Update README.md

README.md

@@ -5,4 +5,24 @@ tags:
 - multilabel-token-classification
 base_model:
 - google-bert/bert-large-cased
----
+---
+# Overview
+- This is an extension of the `bert-large-cased` model to enable **multi-label token classification**.
+- The training objective is BCELoss.
+- Labels are one-hot encoded.
+- Model output logits can be normalized using sigmoid activation.
+
+# Usage
+## Training
+To initialize the model for training, simply provide `id2label` and `label2id`, similarly to standard token classification fine tuning:
+```python
+from transformers import AutoModelForTokenClassification
+
+model = AutoModelForTokenClassification.from_pretrained('jvaquet/multilabel-classification-bert', 
+  id2label = id2label, 
+  label2id = label2id,
+  trust_remote_code=True)
+```
+
+
+