-
Notifications
You must be signed in to change notification settings - Fork 397
new image comparison unit tests #253
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
Comments
I've started the above unit tests on my img-tests fork. This is not installed by setup.py. The results seem pretty good. The test with matplotlib 1.4.3 is failing due to a high RMS between the images. (I've not seen how different the image looks.) "nightly" is failing because contourf has another instance of a numpy array being indexed by a casted floating point value due to division. Before I increased the compare_image tolerance to a RMS of 10 all the test failed because the tolerance was a RMS of 0.001 RMS tolerance of 10 was determined by looking at what is typical for unit testing in matplotlib and geopandas. |
I installed matplotlib 1.4.3. At a glance, I can't really tell the difference between this and other generated versions. I've included images from that version. The img-tests fork is using a later version (1.5.x?). Some options.
Let me know. Thanks! |
There are some lessons that can be learned from recent work in matplotlib: Not all of it can be taken advantage of right now, but some of it can be. On Mon, Jan 25, 2016 at 5:10 PM, Micah Cochran [email protected]
|
Here are some ideas that I have about adding image comparison unit test. I want to get some input about this before I do too much coding on this.
--plot-file
, and not display the plot to the screen when given--no-display
. (--no-display
is so that Travis will not get hung by any required interactivity.)Or should the unit tests be installed? If they are installed it will add a few megabytes of test PNG images that will only be used when test() is called.
I am leaning towards not installed image comparison unit tests. Any thoughts? Is there something that I may not be considering?
I hope that this might lead to enough unittesting that could allow us to start the process of rewriting portions of the basemap code.
The text was updated successfully, but these errors were encountered: