Instructions to use TransWiC/xlmr-large-zh-ET with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use TransWiC/xlmr-large-zh-ET with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="TransWiC/xlmr-large-zh-ET")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("TransWiC/xlmr-large-zh-ET") model = AutoModelForSequenceClassification.from_pretrained("TransWiC/xlmr-large-zh-ET") - Notebooks
- Google Colab
- Kaggle
| 0.6875Default classification report: | |
| precision recall f1-score support | |
| F 0.5844 0.6300 0.6064 500 | |
| T 0.5987 0.5520 0.5744 500 | |
| accuracy 0.5910 1000 | |
| macro avg 0.5916 0.5910 0.5904 1000 | |
| weighted avg 0.5916 0.5910 0.5904 1000 | |
| ADJ | |
| Accuracy = 0.5806451612903226 | |
| Weighted Recall = 0.5806451612903226 | |
| Weighted Precision = 0.6430390492359932 | |
| Weighted F1 = 0.5806451612903226 | |
| Macro Recall = 0.6118421052631579 | |
| Macro Precision = 0.6118421052631579 | |
| Macro F1 = 0.5806451612903226 | |
| ADV | |
| Accuracy = 0.5 | |
| Weighted Recall = 0.5 | |
| Weighted Precision = 0.8571428571428571 | |
| Weighted F1 = 0.5252525252525253 | |
| Macro Recall = 0.6875 | |
| Macro Precision = 0.6428571428571428 | |
| Macro F1 = 0.4949494949494949 | |
| NOUN | |
| Accuracy = 0.6010830324909747 | |
| Weighted Recall = 0.6010830324909747 | |
| Weighted Precision = 0.6015173041701515 | |
| Weighted F1 = 0.6011415264303145 | |
| Macro Recall = 0.6012389149713093 | |
| Macro Precision = 0.6011795373085695 | |
| Macro F1 = 0.6010505358580083 | |
| VERB | |
| Accuracy = 0.5824175824175825 | |
| Weighted Recall = 0.5824175824175825 | |
| Weighted Precision = 0.5820247280028809 | |
| Weighted F1 = 0.5796676494237469 | |
| Macro Recall = 0.5797872340425532 | |
| Macro Precision = 0.5819265143992056 | |
| Macro F1 = 0.5782926829268293 | |