opus-mt-en-bkm-10e4encdec

This model is a fine-tuned version of kalese/opus-mt-en-bkm-10e64 on the arrow dataset. It achieves the following results on the evaluation set:

• Loss: 1.2402
• Bleu: 11.9039
• Gen Len: 60.1387

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: 2e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	487	1.4043	8.9062	60.1369
1.5605	2.0	974	1.3602	9.6028	60.5023
1.4769	3.0	1461	1.3262	9.99	59.6013
1.414	4.0	1948	1.3001	10.5633	59.8315
1.3618	5.0	2435	1.2779	11.0103	60.4428
1.319	6.0	2922	1.2637	11.3109	59.9477
1.2873	7.0	3409	1.2545	11.5952	60.0526
1.2611	8.0	3896	1.2471	11.7607	60.3554
1.2407	9.0	4383	1.2418	11.8586	60.0292
1.2274	10.0	4870	1.2402	11.9039	60.1387

Framework versions

• Transformers 4.39.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2