nep-spell-mt5-small-0

This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0009
• Accuracy: 0.8446
• Precision: 0.889
• Recall: 0.8446
• F1: 0.8603
• Wer: 0.0105
• Cer: 0.0024
• Chrf: 99.2925
• Exact Match: 0.8446
• Bertscore:precision: 0.9972
• Bertscore:recall: 0.9975
• Bertscore:f1: 0.9973
• Ter: 1.0546
• Blerurt: 0.8889

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 3
• eval_batch_size: 3
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Wer	Cer	Chrf	Exact Match	Bertscore:precision	Bertscore:recall	Bertscore:f1	Ter	Blerurt
0.006	0.75	10000	0.0009	0.8446	0.889	0.8446	0.8603	0.0105	0.0024	99.2925	0.8446	0.9972	0.9975	0.9973	1.0546	0.8889

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2