opus-mt-mr-en_mr_en

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

• Loss: 0.3867

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: 0.0005
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 2048
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 16
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	32	0.5244
No log	1.99	64	0.4824
No log	2.99	96	0.4601
No log	3.99	128	0.4412
No log	4.99	160	0.4330
No log	5.98	192	0.4258
No log	6.98	224	0.4214
No log	7.98	256	0.4193
No log	9.0	288	0.4108
No log	9.99	320	0.4034
No log	10.99	352	0.3983
No log	11.99	384	0.3938
No log	12.99	416	0.3901
No log	13.98	448	0.3889
No log	14.98	480	0.3871
No log	15.98	512	0.3867

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.2
• Datasets 2.15.0
• Tokenizers 0.15.2