Update README.md
Browse files
README.md
CHANGED
|
@@ -9,4 +9,16 @@ base_model:
|
|
| 9 |
## Introduction
|
| 10 |
|
| 11 |
This repository hosts the [Kokoro](https://huggingface.co/hexgrad/Kokoro-82M) model for the [React Native Executorch](https://www.npmjs.com/package/react-native-executorch) library.
|
| 12 |
-
It includes the model divided into 4 parts, each of them exported for xnnpack backend in .pte format, ready for use in the ExecuTorch runtime.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
## Introduction
|
| 10 |
|
| 11 |
This repository hosts the [Kokoro](https://huggingface.co/hexgrad/Kokoro-82M) model for the [React Native Executorch](https://www.npmjs.com/package/react-native-executorch) library.
|
| 12 |
+
It includes the model divided into 4 parts, each of them exported for xnnpack backend in .pte format, ready for use in the ExecuTorch runtime.
|
| 13 |
+
|
| 14 |
+
As it stands for now, the models are exported with static input shapes: for 32, 64 and 128 input tokens, with methods
|
| 15 |
+
`forward_32`, `forward_64`, and `forward_128` respectively.
|
| 16 |
+
|
| 17 |
+
## Compatibility
|
| 18 |
+
|
| 19 |
+
These models were exported using v1.0.0 version of ExecuTorch and no forward compatibility is guaranteed.
|
| 20 |
+
Older versions of the runtime may not work with these files.
|
| 21 |
+
|
| 22 |
+
The models are intended to be used within the React Native ExecuTorch package. If you want to use them outside the package,
|
| 23 |
+
make sure your runtime is compatible with the ExecuTorch version used to export the .pte files and follow the
|
| 24 |
+
[example script](https://github.com/NorbertKlockiewicz/kokoro-export/blob/main/demo/inference_example.py).
|