Update README.md
Browse files
README.md
CHANGED
|
@@ -51,7 +51,25 @@ generation you should look at model like GPT2.
|
|
| 51 |
|
| 52 |
### How to use
|
| 53 |
|
| 54 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
|
| 56 |
### Limitations and bias
|
| 57 |
|
|
|
|
| 51 |
|
| 52 |
### How to use
|
| 53 |
|
| 54 |
+
If you haven't already, you can install the [Transformers.js](https://huggingface.co/docs/transformers.js) JavaScript library from [NPM](https://www.npmjs.com/package/@huggingface/transformers) using:
|
| 55 |
+
```bash
|
| 56 |
+
npm i @huggingface/transformers
|
| 57 |
+
```
|
| 58 |
+
|
| 59 |
+
You can then use this model directly with a pipeline for masked language modeling:
|
| 60 |
+
|
| 61 |
+
```js
|
| 62 |
+
import { pipeline } from "@huggingface/transformers";
|
| 63 |
+
|
| 64 |
+
const unmasker = await pipeline("fill-mask", "onnx-community/bert-base-cased-ONNX");
|
| 65 |
+
const result = await unmasker("The capital of France is [MASK].", { top_k: 3 });
|
| 66 |
+
console.log(result);
|
| 67 |
+
// [
|
| 68 |
+
// { score: 0.44467854499816895, token: 2123, token_str: 'Paris', sequence: 'The capital of France is Paris.' },
|
| 69 |
+
// { score: 0.09395839273929596, token: 10067, token_str: 'Lyon', sequence: 'The capital of France is Lyon.' },
|
| 70 |
+
// { score: 0.08234485983848572, token: 18367, token_str: 'Toulouse', sequence: 'The capital of France is Toulouse.' }
|
| 71 |
+
// ]
|
| 72 |
+
```
|
| 73 |
|
| 74 |
### Limitations and bias
|
| 75 |
|