https://github.com/ianhi/napari-broadcastable-points

Points layer with broadcasting over arbitrary dims
https://github.com/ianhi/napari-broadcastable-points

Last synced: 12 days ago
JSON representation

Points layer with broadcasting over arbitrary dims

• Host: GitHub
• URL: https://github.com/ianhi/napari-broadcastable-points
• Owner: ianhi
• License: bsd-3-clause
• Created: 2022-04-29T20:52:17.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2024-04-01T22:02:48.000Z (9 months ago)
• Last Synced: 2024-04-25T06:09:52.409Z (8 months ago)
• Language: Python
• Size: 89.8 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# napari-broadcastable-points

[![License](https://img.shields.io/pypi/l/napari-broadcastable-points.svg?color=green)](https://github.com/ianhi/napari-broadcastable-points/raw/main/LICENSE)

[![PyPI](https://img.shields.io/pypi/v/napari-broadcastable-points.svg?color=green)](https://pypi.org/project/napari-broadcastable-points)

[![Python Version](https://img.shields.io/pypi/pyversions/napari-broadcastable-points.svg?color=green)](https://python.org)

Points layer that can broadcast over arbitrary dimensions. Available here as a workaround until something more complete is implemented in napari core (https://github.com/napari/napari/issues/2343).

**Warning!** This is likely to be very brittle for all the reasons discussed in the napari issue - so while it is useful it should also be used with caution. So don't muck around much with the viewer dims after creating, because who knows what will happen.

### Installation

```bash

pip install napari-broadcastable-points

```

### Usage

Here is an example where we have an  image sequence of (TPCZYX) and we broadcast points over the `C` and `Z` axes.

```python

import napari

from napari_broadcastable_points import BroadcastablePoints

import numpy as np

v = napari.Viewer()

# create a fake timelapse

T = 5

P = 4

C = 3

Z = 2

Y, X = 512, 512

images = np.zeros([T, P, C, Z, Y, X])

v.add_image(images)

# Add the relevant components of points data

# fmt: off

dat = np.array([

       # T,  P,       Y,              X

       [ 0,  0., 340.25071184, 284.13186557],

       [ 0,  0., 312.66551847, 309.95630191],

       [ 0,  0., 240.76794003, 266.81775485],

       [ 0,  0., 240.47448053, 239.81948049],

       [ 0,  1., 261.60356481, 260.36164576],

       [ 0,  1., 309.43746393, 215.16888217],

       [ 0,  1., 371.06395974, 235.12412843]])

# fmt: on

v.dims.axis_labels = ('T', 'P', 'C', 'Z', 'Y', 'X')

points = BroadcastablePoints(dat, broadcast_dims = (2, 3))

v.add_layer(points)

napari.run()

```

![example usage](images/points-broadcasting.gif)

**Creating an empty layer**

You can also create an empty layer - but be sure to specify `ndim` otherwise you may run into an error.

```python

points = BroadcastablePoints(None, broadcast_dims = (2, 3), ndim=6)

```