Ce modèle a été finetuné avec 4000 + 1000 offres. (Train + Validation) Le jeu de données a été oversamplé de façon à créer une répartition des classes homogène.
t5-base-fr-sum-cnndm_finetuned_12_06_13_40
This model is a fine-tuned version of plguillou/t5-base-fr-sum-cnndm on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.5496
- Rouge1: 0.161
- Rouge2: 0.0388
- Rougel: 0.1319
- Rougelsum: 0.1319
- Gen Len: 20.0
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: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 1.7344 | 1.0 | 4000 | 1.5496 | 0.161 | 0.0388 | 0.1319 | 0.1319 | 20.0 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1
- Downloads last month
- 1