Text Classification
Transformers
TensorBoard
English
Swahili
multi-task-learning
fraud-detection
sentiment-analysis
call-quality
question-answering
jenga-ai
nlp-for-africa
security
attention-fusion
Eval Results (legacy)
Instructions to use Rogendo/JengaAI-multi-task-distilbert-base-uncased with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Rogendo/JengaAI-multi-task-distilbert-base-uncased with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="Rogendo/JengaAI-multi-task-distilbert-base-uncased")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Rogendo/JengaAI-multi-task-distilbert-base-uncased", dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "epoch": 8, | |
| "metrics": { | |
| "fraud_detection_fraud_accuracy": 1.0, | |
| "fraud_detection_fraud_precision": 1.0, | |
| "fraud_detection_fraud_recall": 1.0, | |
| "fraud_detection_fraud_f1": 1.0, | |
| "sentiment_analysis_sentiment_accuracy": 0.3333333333333333, | |
| "sentiment_analysis_sentiment_precision": 0.16666666666666666, | |
| "sentiment_analysis_sentiment_recall": 0.3333333333333333, | |
| "sentiment_analysis_sentiment_f1": 0.2222222222222222, | |
| "call_quality_opening_precision": 0.964, | |
| "call_quality_opening_recall": 0.964, | |
| "call_quality_opening_f1": 0.964, | |
| "call_quality_listening_precision": 0.9018963337547409, | |
| "call_quality_listening_recall": 0.9374507227332457, | |
| "call_quality_listening_f1": 0.9193298969072164, | |
| "call_quality_proactiveness_precision": 0.7551020408163265, | |
| "call_quality_proactiveness_recall": 0.8293103448275863, | |
| "call_quality_proactiveness_f1": 0.790468364831553, | |
| "call_quality_resolution_precision": 0.9021229666390956, | |
| "call_quality_resolution_recall": 0.9157570668905681, | |
| "call_quality_resolution_f1": 0.9088888888888889, | |
| "call_quality_hold_precision": 0.7575757575757576, | |
| "call_quality_hold_recall": 0.6067961165048543, | |
| "call_quality_hold_f1": 0.6738544474393531, | |
| "call_quality_closing_precision": 0.869, | |
| "call_quality_closing_recall": 0.869, | |
| "call_quality_closing_f1": 0.869, | |
| "eval_loss": 2.068362368643284, | |
| "train_loss_avg": 1.4034666525022614 | |
| }, | |
| "is_best": false | |
| } |