Update example
Browse files
README.md
CHANGED
|
@@ -59,7 +59,39 @@ You can use this model directly with a pipeline for masked language modeling:
|
|
| 59 |
```python
|
| 60 |
>>> from transformers import pipeline
|
| 61 |
>>> unmasker = pipeline('fill-mask', model='qwant/fralbert-base')
|
| 62 |
-
>>> unmasker("
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 63 |
```
|
| 64 |
|
| 65 |
Here is how to use this model to get the features of a given text in PyTorch:
|
|
|
|
| 59 |
```python
|
| 60 |
>>> from transformers import pipeline
|
| 61 |
>>> unmasker = pipeline('fill-mask', model='qwant/fralbert-base')
|
| 62 |
+
>>> unmasker("Paris est la capitale de la [MASK] .")
|
| 63 |
+
[
|
| 64 |
+
{
|
| 65 |
+
"sequence": "paris est la capitale de la france.",
|
| 66 |
+
"score": 0.6231236457824707,
|
| 67 |
+
"token": 3043,
|
| 68 |
+
"token_str": "france"
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"sequence": "paris est la capitale de la region.",
|
| 72 |
+
"score": 0.2993471622467041,
|
| 73 |
+
"token": 10531,
|
| 74 |
+
"token_str": "region"
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"sequence": "paris est la capitale de la societe.",
|
| 78 |
+
"score": 0.02028230018913746,
|
| 79 |
+
"token": 24622,
|
| 80 |
+
"token_str": "societe"
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"sequence": "paris est la capitale de la bretagne.",
|
| 84 |
+
"score": 0.012089950032532215,
|
| 85 |
+
"token": 24987,
|
| 86 |
+
"token_str": "bretagne"
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"sequence": "paris est la capitale de la chine.",
|
| 90 |
+
"score": 0.010002839379012585,
|
| 91 |
+
"token": 14860,
|
| 92 |
+
"token_str": "chine"
|
| 93 |
+
}
|
| 94 |
+
]
|
| 95 |
```
|
| 96 |
|
| 97 |
Here is how to use this model to get the features of a given text in PyTorch:
|