https://github.com/dokato/navisml

interface between navis and scikit learn
https://github.com/dokato/navisml

Last synced: 18 days ago
JSON representation

interface between navis and scikit learn

• Host: GitHub
• URL: https://github.com/dokato/navisml
• Owner: dokato
• License: mit
• Created: 2022-04-08T13:01:28.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2023-11-22T13:43:55.000Z (about 1 year ago)
• Last Synced: 2024-10-16T12:39:40.366Z (2 months ago)
• Language: Python
• Size: 4.74 MB
• Stars: 2
• Watchers: 3
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
navisML

=======

`navisML` provides an easy interface between `navis` module for neurons analysis and `scikit-learn` for machine learning.

For every neuron object (mesh and skeleton) from a neuronlist it enables you to extract scalar information to use as a feature in a ML modelling.

## Installation

```

$ pip install git+https://github.com/dokato/navisML

```

## Example

```python

import navis

from navisML.extractor import NeuralFeatures

from sklearn.cluster import KMeans

neurons = navis.read_swc("path/to/data.zip", read_meta=True)

nrnfeats = NeuralFeatures({

    'upstream' : 'upstream',

    'downstream' : 'downstream',

    'has_soma' : 'has_soma',

    'custom_feature' : custom_function

})

X = nrnfeats.fit_transform(neurons)

kmeans = KMeans(n_clusters=2, random_state=0).fit(X)

```

## References

- [Navis](https://github.com/navis-org/navis)

- [scikit-learn](https://github.com/scikit-learn/scikit-learn/)