gechim/cls-metadaweb-dataGemini

cls-meta-test

This model is a fine-tuned version of vinai/phobert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7757
• Accuracy: 0.8259
• F1: 0.8036

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: 2e-05
• train_batch_size: 64
• eval_batch_size: 64
• 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	Accuracy	F1
0.9834	1.8182	200	0.5062	0.8314	0.6869
0.4462	3.6364	400	0.9367	0.6784	0.7354
0.3124	5.4545	600	0.9898	0.5084	0.6351
0.2517	7.2727	800	0.5842	0.8350	0.8134
0.1859	9.0909	1000	0.5760	0.8339	0.8138
0.1493	10.9091	1200	0.6501	0.8299	0.8145
0.1301	12.7273	1400	0.6403	0.8310	0.8219
0.1126	14.5455	1600	0.7262	0.8296	0.8110
0.0946	16.3636	1800	0.7555	0.8237	0.8048
0.0861	18.1818	2000	0.7576	0.8270	0.8085
0.0756	20.0	2200	0.7757	0.8259	0.8036

Framework versions

• Transformers 4.42.4
• Pytorch 2.1.2
• Datasets 2.20.0
• Tokenizers 0.19.1