metadata-cls-no-gov-8k

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: 0.3385
• Accuracy: 0.9481
• F1: 0.8232

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
0.553	1.6129	200	0.2901	0.9226	0.6952
0.17	3.2258	400	0.2574	0.9396	0.7657
0.1421	4.8387	600	0.2292	0.9421	0.8178
0.0916	6.4516	800	0.2295	0.9472	0.8307
0.0554	8.0645	1000	0.2675	0.9464	0.8014
0.053	9.6774	1200	0.2816	0.9498	0.8131
0.029	11.2903	1400	0.3140	0.9498	0.8223
0.0207	12.9032	1600	0.3279	0.9464	0.8176
0.0186	14.5161	1800	0.3495	0.9489	0.8263
0.0137	16.1290	2000	0.3347	0.9515	0.8313
0.0135	17.7419	2200	0.3311	0.9506	0.8402
0.0093	19.3548	2400	0.3385	0.9481	0.8232

Framework versions

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