t5_es_weight_1_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: 0.0494
• Accuracy: 0.991
• F1: 0.9915

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
0.7041	6.8817	50	0.6701	0.706	0.7330
0.6404	13.7634	100	0.4985	0.832	0.8397
0.3013	20.6452	150	0.1406	0.9535	0.9555
0.0832	27.5269	200	0.0844	0.9745	0.9757
0.041	34.4086	250	0.0555	0.9835	0.9844
0.0226	41.2903	300	0.0494	0.9845	0.9853
0.0126	48.1720	350	0.0315	0.99	0.9906
0.0075	55.0538	400	0.0241	0.9945	0.9948
0.0045	61.9355	450	0.0380	0.9915	0.9920
0.0025	68.8172	500	0.0324	0.9925	0.9929
0.0018	75.6989	550	0.0487	0.9905	0.9910
0.0016	82.5806	600	0.0395	0.992	0.9925
0.0011	89.4624	650	0.0409	0.993	0.9934
0.0011	96.3441	700	0.0494	0.991	0.9915

Framework versions

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