phobert-large-uit-vsmec

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

• Loss: 1.5110
• