MeMo_BERT-SA_3

This model is a fine-tuned version of MiMe-MeMo/MeMo-BERT-03 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.4776
• F1-score: 0.7659

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-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1-score
No log	1.0	297	0.7872	0.7116
0.5803	2.0	594	0.9097	0.7575
0.5803	3.0	891	1.2182	0.7518
0.2342	4.0	1188	1.3653	0.7618
0.2342	5.0	1485	1.4776	0.7659
0.0843	6.0	1782	1.7107	0.7535
0.0179	7.0	2079	1.7869	0.7655
0.0179	8.0	2376	1.9367	0.7422
0.0106	9.0	2673	2.2548	0.7404
0.0106	10.0	2970	1.9947	0.7636
0.0078	11.0	3267	2.2236	0.7555
0.0013	12.0	3564	2.1156	0.7573
0.0013	13.0	3861	2.2286	0.7413
0.0039	14.0	4158	2.2651	0.7411
0.0039	15.0	4455	2.3064	0.7411
0.0	16.0	4752	2.2598	0.7500
0.0004	17.0	5049	2.1443	0.7529
0.0004	18.0	5346	2.1092	0.7637
0.0021	19.0	5643	2.3133	0.7619
0.0021	20.0	5940	2.2768	0.7656

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2