Instructions to use TransWiC/bert-large-E with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use TransWiC/bert-large-E with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="TransWiC/bert-large-E")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("TransWiC/bert-large-E") model = AutoModelForSequenceClassification.from_pretrained("TransWiC/bert-large-E", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| Default classification report: | |
| precision recall f1-score support | |
| F 0.8941 0.8780 0.8860 500 | |
| T 0.8802 0.8960 0.8880 500 | |
| accuracy 0.8870 1000 | |
| macro avg 0.8871 0.8870 0.8870 1000 | |
| weighted avg 0.8871 0.8870 0.8870 1000 | |
| ADJ | |
| Accuracy = 0.9027777777777778 | |
| Weighted Recall = 0.9027777777777778 | |
| Weighted Precision = 0.9027777777777778 | |
| Weighted F1 = 0.9027777777777778 | |
| Macro Recall = 0.9024767801857585 | |
| Macro Precision = 0.9024767801857585 | |
| Macro F1 = 0.9024767801857585 | |
| ADV | |
| Accuracy = 0.7333333333333333 | |
| Weighted Recall = 0.7333333333333333 | |
| Weighted Precision = 0.7333333333333333 | |
| Weighted F1 = 0.7333333333333333 | |
| Macro Recall = 0.7222222222222222 | |
| Macro Precision = 0.7222222222222222 | |
| Macro F1 = 0.7222222222222222 | |
| NOUN | |
| Accuracy = 0.8787878787878788 | |
| Weighted Recall = 0.8787878787878788 | |
| Weighted Precision = 0.8789940425628158 | |
| Weighted F1 = 0.8787774432611717 | |
| Macro Recall = 0.8788291843030347 | |
| Macro Precision = 0.8789596819336596 | |
| Macro F1 = 0.878780921770074 | |
| VERB | |
| Accuracy = 0.9093959731543624 | |
| Weighted Recall = 0.9093959731543624 | |
| Weighted Precision = 0.9095620043258831 | |
| Weighted F1 = 0.9093867897967228 | |
| Macro Recall = 0.9093959731543624 | |
| Macro Precision = 0.9095620043258832 | |
| Macro F1 = 0.9093867897967228 | |