wcBERTaAttSpmm-ggTranslate-senticPolarEmotion-bully-f1

This model is a fine-tuned version of airesearch/wangchanberta-base-att-spm-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5229
• Accuracy: 0.7448
• Precision: 0.7291
• Recall: 0.7448
• F1 Score: 0.7303

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: 5e-06
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 Score
0.551	1.0	120	0.5535	0.7427	0.7285	0.7427	0.7311
0.5476	2.0	240	0.5530	0.7406	0.7259	0.7406	0.7285
0.5352	3.0	360	0.5528	0.7354	0.7261	0.7354	0.7294
0.5482	4.0	480	0.5531	0.7312	0.7278	0.7312	0.7293
0.5386	5.0	600	0.5547	0.7228	0.7236	0.7228	0.7232
0.5391	6.0	720	0.5467	0.7427	0.7303	0.7427	0.7335
0.5495	7.0	840	0.5506	0.7395	0.7305	0.7395	0.7337
0.5305	8.0	960	0.5444	0.7427	0.7321	0.7427	0.7353
0.5183	9.0	1080	0.5326	0.7448	0.7320	0.7448	0.7349
0.5065	10.0	1200	0.5218	0.7479	0.7314	0.7479	0.7297
0.4753	11.0	1320	0.5207	0.7469	0.7317	0.7469	0.7330
0.4731	12.0	1440	0.5233	0.7458	0.7302	0.7458	0.7312
0.4828	13.0	1560	0.5243	0.7458	0.7302	0.7458	0.7312
0.4662	14.0	1680	0.5229	0.7458	0.7306	0.7458	0.7321
0.472	15.0	1800	0.5229	0.7448	0.7291	0.7448	0.7303

Framework versions

• Transformers 4.42.3
• Pytorch 2.1.2
• Datasets 2.20.0
• Tokenizers 0.19.1