- 
 
- 
  Notifications
 You must be signed in to change notification settings 
- Fork 2.7k
Violin #471
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
Conversation
ping @empet, who wrote the 🎻 vignette this is based on
Right now, colorscaling just colors multiple violin plots from left to right with equal weight across. Good ideas could be colorscaling by some of the stat_params including median, lowest value, quartiles, etc.
Can we please take out the color bar by default and use qualitative coloring? Sequential coloring by stats_params sounds potentially useful as an option. I don't see any reason to ever arbitrarily color from left-to-right though.
Looks like this seaborn violin plot is may be coloring sequentially based on mean:
https://stanford.edu/~mwaskom/software/seaborn/examples/simple_violinplots.html 
I personally like the style of these a lot:
image 
I will be adding a colorscale that works on statistic parameters, user inputted stats on the data, etc.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
🐄 why the CAPS on this? Let's try to reserve those for module constants.
I commented out the last all_args test just in case I have to redo the function once again and need to rewrite the exp_violin_plot
New Docs, new examples in doc string, and most importantly, you can input a dictionary of stats assigned to values in the indexing column and get some colorscale stuff happening.
^This is colorscaling on median
Nice! Lookin' promising! I think the colorscale should always work off
absolute values rather than a 0-100 percentage scale. So, in this case, it
looks like the colorscale range would be ~(-1, 0.5).
On Mon, Jun 6, 2016 at 3:40 PM, Adam Kulidjian notifications@github.com 
wrote:
[image: violin-plot]
https://cloud.githubusercontent.com/assets/10369095/15835207/e2931c58-2bfc-11e6-92d5-1156c26bf28f.png^This is colorscaling on median
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
#471 (comment), or mute
the thread
https://github.com/notifications/unsubscribe/ABx4avQ4AM_-UTd2yqElrmkch1xrqOOgks5qJHeSgaJpZM4IjMGG
.
@Kully I think your entry method could use a little refactoring. I repeats a lot of the same color validations. A good first step would be to create a _validate_color_string method or something.
Also, in general, I think some of your function calls and error messages take up one or two more lines than are probably necessary. That's not such a big deal, but just something I noticed.
 
 
 plotly/tools.py
 
 Outdated
 
 
 There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
🐄 This is pretty un-duck-type-y. What's preventing a future tuple of tuples here?
@Kully overall, I'm mostly just psyched to see these violin plots!
That said it'd be great if you could address two things and I'll give this PR a final review:
- please remove the duplication of code related to whether colorsis an array or not, let me know if you need help with that.
- please uncomment that test 🐅
Thanks for cranking through all these new FigureFactory methods!
Thanks for cranking through all these new FigureFactory methods!
Thanks for the review.
please remove the duplication of code related to whether colors is an array or not, let me know if you need help with that.
I will need some help approaching this.
@theengineear Help! =)
@theengineear I was right!
 
 
 plotly/tools.py
 
 Outdated
 
 
 There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
🐈 Did you mean to commit this?
...] == ','...' line in _unlabel_rgb
@theengineear Ready for takeoff?
Yep 💃 after tests pass!
🚀
@jackparmer @theengineear @cldougl
First Pull Request for Violin Plots.
Lot of the work is done. Some notes:
-tests need to be added
-organize some of the color testing a little better
-I changed some of the preexisting color-converting functions to allow use for singleton color-strings
Right now, colorscaling just colors multiple violin plots from left to right with equal weight across. Good ideas could be colorscaling by some of the stat_params including median, lowest value, quartiles, etc.