Upload ESG Topic Classifier (Macro-F1: 0.7070)
Browse files
README.md
CHANGED
|
@@ -113,6 +113,8 @@ with torch.no_grad():
|
|
| 113 |
print("Prediction:", model.config.id2label[pred_id])
|
| 114 |
print("Confidence:", float(probs[0, pred_id]))
|
| 115 |
|
|
|
|
|
|
|
| 116 |
## Limitations
|
| 117 |
|
| 118 |
Trained specifically on Vietnamese banking reports
|
|
@@ -123,6 +125,8 @@ Some ambiguity exists between Policy, Environmental, and Financing categories du
|
|
| 123 |
|
| 124 |
Minority classes (E, Policy) have fewer samples than Non-ESG and Governance
|
| 125 |
|
|
|
|
|
|
|
| 126 |
## Citation
|
| 127 |
|
| 128 |
If you use this model, please cite:
|
|
|
|
| 113 |
print("Prediction:", model.config.id2label[pred_id])
|
| 114 |
print("Confidence:", float(probs[0, pred_id]))
|
| 115 |
|
| 116 |
+
---
|
| 117 |
+
|
| 118 |
## Limitations
|
| 119 |
|
| 120 |
Trained specifically on Vietnamese banking reports
|
|
|
|
| 125 |
|
| 126 |
Minority classes (E, Policy) have fewer samples than Non-ESG and Governance
|
| 127 |
|
| 128 |
+
---
|
| 129 |
+
|
| 130 |
## Citation
|
| 131 |
|
| 132 |
If you use this model, please cite:
|