sew-timit-upsample_stage1

This model is a fine-tuned version of asapp/sew-tiny-100k-ft-ls100h on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.9682
• Wer: 0.9880

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: 0.001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
200.2864	1.2376	250	159.9157	1.0754
51.9265	2.4752	500	42.0889	1.0003
6.0049	3.7129	750	5.3474	0.9997
3.1387	4.9505	1000	3.1381	1.0
3.0981	6.1881	1250	3.0655	0.9993
3.0237	7.4257	1500	3.0381	0.9754
3.0058	8.6634	1750	2.9859	0.9877
2.9776	9.9010	2000	2.9682	0.9880

Framework versions

• Transformers 4.51.3
• Pytorch 2.2.1
• Datasets 3.6.0
• Tokenizers 0.21.1