t5_es_weight_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: 0.0151
• Accuracy: 0.9975
• F1: 0.9977

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
0.7279	6.8817	50	0.6709	0.583	0.7070
0.6363	13.7634	100	0.4762	0.888	0.8886
0.2667	20.6452	150	0.0757	0.973	0.9746
0.0582	27.5269	200	0.0399	0.9875	0.9882
0.0273	34.4086	250	0.0253	0.9925	0.9929
0.0147	41.2903	300	0.0179	0.995	0.9953
0.0077	48.1720	350	0.0177	0.996	0.9963
0.0049	55.0538	400	0.0152	0.9965	0.9967
0.0031	61.9355	450	0.0153	0.9975	0.9977
0.0023	68.8172	500	0.0170	0.997	0.9972
0.0015	75.6989	550	0.0161	0.998	0.9981
0.0017	82.5806	600	0.0192	0.997	0.9972
0.0012	89.4624	650	0.0148	0.998	0.9981
0.0005	96.3441	700	0.0151	0.9975	0.9977

Framework versions

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