metadata-cls-no-gov

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: 0.3806
• Accuracy: 0.9303
• F1: 0.8372

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
0.5915	1.9608	200	0.2063	0.9406	0.7585
0.1888	3.9216	400	0.1895	0.9426	0.8653
0.1081	5.8824	600	0.2354	0.9344	0.8188
0.0761	7.8431	800	0.2972	0.9324	0.8176
0.0483	9.8039	1000	0.2985	0.9324	0.8385
0.0348	11.7647	1200	0.3409	0.9283	0.8521
0.0271	13.7255	1400	0.3593	0.9283	0.8211
0.019	15.6863	1600	0.3608	0.9324	0.8352
0.0136	17.6471	1800	0.3729	0.9324	0.8358
0.0095	19.6078	2000	0.3806	0.9303	0.8372

Framework versions

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