Mapping values to foci. #259

cristidonos · 2018-10-20T10:46:13Z

I made the following changes to plot SEEG electrodes and task-related activations:

Brain.add_foci can now map values using foci color and size.
added Brain.add_text3d for labeling foci in 3d

- added Brain.add_text3d for labeling foci in 3d

agramfort · 2018-10-20T11:48:02Z

surfer/viz.py

+        data : numpy array
+            1D array the same size as the number of foci. Spheres will be
+            colorcoded acording to data values, whereas spheres' sizes will be
+            coded using the absolute values of data.


API seems quite complicated. Can you achieve what you want using only one new parameter? and without adding a new method?

also you miss tests and some new example for documentation.

I agree with @agramfort. Can allow_data as an external parameter not be replaced with an internal check for data is not None?

I also am not sure that I like forcing the color and size of the spheres by default. There are lots of cases when you might want colors to reflect a scalar but a uniform size. Or (probably less commonly) vice versa. Also I don't see any way to control the size range, colormap, etc. through this API, which would be important.

Thanks for pointing out the issues, i'll try to fix them. This is my first attempt at a pull request, so bear with me, especially with the tests.

mwaskom · 2018-10-20T15:31:52Z

surfer/viz.py

@@ -1588,8 +1588,8 @@ def add_foci(self, coords, coords_as_verts=False, map_surface=None,
            whether the coords parameter should be interpreted as vertex ids
        map_surface : Freesurfer surf or None
            surface to map coordinates through, or None to use raw coords
-        scale_factor : float
-            Controls the size of the foci spheres (relative to 1cm).
+        scale_factor : int


Why the change to int here? Scale factors can be floats, as seen in the example: http://pysurfer.github.io/auto_examples/plot_foci.html Do your changes break that, or is this just an errant attempt to fix the documentation?

mwaskom · 2018-10-20T15:34:44Z

surfer/viz.py

@@ -1737,6 +1748,45 @@ def add_text(self, x, y, text, name, color=None, opacity=1.0,
        if justification is not None:
            text.property.justification = justification

+    def add_text3d(self, x, y, z, text, name, color=None, opacity=1.0,
+                 row=-1, col=-1, font_size=None, justification=None):


Do we use this row, col parameterization anywhere else in pysurfer? Ideally if you're going to add a new method it should match the existing API as closely as possible.

mwaskom · 2018-10-20T15:37:44Z

surfer/viz.py

+        if font_size is not None:
+            text.property.font_size = font_size
+            text.actor.text_scale_mode = 'viewport'
+        if justification is not None:


This parameter doesn't appear in the docstring. Also despite it being the property name in mayavi, "justification" is a weird term and "alignment" would be more correct.

…t_foci example

codecov-io · 2018-10-23T10:42:15Z

Codecov Report

Merging #259 into master will decrease coverage by 8.25%.
The diff coverage is 90.32%.

@@            Coverage Diff             @@
##           master     #259      +/-   ##
==========================================
- Coverage    74.9%   66.64%   -8.26%     
==========================================
  Files           7        7              
  Lines        2506     2536      +30     
  Branches      501      505       +4     
==========================================
- Hits         1877     1690     -187     
- Misses        454      664     +210     
- Partials      175      182       +7

cristidonos · 2018-10-23T10:54:26Z

Hi,
Made some changes, not there's only one add_foci function that uses quiver3d instead of points3d to draw the spheres. Implemented internal checks for data and made color-coded spheres optional.

Regarding text functions: i only copied add_text function from above and changed it to uses the text3d function in mayavi. All 4 arguments (row, col, justification and font_size) were already there.

Couldn't figure out failed tests myself, please help me fix those.

* FIX: Render tweaks * FIX: Use correct data_dict * FIX: test backend * FIX: Better check * FIX: Old VTK

* FIX: width/height in offscreen mode * MAINT: Sisyphus * BUG: Modernize labels * FIX: Circle * FIX: Stretch * FIX: More * MAINT: Push the rock * FIX: Skip for 3

* Integrate better with Jupyuter notebook Mayavi can render things inside a jupyter notebook as either PNG or X3D. This functionality can be enabled with `mlab.init_notebook()`. This PR adds an `_ipython_display_` hook to the `Brain` class that renders the brain accordgin to the Mayavi notebook integration settings. * Add note on Jupyter notebook integration to docs * Fix links in doc

* MAINT: Update fsaverage * FIX: Bump to 3.5 * ENH: Run more on CircleCI * MAINT: CI * MAINT: CI * MAINT: CI * MAINT: Circle * FIX: Pattern * FIX: Circle * FIX: Try again * FIX: Better * FIX: One more

