Skip to content

Let MetricsView accept metrics which have some nans #217

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.

Sign up for GitHub

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

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Let MetricsView accept metrics which have some nans #217

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a0cc44a
allow for nan metrics
chrishalcrow Nov 27, 2025
1561fe8
do panel
chrishalcrow Nov 27, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
31 changes: 17 additions & 14 deletions spikeinterface_gui/metricsview.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -85,6 +85,7 @@ def _qt_creat_grid(self):

def _qt_refresh(self):
import pyqtgraph as pg
import pandas as pd
Copy link
Member

@samuelgarcia samuelgarcia Dec 3, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

what ?

from .myqt import QT


Expand All @@ -111,27 +112,25 @@ def _qt_refresh(self):

scatter.setData(x=values2, y=values1)

visible_unit_ids = self.controller.get_visible_unit_ids()
visible_unit_ids = self.controller.get_visible_unit_indices()

for unit_ind, unit_id in self.controller.iter_visible_units():
color = self.get_unit_color(unit_id)
scatter.addPoints(x=[values2[unit_ind]], y=[values1[unit_ind]], pen=pg.mkPen(None), brush=color)
if (not pd.isna(values2[unit_ind])) and (not pd.isna(values1[unit_ind])):
scatter.addPoints(x=[values2[unit_ind]], y=[values1[unit_ind]], pen=pg.mkPen(None), brush=color)

# self.scatter.addPoints(x=scatter_x[unit_id], y=scatter_y[unit_id], pen=pg.mkPen(None), brush=color)
# self.scatter_select.setData(selected_scatter_x, selected_scatter_y)
elif c == r:
values1 = units_table[visible_metrics[r]].values
values1_no_nans = values1[~np.isnan(values1)]

count, bins = np.histogram(values1, bins=self.settings['num_bins'])
count, bins = np.histogram(values1_no_nans, bins=self.settings['num_bins'])
curve = pg.PlotCurveItem(bins, count, stepMode='center', fillLevel=0, brush=white_brush, pen=white_brush)
plot.addItem(curve)

for unit_ind, unit_id in self.controller.iter_visible_units():
x = values1[unit_ind]
color = self.get_unit_color(unit_id)
line = pg.InfiniteLine(pos=x, angle=90, movable=False, pen=color)
plot.addItem(line)
if not pd.isna(x):
Copy link
Member Author

@chrishalcrow chrishalcrow Dec 3, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@samuelgarcia when metrics is read, nans are read in as some weird pandas type, which we check here. That's why we need to import pandas. If the user has got quality metrics loaded, they already have used pandas in the unit_tables creation.

line = pg.InfiniteLine(pos=x, angle=90, movable=False, pen=color)
plot.addItem(line)

def _qt_select_metrics(self):
if not self.tree_visible_metrics.isVisible():
Expand Down Expand Up @@ -187,6 +186,7 @@ def _panel_on_metrics_changed(self, event):
self.refresh()

def _panel_refresh(self):
import pandas as pd
import panel as pn
import bokeh.plotting as bpl
from bokeh.layouts import gridplot
Expand All @@ -212,6 +212,8 @@ def _panel_refresh(self):
col2 = visible_metrics[c]
values1 = units_table[col1].values
values2 = units_table[col2].values
values1_no_nans = values1[~np.isnan(values1)]
values2_no_nans = values2[~np.isnan(values2)]

plot = bpl.figure(
width=plot_size, height=plot_size,
Expand All @@ -227,7 +229,7 @@ def _panel_refresh(self):
plot.xaxis.axis_label = col1
plot.yaxis.axis_label = "Count"
# Create histogram
hist, edges = np.histogram(values1, bins=self.settings['num_bins'])
hist, edges = np.histogram(values1_no_nans, bins=self.settings['num_bins'])
if len(hist) > 0 and max(hist) > 0:
plot.quad(
top=hist, bottom=0, left=edges[:-1], right=edges[1:],
Expand All @@ -238,8 +240,9 @@ def _panel_refresh(self):
max_hist = max(hist)
for unit_ind, unit_id in self.controller.iter_visible_units():
x = values1[unit_ind]
color = self.get_unit_color(unit_id)
plot.line([x, x], [0, max_hist], line_width=2, color=color, alpha=0.8)
if not pd.isna(x):
color = self.get_unit_color(unit_id)
plot.line([x, x], [0, max_hist], line_width=2, color=color, alpha=0.8)
else:
# Off-diagonal - scatter plot
plot.xaxis.axis_label = col2
Expand All @@ -251,8 +254,8 @@ def _panel_refresh(self):

# Plot all points in light color first
all_source = ColumnDataSource({
'x': values2,
'y': values1,
'x': values2_no_nans,
'y': values1_no_nans,
'color': colors
})
plot.scatter('x', 'y', source=all_source, size=8, color='color', alpha=0.5)
Expand Down