Terjman-Large-v0 / README.md
BounharAbdelaziz's picture
Update README.md
5faf25b verified
|
raw
history blame
986 Bytes
metadata
tags:
  - generated_from_trainer
model-index:
  - name: results
    results: []
datasets:
  - BounharAbdelaziz/English-to-Moroccan-Darija
language:
  - ar
  - en
metrics:
  - bleu

Terjman-Large-v2

This is a translation model from English to Moroccan darija. It is a finetuned version of "Helsinki-NLP/opus-mt-tc-big-en-ar" on the "BounharAbdelaziz/English-to-Moroccan-Darija" dataset.

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 3e-05
  • train_batch_size: 96
  • eval_batch_size: 96
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.03
  • num_epochs: 150

Framework versions

  • Transformers 4.39.2
  • Pytorch 2.2.2+cpu
  • Datasets 2.18.0
  • Tokenizers 0.15.2