Gopal-finetuned-custom-en-to-ru

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-ru on an unknown dataset.

Model description

This is the model fine tuned by me on my custom dataset, the dataset contains communication domain parallel corpuses.

Intended uses & limitations

This model is used for customised purposes and people are advised to fine tune it on the basis of there requirement

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-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: 200
• mixed_precision_training: Native AMP

Training results

The bleu score: 31.08

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2