t5_es_weight_2_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: 0.0241
• Accuracy: 0.997
• F1: 0.9972

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
0.7035	6.8817	50	0.6738	0.7045	0.7288
0.6463	13.7634	100	0.5114	0.8975	0.9015
0.2909	20.6452	150	0.0785	0.977	0.9783
0.0595	27.5269	200	0.0455	0.987	0.9878
0.0286	34.4086	250	0.0283	0.992	0.9925
0.0158	41.2903	300	0.0219	0.9945	0.9948
0.0086	48.1720	350	0.0180	0.996	0.9962
0.0048	55.0538	400	0.0172	0.9955	0.9958
0.0031	61.9355	450	0.0223	0.9955	0.9958
0.002	68.8172	500	0.0199	0.9955	0.9958
0.0012	75.6989	550	0.0201	0.9965	0.9967
0.0008	82.5806	600	0.0190	0.997	0.9972
0.0008	89.4624	650	0.0205	0.997	0.9972
0.0007	96.3441	700	0.0241	0.997	0.9972

Framework versions

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