t5-absa-plusplus-atsc

This model is a fine-tuned version of google-t5/t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4055
• F1: 58.38
• Precision: 57.62
• Recall: 59.15
• N Tp: 223
• N Pred: 387
• N Gold: 377

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 26
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Precision	Recall	N Tp	N Pred	N Gold
0.6296	4.3860	500	0.3825	53.96	54.62	53.32	201	368	377
0.3482	8.7719	1000	0.4051	58.27	57.66	58.89	222	385	377
0.3542	10.0	1140	0.4055	58.38	57.62	59.15	223	387	377

Framework versions

• Transformers 5.6.2
• Pytorch 2.11.0+cu130
• Datasets 4.8.4
• Tokenizers 0.22.2