Create boxplot function in visualize #218
Conversation
…so it actually acts like a save_dir
|
Also looks like you took care of some of #183, can you see if there's anything else? |
| random_data = test_utils.make_segmented_csv(100) | ||
| random_data = random_data[random_data['PatientID'].isin(np.arange(1, 5))] |
There was a problem hiding this comment.
we could make this an option in the test_utils if you think cases like this will become more common
There was a problem hiding this comment.
I only did it because I didn't want the boxplot to display like 10 different categories on one chart. I don't imagine we'll be doing this too often unless we plan to add a lot more faceted graphing functions in visualize.py, for example.
|
Yeah, and then all of the plots can get saved there
…On Thu, Sep 10, 2020 at 6:06 PM alex-l-kong ***@***.***> wrote:
***@***.**** commented on this pull request.
------------------------------
In ark/analysis/visualize.py
<#218 (comment)>
:
> + if save_dir is not None:
+ if not os.path.exists(save_dir):
+ raise ValueError("save_dir %s does not exist" % save_dir)
+
+ plt.savefig(os.path.join(save_dir, "sample_boxplot_viz.png"))
@ngreenwald <https://github.com/ngreenwald> OK, then would it be best to
add a command in the notebook to create data/example_dataset/visualization
if it doesn't exist?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#218 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADJB47KDA6JY5HQHMLYCIZDSFFZY7ANCNFSM4RGBIY5A>
.
|
There was a problem hiding this comment.
Seaborn defaults look quite nice. However, you picked Na to plot, aka Sodium, which is not actually a protein target: it's a naturally abundant metal. AKA we don't really care at all about sodium expression. Try picking CD20, CD45, PD-1, or one of the other biological targets.
NaBrO (sorry, I couldn't resist). But yeah, I'll fix that up. |
|
woops sorry, forgot to click go. You guys can merge stuff in once it's approved |
* Add boxplot viz function * Add test function for draw_boxplot and change save_dir functionality so it actually acts like a save_dir * Fix leftover PYCODESTYLE errors * Increase coverage with error tests, and fix split_vals check * PYCODESTYLE * Add new changes, including demo of draw_boxplot * Fix visualize cells test and add plot barchart tests * Finish up visualization cleanup, and adding creation of viz directory in notebook * Fix documentation for get_sorted_data * Change column to visualize boxplot for * Make get_sorted_data args and variable names more clear for documentation purposes
* Add boxplot viz function * Add test function for draw_boxplot and change save_dir functionality so it actually acts like a save_dir * Fix leftover PYCODESTYLE errors * Increase coverage with error tests, and fix split_vals check * PYCODESTYLE * Add new changes, including demo of draw_boxplot * Fix visualize cells test and add plot barchart tests * Finish up visualization cleanup, and adding creation of viz directory in notebook * Fix documentation for get_sorted_data * Change column to visualize boxplot for * Make get_sorted_data args and variable names more clear for documentation purposes
What is the purpose of this PR?
Addresses and closes #204. Addresses and closes #183.
How did you implement your changes
Add a function draw_boxplot to visualize.py which will allow the user to draw (and customize) a boxplot. In the meantime, clean up visualize.py and visualize_test.py.
Remaining issues
The previous way save_dir was being used was more of like a save_prefix. I changed the nature of save_dir and added a little more error checking to make sure we weren't trying to save to a save_dir that didn't exist.