Token Classification
Transformers
Safetensors
English
modernbert
fill-mask
orality
linguistics
multi-label
custom_code
Instructions to use HavelockAI/bert-token-classifier with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use HavelockAI/bert-token-classifier with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="HavelockAI/bert-token-classifier", trust_remote_code=True)# Load model directly from transformers import AutoTokenizer, AutoModelForMaskedLM tokenizer = AutoTokenizer.from_pretrained("HavelockAI/bert-token-classifier", trust_remote_code=True) model = AutoModelForMaskedLM.from_pretrained("HavelockAI/bert-token-classifier", trust_remote_code=True) - Notebooks
- Google Colab
- Kaggle
Upload folder using huggingface_hub
Browse files- modeling_havelock.py +1 -1
modeling_havelock.py
CHANGED
|
@@ -15,7 +15,7 @@ class MultiLabelCRF(nn.Module):
|
|
| 15 |
self.start_transitions = nn.Parameter(torch.empty(num_types, 3))
|
| 16 |
self.end_transitions = nn.Parameter(torch.empty(num_types, 3))
|
| 17 |
# Placeholder — will be overwritten by loaded weights if present
|
| 18 |
-
self.register_buffer("emission_bias",
|
| 19 |
self._reset_parameters()
|
| 20 |
|
| 21 |
def _reset_parameters(self) -> None:
|
|
|
|
| 15 |
self.start_transitions = nn.Parameter(torch.empty(num_types, 3))
|
| 16 |
self.end_transitions = nn.Parameter(torch.empty(num_types, 3))
|
| 17 |
# Placeholder — will be overwritten by loaded weights if present
|
| 18 |
+
self.register_buffer("emission_bias", torch.zeros(1, 1, 1, 3))
|
| 19 |
self._reset_parameters()
|
| 20 |
|
| 21 |
def _reset_parameters(self) -> None:
|