You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The issues is with matplotlib.tests.test_streamplot.test_streamplot_limits and matplotlib.tests.test_collections.test_barb_limits which passed on 1.3.x before the merge.
Arrays are not almost equal to 2 decimals
(mismatch 25.0%)
x: array([ 19.99259073, 29.99042969, 15.01481855, 6.01914062])
y: array([20, 30, 15, 6])
Dropping this from 2 decimals to 1 would be sufficient - the fact is that streamplot will not result in exact limits as it is the extents of the resulting paths which are used, not the input data itself.
Activity
pelson commentedon Mar 10, 2014
Sure.
Looks like the numbers are pretty close:
Dropping this from 2 decimals to 1 would be sufficient - the fact is that streamplot will not result in exact limits as it is the extents of the resulting paths which are used, not the input data itself.
The same issue applies to the barb test.
tacaswell commentedon Mar 10, 2014
Merged the PR, but I am still a bit concerned that it passed on 1.3.x before the merge, but not on master.