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

Mask resampling + binary mask option #142

Closed
wants to merge 5 commits into from
Closed

Conversation

vincentsarago
Copy link
Member

This is a second try to fix #105 and also makes alpha/mask band read more consistent for all dataset.

This is a breaking change and will need to be tested before release

@vincentsarago
Copy link
Member Author

vincentsarago commented Feb 17, 2020

Note: benchmarks show a slight speed decrease (~10ms) but my guess it that bilinear resampling is a bit slower then nearest.

☝️ confirmed, performances are equivalent to master values

vincentsarago added a commit that referenced this pull request Mar 10, 2020
@vincentsarago
Copy link
Member Author

this now lives in #154

vincentsarago added a commit that referenced this pull request May 18, 2020
* v2 sketch

* update from #142

* Update CHANGES.txt

Co-Authored-By: jqtrde <hi@jacquestardie.org>

* add unscale option to apply internal scale/offset (#157)

* Terracotta default transform (#155)

* set padding default to 0

* switch to terracotta "calculate_default_transform"

* s/breacking/Breaking/g [skip-ci]

* add more metadata info (#159)

* add more metadata info

* add missing test

* update changes [skip-ci]

* update docs [skip-ci]

* Colormaps v2 (#160)

* 4bands colormap and new utility functions

* update old cmaps and add tests

* add migration docs

* add mypy to pre-commit and add missing types

* remove alpha band from output data array (#162)

* remove alpha band from output data array

* update changelog [skip ci]

* typo

* Gdal fall back (#165)

* fall back to gdal calculate_default_transform for dateline separation crossing dataset

* release note

* re-order part function to remove useless transform_bounds (#166)

* switch back to gdal calculate default transform and adapt VRT resolution to tile resolution (#169)

* update docs

* Smol documentation fixes (#170)

Merci Jacques!

* add simple COG info method (#171)

* add io.stack submodule (#172)

* add stack submodule

* update docs

* Landsat as uint16 (#174)

* revert float32 landsat output

* update changelog [skip ci]

* make image_profile return a copy of a dict (#178)

* Colormap docs (#179)

* Add colormap thumbnail script

* Add colormaps doc

* Coerce name to lower case

* Apply black on data

* Escape spaces in markdown

* Replace space with underscore

* Add docstring

* rename geographic colormaps

* add refs for custom colormaps

* fix link

* format

Co-authored-by: Kyle Barron <barronk@mit.edu>

* doc update

* fix unwanted breacking change with `img_profiles.get` not allowing default values (#180)

* allow reading part of data at hight resolution and add area/point function in io.cogeo (#182)

* allow reading part of data at hight resolution and add area/point function in io.cogeo

* make flake8 happy

* update change log [skip ci]

* remove print statement [skip ci]

* release

* add --pre option for pypi install [skip ci]

Co-authored-by: jqtrde <hi@jacquestardie.org>
Co-authored-by: Kyle Barron <barronk@mit.edu>
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.

Apply same resampling method for mask and tile
1 participant