he-cantillation

• Loss: 0.1410
• Wer: 11.6348
• Avg Precision Exact: 0.8905
• Avg Recall Exact: 0.8961
• Avg F1 Exact: 0.8928
• Avg Precision Letter Shift: 0.9069
• Avg Recall Letter Shift: 0.9133
• Avg F1 Letter Shift: 0.9096
• Avg Precision Word Level: 0.9091
• Avg Recall Word Level: 0.9146
• Avg F1 Word Level: 0.9114
• Avg Precision Word Shift: 0.9658
• Avg Recall Word Shift: 0.9719
• Avg F1 Word Shift: 0.9683
• Precision Median Exact: 0.9375
• Recall Median Exact: 0.9412
• F1 Median Exact: 0.9630
• Precision Max Exact: 1.0
• Recall Max Exact: 1.0
• F1 Max Exact: 1.0
• Precision Min Exact: 0.0
• Recall Min Exact: 0.0
• F1 Min Exact: 0.0
• Precision Min Letter Shift: 0.0
• Recall Min Letter Shift: 0.0
• F1 Min Letter Shift: 0.0
• Precision Min Word Level: 0.0
• Recall Min Word Level: 0.0
• F1 Min Word Level: 0.0
• Precision Min Word Shift: 0.6
• Recall Min Word Shift: 0.6429
• F1 Min Word Shift: 0.6429

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

• learning_rate: 1e-05
• train_batch_size: 8
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.39.0.dev0
• Pytorch 2.2.1+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0