w2v2_bert_ru

This model is a fine-tuned version of facebook/w2v-bert-2.0 on the None dataset. It achieves the following results on the evaluation set:

• Loss: inf
• Wer: 0.0538

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: 32
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.711	0.73	300	inf	0.1267
0.1026	1.46	600	inf	0.0925
0.0748	2.18	900	inf	0.0732
0.0591	2.91	1200	inf	0.0710
0.0437	3.64	1500	inf	0.0675
0.0382	4.37	1800	inf	0.0675
0.0302	5.1	2100	inf	0.0620
0.0243	5.83	2400	inf	0.0590
0.0219	6.55	2700	inf	0.0584
0.0173	7.28	3000	inf	0.0577
0.015	8.01	3300	inf	0.0560
0.0115	8.74	3600	inf	0.0551
0.0099	9.47	3900	inf	0.0538

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1