Skip to content
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

Interactive functions may need to be retested #65

Open
josephcslater opened this issue Jul 22, 2017 · 11 comments
Open

Interactive functions may need to be retested #65

josephcslater opened this issue Jul 22, 2017 · 11 comments

Comments

@josephcslater
Copy link
Member

I've found other cases where interactive has failed. Solution is to add plt.show() at the end of all plotting.

@raphaeltimbo
Copy link
Collaborator

Which functions are failing?

@josephcslater
Copy link
Member Author

I'm not saying they are. All of my interactive functions in my notes for another package failed due to this. They may fail here. They need to be tested. I just needed to make a note of that- run all interactives to test again- so I didn't forget.

@josephcslater
Copy link
Member Author

They are very broken, but for a different reason. They report "time_plot_i can only be used in an iPython notebook."

Our checking code doesn't work anymore.

@josephcslater
Copy link
Member Author

So I found out that the posted potential issue doesn't appear today. I say that because I had to revert to reinstalling Anaconda (long story). However, the error requiring usage in an iPython notebook DOES exist on my linux installation. So, the check is not reliable.

@josephcslater
Copy link
Member Author

Please pay no attention... I'm apparently tracking down my own installation gremlins.

@josephcslater
Copy link
Member Author

So, I was partly correct. I don't know what causes it, but sometimes the plt.show() command is needed, sometimes not. I think the requirement may have come about between the installed version on my mac (old) and linux (new). plt.show() shouldn't cause any harm if not needed, so it's fine.

Further, average users will need to know to do jupyter nbextension enable --py --sys-prefix widgetsnbextension. This needs to be added to the docs somewhere.

@josephcslater
Copy link
Member Author

sdof_notebook doesn't act right. All plots don't initially display... not until one moves the slides. Others do. I can't tell the difference. Very odd. Thoughts?

@raphaeltimbo
Copy link
Collaborator

One way to change this is by using: %matplotlib inline instead of %matplotlib notebook.
The advantage for %matplotlib notebook is that we can zoom in etc... when working in the notebooks.
But for both cases we don't have the interactive plots loading in the docs...

@raphaeltimbo
Copy link
Collaborator

I have just noticed that the images in the web page tutorial are not displaying.
Any idea about why this is happening? Maybe we just need to run all the cells again and update the web page again...

@josephcslater
Copy link
Member Author

I don't. I spent most of my time, after getting level headings right, trying to get gh-pages to work again. I had this error.

As I recall, early on, we both made changes to preclude graphics from showing up in notebooks. The reason was that they cause problems when both you and I work on the same notebooks. I don't recall the problems, but I do recall us both trying to stop notebook graphics from being committed with the notebooks. Maybe this is the problem.

@raphaeltimbo
Copy link
Collaborator

Images on the sdof tutorial notebook are now displaying.

The problem that we had was that a change in the notebook output, which could be something not related to the code (e.g. change in the cell number), would be considered as a change in the code by git.
This would cause lots of problems because we were only using notebooks and at the time I didn't know nothing about git (still learning)... To avoid these problems we were always clearing the outputs before a git push.

The problem now is that if we clear the output images won't display in the tutorials.

I think that now we have more control on how we are using git and the outputs won't be a problem.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants