Instructions to use osanseviero/test_model_bertmesh with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use osanseviero/test_model_bertmesh with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("osanseviero/test_model_bertmesh", trust_remote_code=True) model = AutoModel.from_pretrained("osanseviero/test_model_bertmesh", trust_remote_code=True) - Notebooks
- Google Colab
- Kaggle
Nick Sorros commited on
Commit ·
7592134
1
Parent(s): 1fc3d25
Add warning that you need latest transformers
Browse files
README.md
CHANGED
|
@@ -20,6 +20,8 @@ The code for developing the model is open source and can be found in https://git
|
|
| 20 |
|
| 21 |
# How to use
|
| 22 |
|
|
|
|
|
|
|
| 23 |
You can use the model straight from the hub but because it contains a custom forward function due to the multilabel attention head you have to pass `trust_remote_code=True`. You can get access to the probabilities for all labels by omitting `return_labels=True`.
|
| 24 |
|
| 25 |
```
|
|
|
|
| 20 |
|
| 21 |
# How to use
|
| 22 |
|
| 23 |
+
⚠️ You need transformers 4.17+ for the example to work due to its recent support for custom models.
|
| 24 |
+
|
| 25 |
You can use the model straight from the hub but because it contains a custom forward function due to the multilabel attention head you have to pass `trust_remote_code=True`. You can get access to the probabilities for all labels by omitting `return_labels=True`.
|
| 26 |
|
| 27 |
```
|