st5s-es-inclusivo

This model is a fine-tuned version of flax-community/spanish-t5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1779
• Score: 55.1958

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: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 12
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Score
No log	1.0	105	0.5096	21.7567
No log	2.0	210	0.3781	38.3010
No log	3.0	315	0.2693	49.1274
No log	4.0	420	0.2274	52.3318
0.5095	5.0	525	0.2107	52.9045
0.5095	6.0	630	0.1982	54.1138
0.5095	7.0	735	0.1903	54.5379
0.5095	8.0	840	0.1861	54.7896
0.5095	9.0	945	0.1815	55.0080
0.1941	10.0	1050	0.1796	55.2191
0.1941	11.0	1155	0.1784	55.1849
0.1941	12.0	1260	0.1779	55.1958

Framework versions

• Transformers 4.39.2
• Pytorch 2.2.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2