sm-subgroup-classifier / sv_NA-nb /training_details.txt
erikhenriksson's picture
Upload folder using huggingface_hub
f71be9a verified
raw
history blame contribute delete
889 Bytes
Training Details for sv_NA-nb
========================================
Language: sv
Register: NA-nb
Training Date: 2025-09-26 14:06:54
Data Summary:
- Total samples: 222925
- Training samples: 178340
- Test samples: 44585
- Embedding dimension: 1024
Classes:
- Number of classes: 2
- Class names: '', 'comments'
- Class distribution: {'': 203958, 'comments': 18967}
Cross-Validation Results:
- CV folds: 5
- CV scores: [0.99290680722216, 0.9922339351799933, 0.9935516429292363, 0.9936918246046876, 0.993523606594146]
- CV mean: 0.9932
- CV std: 0.0005
- CV confidence interval: 0.9932 ± 0.0011
Final Performance:
- Test accuracy: 0.9940
Model Configuration:
- Algorithm: Logistic Regression
- Regularization (C): 1.0
- Feature scaling: StandardScaler
- Random state: 42
Files:
- Classifier: model.pkl
- Scaler: scaler.pkl
- Metadata: metadata.pkl
- This file: training_details.txt