t5_es_weight_4_2

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

• Loss: 0.0201
• Accuracy: 0.9955
• F1: 0.9958

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 64
• total_train_batch_size: 4096
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.7207	6.8817	50	0.6639	0.631	0.7122
0.6227	13.7634	100	0.4261	0.8935	0.9021
0.2336	20.6452	150	0.0745	0.979	0.9803
0.0593	27.5269	200	0.0453	0.9845	0.9853
0.0286	34.4086	250	0.0287	0.9915	0.992
0.0149	41.2903	300	0.0201	0.995	0.9953
0.0078	48.1720	350	0.0194	0.996	0.9962
0.0047	55.0538	400	0.0203	0.9965	0.9967
0.0033	61.9355	450	0.0203	0.996	0.9962
0.0024	68.8172	500	0.0192	0.996	0.9962
0.0016	75.6989	550	0.0194	0.996	0.9962
0.0008	82.5806	600	0.0219	0.996	0.9962
0.0008	89.4624	650	0.0249	0.996	0.9963
0.0026	96.3441	700	0.0201	0.9955	0.9958

Framework versions

• Transformers 4.40.0
• Pytorch 2.4.1+cu121
• Datasets 3.1.0
• Tokenizers 0.19.1