README.md

---
license: mit
---

## Model description

A toy model, trained on toy data generated from spikeinterface. 

# Intended use

Used to try out the download features of hugging face hub.

# How to Get Started with the Model

This can be used to automatically label a sorting in spikeinterface. Provided you have a `sorting_analyzer`, it is used as follows

``` python
from huggingface_hub import hf_hub_download
import skops.io as skio
from spikeinterface.curation import auto_label_units

new_model_path = hf_hub_download(repo_id="chrishalcrow/toy_data_model", filename="best_model.skops")
pipeline = skio.load(new_model_path, trusted='numpy.dtype')

auto_label_units(sorting_analyzer = sorting_analyzer, pipeline = pipeline)
```

or you can download the entire repositry to `a_folder_for_a_pipeline`, and use

``` python
from spikeinterface.curation import auto_label_units, load_pipeline

pipeline = load_pipeline('a_folder_for_a_pipeline')
auto_label_units(sorting_analyzer = sorting_analyzer, pipeline = pipeline)
```

# Authors

Chris Halcrow