Hugging Face's logo

clui
/
opus-it-pl-v1

Text2Text Generation
Transformers
Safetensors
marian
Generated from Trainer
Inference Endpoints
Model card Files Community
opus-it-pl-v1 / README.md
clui's picture
clui
End of training
3c28567 verified
preview code
|
raw
history blame contribute delete
1.82 kB
metadata
library_name: transformers
license: apache-2.0
base_model: Helsinki-NLP/opus-mt-it-en
tags:
  - generated_from_trainer
metrics:
  - bleu
model-index:
  - name: opus-it-pl-v1
    results: []

opus-it-pl-v1

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

  • Loss: 1.3595
  • Bleu: 9.6913
  • Gen Len: 36.0054

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
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
2.0728 1.0 8066 1.6218 7.3281 36.9712
1.8058 2.0 16132 1.4415 8.9077 36.0686
1.6925 3.0 24198 1.3778 9.5031 36.0025
1.6379 4.0 32264 1.3595 9.6913 36.0054

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0
  • Datasets 2.19.1
  • Tokenizers 0.19.1