LMPT_project / README.md
rujengelal's picture
End of training
b583491 verified
|
raw
history blame
2.1 kB
metadata
license: cc-by-nc-4.0
base_model: facebook/nllb-200-distilled-600M
tags:
  - generated_from_trainer
metrics:
  - bleu
model-index:
  - name: LMPT_project
    results: []

LMPT_project

This model is a fine-tuned version of facebook/nllb-200-distilled-600M on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.7558
  • Bleu: 12.2835
  • Gen Len: 44.0

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: 10
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
No log 1.0 13 1.9988 0.6722 37.35
No log 2.0 26 1.8930 0.7123 37.55
No log 3.0 39 1.8316 12.8352 36.55
No log 4.0 52 1.7920 17.2952 36.3
No log 5.0 65 1.7743 16.9058 36.8
No log 6.0 78 1.7652 17.4576 36.15
No log 7.0 91 1.7601 11.2528 44.4
No log 8.0 104 1.7578 12.2835 44.0
No log 9.0 117 1.7556 12.2835 44.0
No log 10.0 130 1.7558 12.2835 44.0

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.1.2
  • Datasets 2.18.0
  • Tokenizers 0.15.2