En-Nso_update3

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-nso on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.4218
• Bleu: 24.5765

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: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
3.6568	1.0	867	3.0185	18.4004
2.7574	2.0	1734	2.7774	20.3167
2.4522	3.0	2601	2.6436	22.1868
2.3298	4.0	3468	2.5732	22.6221
2.1563	5.0	4335	2.5225	22.6937
2.0177	6.0	5202	2.4917	23.2204
1.9407	7.0	6069	2.4656	23.3616
1.8758	8.0	6936	2.4509	23.5496
1.8167	9.0	7803	2.4426	23.6263
1.7566	10.0	8670	2.4345	24.0730
1.7254	11.0	9537	2.4281	24.1627
1.7088	12.0	10404	2.4252	24.1109
1.6731	13.0	11271	2.4226	24.1018
1.6574	14.0	12138	2.4211	23.9186
1.6481	15.0	13005	2.4218	24.1323

Framework versions

• Transformers 4.20.1
• Pytorch 1.12.0+cu113
• Datasets 2.3.2
• Tokenizers 0.12.1