Instructions to use UsefulSensors/moonshine with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use UsefulSensors/moonshine with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://UsefulSensors/moonshine") - Notebooks
- Google Colab
- Kaggle
Added ONNX Runtime format versions of the original ONNX files
#4
by petewarden - opened
The ONNX Runtime format offers smaller files than the original ONNX format, so it's helpful to use this approach with some of our web examples, where bandwidth is important.
Thesefiles were created by running these commands on the downloaded ONNX files:
py -m onnxruntime.tools.convert_onnx_models_to_ort .\public\moonshine\tiny\
py -m onnxruntime.tools.convert_onnx_models_to_ort .\public\moonshine\base\
keveman changed pull request status to merged