♻️ Refactor DatashaderRasterizer to be up front about datapipe lengths #39
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
Check during initialization of DatashaderRasterizerIterDataPipe on whether the input canvas and vector datapipes have compatible lengths. This is better than finding out that the zip function doesn't work when the datapipe is being iterated over. Added a unit test to cover the 3:2 ratio case and documented why the ValueError is raised on unmatched lengths. Also renamed the previous ValueError on unsupported geometry types to NotImplementedError to avoid confusion.
weiji14
added a commit
that referenced
this pull request
May 14, 2023
Used to wrong boolean operator (should be and, not or) which meant DatashaderRasterizer didn't work when the vector_datapipe was length 1 and canvas_datapipe was length 2 or more. This prevents false errors like `ValueError: Unmatched lengths for the canvas datapipe (XarrayCanvasIterDataPipe) and vector datapipe (PyogrioReaderIterDataPipe). The vector datapipe's length (1) should either be (1) to allow for broadcasting, or match the canvas datapipe's length of (2)`. Patches the bugfix at #39 from zen3geo v0.3.0. Also updated a unit test to use a 2:1 canvas:vector ratio to prevent regression.
weiji14
added a commit
that referenced
this pull request
May 14, 2023
* 🐛 Fix DatashaderRasterizer to allow N:1 instead of just 1:1 Used to wrong boolean operator (should be and, not or) which meant DatashaderRasterizer didn't work when the vector_datapipe was length 1 and canvas_datapipe was length 2 or more. This prevents false errors like `ValueError: Unmatched lengths for the canvas datapipe (XarrayCanvasIterDataPipe) and vector datapipe (PyogrioReaderIterDataPipe). The vector datapipe's length (1) should either be (1) to allow for broadcasting, or match the canvas datapipe's length of (2)`. Patches the bugfix at #39 from zen3geo v0.3.0. Also updated a unit test to use a 2:1 canvas:vector ratio to prevent regression. * 📌 Temporarily pin numpy<1.24 to prevent AttributeError The latest numpy=1.24 version installed via readthedocs is incompatible with datashader=0.14.4, resulting in an error like `AttributeError: module 'numpy' has no attribute 'warnings'`. This was reported in holoviz/datashader#1209 and fixed in holoviz/datashader#1176, but will need to wait for datashader>=0.14.5 to be released.
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.
Check during initialization of
DatashaderRasterizerIterDataPipeon whether the input canvas and vector datapipes have compatible lengths. This is better than finding out that the zip function doesn't work when the datapipe is being iterated over. Added a unit test to cover the 3:2 ratio case and documented why theValueErroris raised on unmatched lengths. Also renamed the previousValueErroron unsupported geometry types toNotImplementedErrorto avoid confusion.Motivated by issues discovered in #31 (see af489a8) when some DataPipe functions like https://pytorch.org/data/0.4/generated/torchdata.datapipes.iter.BatchMapper.html do not return any lengths.
Patches #35.