roberta-base-go_emotions

This model is a fine-tuned version of FacebookAI/roberta-base on an unknown dataset. It achieves the following results on the validation set:

• Loss: 0.1086
• Accuracy: 0.4561
• Roc Auc: 0.9064
• Micro Precision: 0.6063
• Micro Recall: 0.5340
• Micro F1: 0.5679
• Macro Precision: 0.5800
• Macro Recall: 0.4344
• Macro F1: 0.4649
• Weighted Precision: 0.5994
• Weighted Recall: 0.5340
• Weighted F1: 0.5591

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: 8
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Roc Auc	Micro Precision	Micro Recall	Micro F1	Macro Precision	Macro Recall	Macro F1	Weighted Precision	Weighted Recall	Weighted F1
0.1047	1.0	5427	0.0973	0.3616	0.8668	0.7390	0.3710	0.4940	0.3548	0.1954	0.2192	0.5670	0.3710	0.4098
0.09	2.0	10854	0.0876	0.4195	0.9037	0.7497	0.4276	0.5446	0.5715	0.2731	0.3243	0.6961	0.4276	0.4875
0.0821	3.0	16281	0.0850	0.4477	0.9137	0.7294	0.4627	0.5662	0.5692	0.3174	0.3799	0.6893	0.4627	0.5258
0.0774	4.0	21708	0.0851	0.4591	0.9178	0.6930	0.4876	0.5725	0.5768	0.3765	0.4273	0.6745	0.4876	0.5435
0.0736	5.0	27135	0.0856	0.4657	0.9208	0.6844	0.4989	0.5771	0.5741	0.3909	0.4448	0.6715	0.4989	0.5557
0.0714	6.0	32562	0.0866	0.4619	0.9171	0.6674	0.4991	0.5711	0.5593	0.3845	0.4386	0.6529	0.4991	0.5515
0.0673	7.0	37989	0.0883	0.4607	0.9209	0.6585	0.5038	0.5708	0.5197	0.4151	0.4522	0.6417	0.5038	0.5539
0.0604	8.0	43416	0.0902	0.4773	0.9171	0.6530	0.5252	0.5822	0.5623	0.4192	0.4629	0.6316	0.5252	0.5646
0.0593	9.0	48843	0.0926	0.4714	0.9165	0.6319	0.5263	0.5743	0.5850	0.4208	0.4612	0.6235	0.5263	0.5625
0.0557	10.0	54270	0.0959	0.4639	0.9155	0.6319	0.5229	0.5723	0.5710	0.4340	0.4705	0.6227	0.5229	0.5602
0.0512	11.0	59697	0.0985	0.4631	0.9147	0.6203	0.5266	0.5696	0.5656	0.4470	0.4754	0.6162	0.5266	0.5605
0.0478	12.0	65124	0.1013	0.4644	0.9116	0.6191	0.5279	0.5699	0.5588	0.4426	0.4776	0.6159	0.5279	0.5607
0.0449	13.0	70551	0.1036	0.4696	0.9080	0.6188	0.5354	0.5741	0.5594	0.4395	0.4729	0.6073	0.5354	0.5618
0.042	14.0	75978	0.1055	0.4700	0.9071	0.6131	0.5409	0.5747	0.5761	0.4399	0.4698	0.6013	0.5409	0.5638
0.0392	15.0	81405	0.1086	0.4561	0.9064	0.6063	0.5340	0.5679	0.5800	0.4344	0.4649	0.5994	0.5340	0.5591

Framework versions

• Transformers 4.45.2
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.20.3