yemen2016/MeMo_BERT-WSD-01

MeMo_BERT-01

This model is a fine-tuned version of MiMe-MeMo/MeMo_BERT-01 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.8264
• F1-score: 0.4148

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	61	1.5574	0.1229
No log	2.0	122	1.4491	0.1477
No log	3.0	183	1.3661	0.2112
No log	4.0	244	1.7131	0.3356
No log	5.0	305	2.0026	0.3592
No log	6.0	366	2.8807	0.2263
No log	7.0	427	2.8357	0.3625
No log	8.0	488	3.5272	0.3150
0.7496	9.0	549	3.5185	0.3402
0.7496	10.0	610	3.8264	0.4148
0.7496	11.0	671	4.5875	0.2999
0.7496	12.0	732	4.3188	0.3392
0.7496	13.0	793	4.7118	0.3098
0.7496	14.0	854	4.5895	0.3319
0.7496	15.0	915	4.6764	0.3033
0.7496	16.0	976	4.6250	0.3319
0.01	17.0	1037	4.7393	0.3176
0.01	18.0	1098	4.7690	0.3176
0.01	19.0	1159	4.7655	0.3176
0.01	20.0	1220	4.7695	0.3176

Framework versions

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