languagemodel

This model is a fine-tuned version of monideep2255/XLRS-torgo on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: inf
• Wer: 1.1173

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: 0.0003
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
2.3015	3.12	400	inf	1.3984
0.6892	6.25	800	inf	1.1059
0.5069	9.37	1200	inf	1.0300
0.3596	12.5	1600	inf	1.0830
0.2571	15.62	2000	inf	1.1981
0.198	18.75	2400	inf	1.1009
0.1523	21.87	2800	inf	1.1803
0.1112	25.0	3200	inf	1.0429
0.08	28.12	3600	inf	1.1173

Framework versions

• Transformers 4.23.1
• Pytorch 1.10.0+cu113
• Datasets 1.18.3
• Tokenizers 0.13.1