Kümmmerer <[email protected]>
* Added filter to remove stimuli with no fixations

* Update test_filter_datasets.py

* Update test_filter_datasets.py
* Added filter to remove stimuli with no fixations

* Update test_filter_datasets.py

* Update test_filter_datasets.py

* Added remove stimuli without fixations to dataset_config.py
* Merge branch 'dev' of github.com:naman0210/pysaliency into dev

* Update test_precomputed_models.py

* Update tests/test_precomputed_models.py

added comment

* Update tests/test_precomputed_models.py

---------

Co-authored-by: Harneet Singh Khanuja <[email protected]>
Co-authored-by: matthias-k <[email protected]>
use integer indices instead of binary masks in image-subject-crossval.

Signed-off-by: Matthias Kümmmerer <[email protected]>
…here in a common sub directory

Signed-off-by: Matthias Kümmmerer <[email protected]>
…timuli

Bugfix hdf5 filenames for substimuli
Signed-off-by: Matthias Kümmmerer <[email protected]>
* Stimuli can be filtered by boolean masks

Signed-off-by: Matthias Kümmmerer <[email protected]>

* fix typo

Signed-off-by: Matthias Kümmmerer <[email protected]>

---------

Signed-off-by: Matthias Kümmmerer <[email protected]>
The CAT2000 dataset comes with additional saliency maps (I think AIM?)
in subdirectories. For CAT2000 test they already have been removed
as part of the pysaliency import, but for the train dataset this
has been forgotten. This is now fixed, also there is a test for
CAT2000_train v1.1 added.

Signed-off-by: Matthias Kümmmerer <[email protected]>
* Cleanup imports

Signed-off-by: Matthias Kümmerer <[email protected]>

* BUG: AUC with average=image fails if there are images without fixations"

Now images without fixations are ignored, in consistency with how
it is handled in information_gain and NSS.

Signed-off-by: Matthias Kümmerer <[email protected]>

* Ignore log warnings in uniform model

Signed-off-by: Matthias Kümmerer <[email protected]>

* ENH: avoid computing saliency maps for AUC on images without fixations

Signed-off-by: Matthias Kümmerer <[email protected]>

---------

Signed-off-by: Matthias Kümmerer <[email protected]>
Signed-off-by: Matthias Kümmerer <[email protected]>
Signed-off-by: Matthias Kümmerer <[email protected]>
@matthias-k matthias-k merged commit e87966a into master Apr 14, 2024
8 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants