update description

README.md

@@ -7,39 +7,61 @@ metrics:
 - recall
 - f1
 model-index:
-- name: xlm-ate-nobi-en-nes
+- name: xlm-ate-nobi-en
   results: []
+language:
+- en
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
+# XLMR Token Classifier for Term Extraction
 
-# xlm-ate-nobi-en-nes
-
-This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on an unknown dataset.
-It achieves the following results on the evaluation set:
-- Loss: 1.2581
-- Precision: 0.5875
-- Recall: 0.4794
-- F1: 0.5280
+This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) for cross-domain term extraction tasks.
 
 ## Model description
 
-More information needed
+This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) for token classification, specifically designed to identify and classify terms within text sequences. The model assigns labels such as B-Term, I-Term, BN-Term, IN-Term, and O to individual tokens, allowing for the extraction of meaningful terms from the text.
+
 
 ## Intended uses & limitations
 
-More information needed
+The model is intended for term extraction tasks. It can be applied in domains like:
+- Named Entity Recognition (NER)
+- Information Extraction
+
+## How to use
+
+Here's a quick example of how to use the model with the Hugging Face `transformers` library:
+
+```python
+from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline
+
+# Load the tokenizer and model
+tokenizer = AutoTokenizer.from_pretrained("tthhanh/xlm-ate-nobi-en-nes")
+model = AutoModelForTokenClassification.from_pretrained("tthhanh/xlm-ate-nobi-en-nes")
+
+# Create a pipeline for token classification
+nlp = pipeline("token-classification", model=model, tokenizer=tokenizer, aggregation_strategy="simple")
+
+# Example text
+text = "Treatment of anemia in patients with heart disease : a clinical practice guideline from the American College of Physicians ."
+
+# Get predictions
+predictions = nlp(text)
+
+# Print predictions
+for prediction in predictions:
+    print(prediction)
+
+```
 
 ## Training and evaluation data
 
-More information needed
+We fine-tuned the English version of the ACTER dataset where Named Entities are included in the gold standard. We trained on the Corruption and Wind Energy domain, validated on the Equitation domain, and tested on the Heart Failure domain.
 
 ## Training procedure
 
-### Training hyperparameters
-
 The following hyperparameters were used during training:
+```
 - learning_rate: 2e-05
 - train_batch_size: 32
 - eval_batch_size: 32
@@ -47,25 +69,30 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 20
+```
 
-### Training results
-
-| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     |
-|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|
-| 0.2205        | 1.85  | 500  | 0.5622          | 0.5705    | 0.4546 | 0.5060 |
-| 0.077         | 3.69  | 1000 | 0.7307          | 0.5715    | 0.4819 | 0.5229 |
-| 0.0421        | 5.54  | 1500 | 0.8561          | 0.5725    | 0.4965 | 0.5318 |
-| 0.0253        | 7.38  | 2000 | 0.8979          | 0.5601    | 0.5181 | 0.5383 |
-| 0.0157        | 9.23  | 2500 | 1.1252          | 0.6047    | 0.4565 | 0.5203 |
-| 0.0099        | 11.07 | 3000 | 1.1651          | 0.5874    | 0.4781 | 0.5271 |
-| 0.0077        | 12.92 | 3500 | 1.0574          | 0.5471    | 0.5270 | 0.5369 |
-| 0.0052        | 14.76 | 4000 | 1.1903          | 0.5879    | 0.4863 | 0.5323 |
-| 0.0034        | 16.61 | 4500 | 1.2581          | 0.5875    | 0.4794 | 0.5280 |
-
-
-### Framework versions
-
+Framework versions:
+```
 - Transformers 4.26.1
 - Pytorch 2.0.1+cu117
 - Datasets 2.9.0
 - Tokenizers 0.13.2
+```
+  
+## Evaluation
+
+We evaluate the performance of the ATE systems by comparing the candidate list extracted from the test set with the manually annotated gold standard term list for that specific test set. We use exact string matching to compare the retrieved terms to the ones in the gold standard and calculate Precision (P), Recall (R), and F1-score (F1). 
+The results are reported in [Can cross-domain term extraction benefit from cross-lingual transfer and nested term labeling?](https://link.springer.com/article/10.1007/s10994-023-06506-7#Sec12).
+
+## Citation
+If you use this model in your research or application, please cite it as follows:
+```
+@inproceedings{tran2022can,
+  title={Can cross-domain term extraction benefit from cross-lingual transfer?},
+  author={Tran, Hanh Thi Hong and Martinc, Matej and Doucet, Antoine and Pollak, Senja},
+  booktitle={International Conference on Discovery Science},
+  pages={363--378},
+  year={2022},
+  organization={Springer}
+}
+```