hybrid_lexical_hosting_64_15_v2

This model is a fine-tuned version of vinai/phobert-base-v2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8164
• Accuracy: 0.8576
• F1: 0.8576

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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2580
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
42.9086	1.0	1032	5.8377	0.5453	0.4382
24.7451	2.0	2064	7.7162	0.5456	0.4388
17.4566	3.0	3096	2.5417	0.6740	0.6637
12.0533	4.0	4128	4.9885	0.8127	0.8126
8.4531	5.0	5160	3.3348	0.8201	0.8197
6.3636	6.0	6192	1.5283	0.8325	0.8324
5.0662	7.0	7224	2.4994	0.8383	0.8382
4.3726	8.0	8256	2.6446	0.8354	0.8351
3.7847	9.0	9288	1.3979	0.8386	0.8386
3.4433	10.0	10320	2.2144	0.8449	0.8449
3.0554	11.0	11352	1.0773	0.8380	0.8367
2.7706	12.0	12384	0.8758	0.8532	0.8533
2.5067	13.0	13416	1.0391	0.8527	0.8526
2.4275	14.0	14448	1.0730	0.8512	0.8513
2.3293	15.0	15480	0.8164	0.8576	0.8576

Framework versions

• Transformers 4.57.1
• Pytorch 2.6.0+cu124
• Datasets 4.1.1
• Tokenizers 0.22.1