fineturning-without-pretraining-2

This model is a fine-tuned version of on the common_voice_1_0 dataset. It achieves the following results on the evaluation set:

• Loss: 779.5451
• Wer: 0.9999

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: 35
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1829.2385	4.29	500	781.0485	0.9999
1459.4001	8.58	1000	777.1782	0.9999
1454.826	12.88	1500	777.3484	0.9999
1448.8867	17.17	2000	788.0052	0.9999
1445.467	21.46	2500	779.9430	0.9999
1438.5691	25.75	3000	786.7927	0.9999
1445.318	30.04	3500	789.1374	0.9999
1442.6181	34.33	4000	779.5451	0.9999

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2