opus-mt-finetuned-id-to-su

This model is a fine-tuned version of hermanshid/opus-mt-finetuned-id-to-su on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.2138
• Bleu: 30.2906
• Gen Len: 61.8136

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.6062	1.0	2500	1.4921	25.6912	62.7002
1.4351	2.0	5000	1.3702	27.6125	62.2532
1.3364	3.0	7500	1.3116	28.515	62.4072
1.2696	4.0	10000	1.2821	29.279	61.8128
1.2245	5.0	12500	1.2532	29.6096	61.631
1.1812	6.0	15000	1.2373	29.8749	61.9136
1.1589	7.0	17500	1.2309	29.8342	62.038
1.1268	8.0	20000	1.2186	30.0483	61.9544
1.1055	9.0	22500	1.2158	30.3904	61.5692
1.1035	10.0	25000	1.2138	30.2906	61.8136

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3