Some YB tests fail with conda Python 3.7 & Matplotlib 3.x
See original GitHub issueI’m trying to run the tests. I get 133 failures, many of them
E AssertionError: 'Qt5Agg' != 'agg'
Does running the tests require setting a specific backend?
Issue Analytics
- State:
- Created 5 years ago
- Reactions:1
- Comments:14 (7 by maintainers)
Top Results From Across the Web
Pip install matplotlib fails on M1 Mac - python - Stack Overflow
After this it tries installing version 3.3.3 after which the same happens again, and so on. What goes wrong? Here is some of...
Read more >Changelog — documentación de Yellowbrick - v1.4
Implement new set_params and get_params on ModelVisualizers to ensure wrapped estimator is being correctly accessed via the new Estimator methods. Fix the test...
Read more >pycse - Python3 Computations in Science and Engineering
This is a collection of examples of using python in the kinds of scientific and engineering computations I have used in classes and...
Read more >Fastai v2 text
Has anyone faced a dead kernel error while using the Tokenizer class? ... Any and all help would be greatly appreciated. cc @jeremy....
Read more >Yellowbrick Documentation - Read the Docs
Yellowbrick is a Python 3 package and works well with 3.4 or later. ... If the test error doesn't match the train error...
Read more >Top Related Medium Post
No results found
Top Related StackOverflow Question
No results found
Troubleshoot Live Code
Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start FreeTop Related Reddit Thread
No results found
Top Related Hackernoon Post
No results found
Top Related Tweet
No results found
Top Related Dev.to Post
No results found
Top Related Hashnode Post
No results found
Top GitHub Comments
Hey there @amueller – nice to see you here in the YB issues!
In response to your question, yes, in order to ensure our image comparison tests work across many different operating systems (or at least don’t fail with backend errors!), we’ve found that we need to use
Agg
.Agg
has to be set before importingpyplot
, so that’s why we’ve got it on our test import (here). We also expect developers will run the tests usingpytest
from theroot
directory of YB (perhaps we should make this more clear in the contributor docs!). Running tests in this way will ensure the backend gets set properly first.But, just out of curiosity (and to help our future contributors!)-- do you have a custom config or backend for Matplotlib? Also, what do you usually run tests with (
nose
?pytest
?)?A sure-fire way to fix this is to set the
MPLBACKEND=agg
env which overrides any other configuration files.