my_awesome_opus_books_model

This model is a fine-tuned version of Helsinki-NLP/opus-mt-tc-big-en-ko on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 4.4857
• Bleu: 0.0931
• Gen Len: 32.375

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
4.8207	1.0	25	4.6601	0.3081	14.015
4.546	2.0	50	4.5422	0.1017	11.4
4.1457	3.0	75	4.4989	0.2841	13.285
4.0581	4.0	100	4.4812	0.1455	16.07
3.9499	5.0	125	4.4730	0.0931	30.79
3.8727	6.0	150	4.4788	0.0814	32.83
3.6131	7.0	175	4.4823	0.0742	29.73
3.5434	8.0	200	4.4830	0.0746	29.71
3.408	9.0	225	4.4857	0.094	31.465
3.4915	10.0	250	4.4857	0.0931	32.375

Framework versions

• Transformers 4.55.0
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.4