t5_es_weight_2_4

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: 0.0167
• Accuracy: 0.9965
• F1: 0.9967

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
0.7403	6.8817	50	0.6830	0.551	0.3455
0.6354	13.7634	100	0.4776	0.9015	0.9015
0.2911	20.6452	150	0.1007	0.9705	0.9721
0.0694	27.5269	200	0.0499	0.983	0.9839
0.0293	34.4086	250	0.0317	0.9915	0.9920
0.0154	41.2903	300	0.0231	0.9925	0.9929
0.008	48.1720	350	0.0187	0.9955	0.9958
0.0045	55.0538	400	0.0180	0.9945	0.9948
0.0028	61.9355	450	0.0195	0.995	0.9953
0.0021	68.8172	500	0.0224	0.9955	0.9958
0.0015	75.6989	550	0.0191	0.996	0.9962
0.0011	82.5806	600	0.0236	0.9965	0.9967
0.001	89.4624	650	0.0240	0.996	0.9962
0.0007	96.3441	700	0.0167	0.9965	0.9967

Framework versions

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