Text Classification
Transformers
Safetensors
English
roberta
legal
multi-label-classification
banking
consumer-protection
contract-analysis
clause-classification
risk-detection
Eval Results (legacy)
text-embeddings-inference
Instructions to use Agreemind/en-banking-roberta with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Agreemind/en-banking-roberta with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="Agreemind/en-banking-roberta")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("Agreemind/en-banking-roberta") model = AutoModelForSequenceClassification.from_pretrained("Agreemind/en-banking-roberta") - Notebooks
- Google Colab
- Kaggle
| { | |
| "default_threshold": 0.5, | |
| "best_global_threshold": 0.7, | |
| "per_label_thresholds": { | |
| "hidden_fees": { | |
| "threshold": 0.7, | |
| "val_f1": 0.8815 | |
| }, | |
| "unilateral_rate_change": { | |
| "threshold": 0.75, | |
| "val_f1": 0.7294 | |
| }, | |
| "unilateral_terms_change": { | |
| "threshold": 0.65, | |
| "val_f1": 0.784 | |
| }, | |
| "overdraft_or_overlimit_penalty": { | |
| "threshold": 0.8, | |
| "val_f1": 0.8395 | |
| }, | |
| "auto_enrollment": { | |
| "threshold": 0.55, | |
| "val_f1": 0.5672 | |
| }, | |
| "data_sharing": { | |
| "threshold": 0.5, | |
| "val_f1": 0.8435 | |
| }, | |
| "dispute_limitation": { | |
| "threshold": 0.55, | |
| "val_f1": 0.8601 | |
| }, | |
| "account_freeze_or_closure": { | |
| "threshold": 0.7, | |
| "val_f1": 0.7845 | |
| }, | |
| "rewards_restriction_or_devaluation": { | |
| "threshold": 0.35, | |
| "val_f1": 0.1429 | |
| } | |
| } | |
| } |