w2v-bert-2.0-yoruba-colab-CV17.0-v2

This model is a fine-tuned version of facebook/w2v-bert-2.0 on the common_voice_17_0 dataset. It achieves the following results on the evaluation set:

• Loss: 1.8450
• Wer: 0.5772

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: 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_ratio: 0.15
• training_steps: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
2.7641	3.0769	200	1.0220	0.7877
0.6684	6.1538	400	0.9003	0.6490
0.4959	9.2308	600	0.9080	0.7072
0.359	12.3077	800	0.9788	0.6147
0.2047	15.3846	1000	1.0914	0.6017
0.0858	18.4615	1200	1.4604	0.5973
0.0426	21.5385	1400	1.5740	0.5988
0.0088	24.6154	1600	1.7418	0.5753
0.0017	27.6923	1800	1.8206	0.5779
0.001	30.7692	2000	1.8450	0.5772

Framework versions

• Transformers 4.42.4
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1