* ENH: Add option to smooth to nearest * DOC: Update docstring * FIX: Flake * FIX: Undordered vertices are the worst * FIX: Orders are hard * DOC: Changes

* Feat: add kwargs in Brain.add_* and pass it to mayavi modules, so that we could directly manipulate mayavi display settings, like point resolution and overlay opacity. * Fix: Fix code styles which failed in the flake test. * Modify test_viz to pass the kwargs to the mayavi module. * fix flake test. * Specifying `kwargs` as `mlab_kws` and make the documentation more clear. * Revert "Specifying `kwargs` as `mlab_kws` and make the documentation more clear." This reverts commit c7a5912. * Link mayavi documents. * Give detailed explanations about the kwargs passed to the functions. * Revert "Link mayavi documents." This reverts commit 2c86a04. * Change the position of line breaking. * Fix the typo and unavailable links. * Fix render errors. * Revert "Fix render errors." This reverts commit ac0b7b8. * Fix render error. * Improve code rendering. * Revert "Fix render error." This reverts commit bbf7631. * Fix line breaking error. * Revert "Fix line breaking error." This reverts commit 6efccac. * Fix line breaking error. * Revert "Fix line breaking error." This reverts commit a7abf61. * Force line breaking to fix the rendering error on long code. * Revert "Force line breaking to fix the rendering error on long code." This reverts commit 6a9cdf1. * Force line breaking to fix the rendering error on long code. * Revert "Force line breaking to fix the rendering error on long code." This reverts commit c7f6a9f. * Force line breaking to fix the rendering error on long code. * Revert "Force line breaking to fix the rendering error on long code." This reverts commit de31e5f. * Fix render error Force line breaking to avoid render error. * Render undocumented function as plain code. * Fix sectioning. * FIX: Formatting

BUG: Properly clean up on del

danielegrattarola · 2020-02-03T15:58:05Z

Hi, was this feature ever implemented in the library?
I need to plot some text on a brain surface but I'm not sure how to achieve this in pysurfer.

Thanks

mwaskom · 2020-02-03T16:57:26Z

@danielegrattarola This has not been merged. There are unresolved questions about how best to integrate into the existing PySurfer API, and the tests are not passing. But looking at the code should give you a pretty good idea of how to add text onto a PySurfer visualization. It will take you writing more code than if it were just a method on the Brain object, but it is definitely possible.

danielegrattarola · 2020-02-03T17:04:20Z

@mwaskom Thanks, I've managed to hack a solution together using mlab.text3d for now.
Cheers

MRG, FIX: Safer exit

* FIX: Fix bug with scale_data_colormap * FIX: Fix for deprecation * FIX: One more

[skip ci]

MRG, FIX: Remove force render

* FIX: Verbose * FIX: MNE vers

* ENH: Move up to 3.8 * FIX: Newer * FIX: Newer

* ENH: Add FXAA option * FIX: Mayavi for VTK9

* MAINT: Simpler vector params * FIX: Undo auto scaling * FIX: Dup * FIX: URL * FIX: Better * FIX: More tolerant of type

… master � Conflicts: � surfer/viz.py

- Brain.add_foci can now map values using foci color and size.

1716aab

- added Brain.add_text3d for labeling foci in 3d

agramfort reviewed Oct 20, 2018

View reviewed changes

mwaskom reviewed Oct 20, 2018

View reviewed changes

re-organized add_foci function to allow data mapping and extended plo…

9fbd69f

…t_foci example

larsoner and others added 13 commits October 26, 2018 08:27

MRG: Fix scale_data_colormap and add FXAA (nipy#261)

16e5ed2

* FIX: Render tweaks * FIX: Use correct data_dict * FIX: test backend * FIX: Better check * FIX: Old VTK

MRG: width/height in offscreen mode (nipy#265)

df68527

* FIX: width/height in offscreen mode * MAINT: Sisyphus * BUG: Modernize labels * FIX: Circle * FIX: Stretch * FIX: More * MAINT: Push the rock * FIX: Skip for 3

MAINT: Update fsaverage (nipy#270)

27f54c6

* MAINT: Update fsaverage * FIX: Bump to 3.5 * ENH: Run more on CircleCI * MAINT: CI * MAINT: CI * MAINT: CI * MAINT: Circle * FIX: Pattern * FIX: Circle * FIX: Try again * FIX: Better * FIX: One more

MAINT: Fix ignores for CIs (nipy#272)

b9a7dd6

Set annotation LUT directly (nipy#271)

b2474cf

Allow painting all annotation vertices in a single color (nipy#273)

