Edit model card

spanish-spanglish

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-es on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.9401
  • Bleu: 29.8764
  • Gen Len: 15.725

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: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
0.5094 1.0 10 0.9295 28.2464 16.325
0.385 2.0 20 0.9574 24.2148 13.925
0.3085 3.0 30 0.9097 30.5954 15.925
0.2548 4.0 40 0.9455 25.5705 14.6
0.2283 5.0 50 0.9268 28.5459 15.675
0.2193 6.0 60 0.9306 28.1665 15.65
0.1902 7.0 70 0.9363 29.2711 15.525
0.1784 8.0 80 0.9349 30.2007 15.75
0.1647 9.0 90 0.9393 29.8764 15.725
0.1532 10.0 100 0.9401 29.8764 15.725

Framework versions

  • Transformers 4.40.1
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
7
Safetensors
Model size
77.5M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for drewcurran/spanish-spanglish

Finetuned
(21)
this model