chinese-hubert-base-CV16

This model is a fine-tuned version of TencentGameMate/chinese-hubert-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 7.0618
• Wer: 1.0

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.PAGED_LION_8BIT and the args are: No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.08333333333333333
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
7.3559	0.5	752	7.1096	1.0
7.368	1.0	1504	7.1509	1.0
7.2041	1.5	2256	7.0803	1.0
7.1962	2.0	3008	7.0860	1.0
7.2482	2.5	3760	7.0603	1.0
7.2172	3.0	4512	7.0618	1.0

Framework versions

• Transformers 4.48.2
• Pytorch 2.2.1+cu121
• Datasets 2.17.1
• Tokenizers 0.21.0