memo3_FGN

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: 1.3348
• F1-score: 0.8316

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	120	0.4934	0.8027
No log	2.0	240	0.5542	0.8121
No log	3.0	360	0.8229	0.8168
No log	4.0	480	0.9300	0.8123
0.4028	5.0	600	1.1413	0.8057
0.4028	6.0	720	1.3327	0.7989
0.4028	7.0	840	1.5178	0.7851
0.4028	8.0	960	1.1979	0.8299
0.0435	9.0	1080	1.2149	0.8307
0.0435	10.0	1200	1.2327	0.8302
0.0435	11.0	1320	1.2521	0.8312
0.0435	12.0	1440	1.2734	0.8304
0.0238	13.0	1560	1.3348	0.8316
0.0238	14.0	1680	1.3715	0.8181
0.0238	15.0	1800	1.4095	0.8064
0.0238	16.0	1920	1.4157	0.8179
0.0099	17.0	2040	1.4234	0.8179
0.0099	18.0	2160	1.4038	0.8127
0.0099	19.0	2280	1.4404	0.8117
0.0099	20.0	2400	1.4436	0.8117

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1