Finetuning

This model is a fine-tuned version of Helsinki/opus-mt-en-mul on the samanantar dataset. source group: English target group: Hindi model: transformer

Model description

Helsinki/opus-mt-en-mul finetuned for translation task in Hindi language

Training and evaluation data

ai4bharath/samanantar

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-5
• warmup_steps: 500
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• num_epochs: 1

Benchamark Evaluation

• BLEU score on Tatoeba: 12.33471341
• BLUE score on IN-22: 26.00960094

Framework versions

• Transformers 4.42.3
• Pytorch 2.1.2
• Datasets 2.20.0
• Tokenizers 0.19.1