metadata
license: apache-2.0
base_model: google/flan-t5-base
tags:
- simplification
- generated_from_trainer
metrics:
- bleu
model-index:
- name: flant5
results: []
flant5
This model is a fine-tuned version of google/flan-t5-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1632
- Bleu: 32.6829
- Gen Len: 17.6119
Model description
El modelo FLAN T5 Base es una variante del modelo de lenguaje T5 (Text-To-Text Transfer Transformer) desarrollado por Google. T5 es una arquitectura de modelo de lenguaje basada en transformers que ha demostrado un rendimiento sobresaliente en una amplia gama de tareas de procesamiento de lenguaje natural (NLP).
Intended uses & limitations
En este caso el modelo es un traductor automático que simplifica el francés.
Training and evaluation data
Se utilizan 266 filas de la tabla como entrenamiento y 67 como test.
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5.6e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| No log | 1.0 | 67 | 1.1644 | 32.6829 | 17.6119 |
| No log | 2.0 | 134 | 1.1632 | 32.6829 | 17.6119 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2