marian-finetuned-kde4-en-to-fr

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-fr on the kde4 dataset. It achieves the following results on the evaluation set:

• Loss: 0.8815
• Score: 52.2204
• Counts: [166010, 120787, 91973, 70929]
• Totals: [228361, 207343, 189354, 173335]
• Precisions: [72.69630103213771, 58.254679444205976, 48.57198686058916, 40.92018345977443]
• Bp: 0.9695
• Sys Len: 228361
• Ref Len: 235434

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.16.2
• Pytorch 1.10.0a0+0aef44c
• Datasets 1.18.3
• Tokenizers 0.11.0