Final-en-ne / README.md
BeebekBhz's picture
Update README.md
ad4b2a1 verified
metadata
license: cc-by-nc-4.0
tags:
  - generated_from_trainer
metrics:
  - bleu
model-index:
  - name: Final-english-nepali
    results: []
pipeline_tag: text2text-generation
library_name: transformers

Final-English-to-Nepali

This model is a fine-tuned version of [ English-to-Nepali] on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.0789
  • Bleu: 13.3322
  • Gen Len: 30.295

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
2.1526 1.0 125 2.1378 10.6717 30.905
2.1332 2.0 250 2.1030 12.5862 31.295
2.1125 3.0 375 2.0842 12.9723 29.675
2.0989 4.0 500 2.0782 13.1558 30.38
2.0989 5.0 625 2.0789 13.3322 30.295

Framework versions

  • Transformers 4.45.2
  • Pytorch 2.5.1+cu121
  • Datasets 3.2.0
  • Tokenizers 0.20.3