Instructions to use Prompsit/paraphrase-bert-en with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Prompsit/paraphrase-bert-en with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="Prompsit/paraphrase-bert-en")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("Prompsit/paraphrase-bert-en") model = AutoModelForSequenceClassification.from_pretrained("Prompsit/paraphrase-bert-en") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -55,6 +55,16 @@ We have used as test dataset 16500 pairs of phrases human tagged.
|
|
| 55 |
Metrics obtained are:
|
| 56 |
|
| 57 |
```
|
| 58 |
-
metrics={
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
|
| 60 |
```
|
|
|
|
| 55 |
Metrics obtained are:
|
| 56 |
|
| 57 |
```
|
| 58 |
+
metrics={
|
| 59 |
+
'test_loss': 0.5660144090652466,
|
| 60 |
+
'test_accuracy': 0.8170742794799527,
|
| 61 |
+
'test_precision': 0.7043977055449331,
|
| 62 |
+
'test_recall': 0.5978578383641675,
|
| 63 |
+
'test_f1': 0.6467696629213483,
|
| 64 |
+
'test_matthews_correlation': 0.5276716223607356,
|
| 65 |
+
'test_runtime': 19.3345,
|
| 66 |
+
'test_samples_per_second': 568.88,
|
| 67 |
+
'test_steps_per_second': 17.792
|
| 68 |
+
}
|
| 69 |
|
| 70 |
```
|