Correct implementation of rescale_discrete_levels #1078
Merged
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
ianthomas23
commented
Apr 29, 2022
| @@ -172,20 +172,22 @@ def eq_hist(data, mask=None, nbins=256*256): | |||
|
|
|||
| # Run more accurate value counting if data is of boolean or integer type | |||
| # and unique value array is smaller than nbins. | |||
| if data2.dtype == bool or (np.issubdtype(data2.dtype, np.integer) and data2.max() < nbins): | |||
| if data2.dtype == bool or (np.issubdtype(data2.dtype, np.integer) and data2.ptp() < nbins): | |||
There was a problem hiding this comment.
Changed from data2.max() to data2,ptp() to cover the unusual but possible use case of a small number of discrete levels but a non-zero minimum.
There was a problem hiding this comment.
Makes sense, though there's always a non-zero minimum because zero values are treated as NaN, so really this better covers the case when there is a minimum other than 1.
ianthomas23
commented
Apr 29, 2022
jbednar
reviewed
Apr 30, 2022
| @@ -172,20 +172,22 @@ def eq_hist(data, mask=None, nbins=256*256): | |||
|
|
|||
| # Run more accurate value counting if data is of boolean or integer type | |||
| # and unique value array is smaller than nbins. | |||
| if data2.dtype == bool or (np.issubdtype(data2.dtype, np.integer) and data2.max() < nbins): | |||
| if data2.dtype == bool or (np.issubdtype(data2.dtype, np.integer) and data2.ptp() < nbins): | |||
There was a problem hiding this comment.
Makes sense, though there's always a non-zero minimum because zero values are treated as NaN, so really this better covers the case when there is a minimum other than 1.
jbednar
approved these changes
May 4, 2022
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.
Original implementation of
rescale_discrete_levelsforhow='eq_hist'used the maximum value in the agg rather than the number of discrete levels. This PR corrects it.The number of discrete levels is only calculated if an agg has bool or int
dtype, and only ifmax(agg)-min(agg)is less than 256. Rescaling is only ever applied if there are fewer than 100 discrete levels. It is possible to fool this logic by using, for example, an agg containing unique values of1and300. This could be fixed, if desired, by performing thenp.unique()call earlier if thedtypeis integer.There are minor changes to images returned by
shadethat are indistinguishable to the naked eye for the standard use case ofagg=ds.count().