80c3ccb

FIX: don't change view when adding foci (nipy#274)

9cd7511

ENH: Add option to smooth to nearest (nipy#275)

1fb7daf

* ENH: Add option to smooth to nearest * DOC: Update docstring * FIX: Flake * FIX: Undordered vertices are the worst * FIX: Orders are hard * DOC: Changes

MAINT: CircleCI for latest pip (nipy#277)

64a1069

BUG: Properly clean up on del

0e1da27

Merge pull request nipy#279 from larsoner/cleanup

3692eb4

BUG: Properly clean up on del

larsoner added 6 commits February 6, 2020 14:38

FIX: Safer exit

e79624d

FIX: Actually close

78950ed

FIX: Try again

e79e15d

FIX: Again again

0b20e8b

FIX: Different

1435b57

FIX: One more

2cf1d73

larsoner and others added 29 commits February 10, 2020 10:41

FIX: Dont reuse by default

5aaf043

Merge pull request nipy#285 from larsoner/safer

21e7183

MRG, FIX: Safer exit

Fix (nipy#286)

dd9f3b5

MRG, FIX: Fix bug with scale_data_colormap (nipy#287)

cd2f074

* FIX: Fix bug with scale_data_colormap * FIX: Fix for deprecation * FIX: One more

fix int (nipy#288)

6962fe9

Update __init__.py

f576145

Update __init__.py

bd2bbfe

FIX: CSS

f67f444

FIX: Update changes

6967306

Set up CI with Azure Pipelines

7952423

[skip ci]

FIX: Remove force render

9c43880

ENH: Azure

38db203

FIX: Missed

b7f25db

FIX: Missed again

963c9b6

FIX: Try again

b3522d4

FIX: Wrong call

11afc4a

FIX: Explicit close/del

283a580

FIX: Safer del

ce7ac2d

Merge pull request nipy#289 from larsoner/force

07dc04b

MRG, FIX: Remove force render

STY: Dont log info messages by default (nipy#293)

1afce4b

* FIX: Verbose * FIX: MNE vers

ENH: Move up to 3.8 (nipy#294)

09513f9

* ENH: Move up to 3.8 * FIX: Newer * FIX: Newer

ENH: Add FXAA option (nipy#295)

b0ca3a1

* ENH: Add FXAA option * FIX: Mayavi for VTK9

MRG, MAINT: Simpler vector params (nipy#291)

9e5fe1f

* MAINT: Simpler vector params * FIX: Undo auto scaling * FIX: Dup * FIX: URL * FIX: Better * FIX: More tolerant of type

MAINT: Bump version

85e9e6c

MAINT: Bump version

f02cca8

option to apply xfrom from surface files

2512fc6

merged changes by CD on top of nipy's f02cca8 version from Jun 23

d3749dc

merged changes by CD on top of nipy's f02cca8 version from Jun 23

6c9cb10

Merge branch 'master' of https://github.com/cristidonos/PySurfer into…

cd3de2f

… master � Conflicts: � surfer/viz.py

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Mapping values to foci. #259

Mapping values to foci. #259

cristidonos commented Oct 20, 2018

agramfort Oct 20, 2018

mwaskom Oct 20, 2018

mwaskom Oct 20, 2018

cristidonos Oct 20, 2018

mwaskom Oct 20, 2018

mwaskom Oct 20, 2018

mwaskom Oct 20, 2018

codecov-io commented Oct 23, 2018

cristidonos commented Oct 23, 2018

danielegrattarola commented Feb 3, 2020

mwaskom commented Feb 3, 2020

danielegrattarola commented Feb 3, 2020

Mapping values to foci. #259

Are you sure you want to change the base?

Mapping values to foci. #259

Conversation

cristidonos commented Oct 20, 2018

agramfort Oct 20, 2018

Choose a reason for hiding this comment

mwaskom Oct 20, 2018

Choose a reason for hiding this comment

mwaskom Oct 20, 2018

Choose a reason for hiding this comment

cristidonos Oct 20, 2018

Choose a reason for hiding this comment

mwaskom Oct 20, 2018

Choose a reason for hiding this comment

mwaskom Oct 20, 2018

Choose a reason for hiding this comment

mwaskom Oct 20, 2018

Choose a reason for hiding this comment

codecov-io commented Oct 23, 2018

Codecov Report

cristidonos commented Oct 23, 2018

danielegrattarola commented Feb 3, 2020

mwaskom commented Feb 3, 2020

danielegrattarola commented Feb 3, 2020