MeMo_BERT-SA_2

This model is a fine-tuned version of MiMe-MeMo/MeMo-BERT-02 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.6537
• F1-score: 0.5806

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	0.9994	0.4947
0.9854	2.0	594	1.0281	0.5454
0.9854	3.0	891	1.1639	0.5563
0.6507	4.0	1188	1.6537	0.5806
0.6507	5.0	1485	1.7769	0.5688
0.404	6.0	1782	2.2283	0.5678
0.1759	7.0	2079	2.7271	0.5659
0.1759	8.0	2376	3.2521	0.5701
0.0906	9.0	2673	3.2777	0.5655
0.0906	10.0	2970	3.3451	0.5696
0.0717	11.0	3267	3.7121	0.5610
0.0299	12.0	3564	3.8619	0.5420
0.0299	13.0	3861	3.9670	0.5654
0.02	14.0	4158	4.2349	0.5409
0.02	15.0	4455	4.2491	0.5549
0.0074	16.0	4752	4.4144	0.5467
0.006	17.0	5049	4.3564	0.5663
0.006	18.0	5346	4.5443	0.5616
0.004	19.0	5643	4.5436	0.5424
0.004	20.0	5940	4.5248	0.5582

Framework versions

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