t5_es_farshad_half_4_1

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: 0.0490
• Accuracy: 0.9916
• F1: 0.9919

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
0.6889	5.8501	50	0.6724	0.6073	0.5334
0.6445	11.7002	100	0.5323	0.8022	0.8091
0.3119	17.5503	150	0.1187	0.9649	0.9656
0.0967	23.4004	200	0.0648	0.9794	0.9800
0.0549	29.2505	250	0.0500	0.9858	0.9862
0.0359	35.1005	300	0.0465	0.9884	0.9888
0.0248	40.9506	350	0.0443	0.9887	0.9891
0.0183	46.8007	400	0.0404	0.9898	0.9902
0.0139	52.6508	450	0.0445	0.9890	0.9893
0.0111	58.5009	500	0.0559	0.9887	0.9890
0.0087	64.3510	550	0.0486	0.9893	0.9896
0.0081	70.2011	600	0.0440	0.9910	0.9913
0.0065	76.0512	650	0.0410	0.9919	0.9921
0.0045	81.9013	700	0.0596	0.9893	0.9896
0.0042	87.7514	750	0.0475	0.9898	0.9902
0.0036	93.6015	800	0.0490	0.9916	0.9919

Framework versions

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