Repository navigation
Datashader shapes outlines and fill_alpha documentation #367
Description
Status after #309, just to document some thoughts/details about the comparison between ds and mpl:
- outlines linewidth with datashader seems to vary slightly between e.g. polygons and circles, which is not the case with matplotlib
- below you see that the outlines of the circles are slightly cut in places where the extent of circles ends, but is later on expanded because of the polygons. This is similar to what we see with points. It should only be a problem with large linewidth values though.
fill_alphain ds looks lighter than in mpl, even though it's0.5both times. Reason: in ds, the fill_alpha is applied two times: once during theshade()step, to make sure elements are blended with each other. Then also during the rendering of the resulting image, to make sure other things that were plotted before still shine through. Possible fix: usesqrt(fill_alpha)both times. Problem: if sth else is rendered below, it would be obvious that the elements are less transparent than the user would expect (e.g. 0.7 instead of 0.5). That could be confusing.
import spatialdata_plot
from spatialdata.datasets import blobs
blob=blobs()
blob.pl.render_shapes(method="datashader", fill_alpha=0.5, outline_alpha=0.7, color="red", outline_color="blue", outline_width=5).pl.show()
blob.pl.render_shapes(method="matplotlib", fill_alpha=0.5, outline_alpha=0.7, color="red", outline_color="blue", outline_width=5).pl.show()
Activity
LucaMarconato commentedon Oct 2, 2024
Thanks for reporting. I comment on 3. Could you please link the two code places where the
alphais used? I wonder if we can set the second alpha to be 1. Maybe it is a bug of datashader that we can report/fix?Sonja-Stockhaus commentedon Oct 2, 2024
Sure:
spatialdata-plot/src/spatialdata_plot/pl/render.py
Lines 256 to 261 in febd424
spatialdata-plot/src/spatialdata_plot/pl/render.py
Lines 277 to 279 in febd424
I think the problem is inherent to the way we render: first, create an image using datashader and then render it as if it was a SpatialImage. These 2 steps lead to fill_alpha being used twice, so I'd say it's not a datashader bug.
If we set the second one to 1, we get the problem again that the shapes will not look transparent if sth else, e.g. an image was rendered before them. See below (e.g. the triangles are not see-through anymore)
LucaMarconato commentedon Oct 2, 2024
Thanks, clear now. I'd keep things like this for the moment and merge the PR, and think about this later on.
I think the problem can be formulated as follows: consider an image
A, an imageB(background) and a white imageW. From my understanding, the image produced by datashader, let's call itD, is the alpha blend between the imageAandW. In formulas it isalpha * A + (1 - alpha) W.The matplotlib image is
alpha * A + (1 - alpha) B.Probably one can find a function
f, and a valuealpha', such that the alpha blend (usingalpha'), betweenf(D)andBis equal to the alpha blend (usingalpha) betweenAandB.The above shows that using
alpha' = alphaandf = identitydoesn't work, but probably the solution can be easily found.LucaMarconato commentedon Oct 2, 2024
I was lazy to do the math, so I gave a try to the new o1-preview from ChatGPT. It gives the following (I checked the calculations and they are correct):
Can you try it out please? Basically the
SpatialImageDshould becomef(D)this code is calledspatialdata-plot/src/spatialdata_plot/pl/render.py
Lines 277 to 279 in febd424
Please note that the assumption is that the image produced by datashader has RGB values between 0 and 1, if not, it should be first scaled. If the datashader image has an alpha channel, maybe the formula would not work and we need to adjust the formulation of the problem and find the formula for the new version of the problem.
LucaMarconato commentedon Oct 2, 2024
This is the answer from the bot for future reference.
Sonja-Stockhaus commentedon Oct 3, 2024
I think I got you, thanks for the input! Brain dump from my side:
shouldn't the datashader redering rather be
alpha * (alpha * A + (1 - alpha) * W) + (1 - alpha) * Bwhich equals toalpha²*A + (alpha-alpha²)*W + (1-alpha)*Bwhich in turn explains why the image is too pale (since we weighAwithalpha²). If we used e.g.sqrt(alpha)instead ofalpha, we would getalpha*A + (sqrt(alpha)-alpha)*W + (1-sqrt(alpha))*Bwhich uses the same weight forAas mpl (alpha*A + (1-alpha)*B), but a different one forBand it still hasWin it.Does that make any sense?
LucaMarconato commentedon Oct 3, 2024
Yes, this would be the case with
fequal to identity andalpha'equal toalpha. But as you observed, both this case, or the case withalpha' = sqrt(alpha)lead to different results as the one obtained by directly blendingAtoBwithout involvingW.Fix datashader fill_alpha applied twice (#367)
Update datashader visual test reference images
Update 2 datashader visual test reference images
Address review: add outline test, align docstrings, fix Sphinx ref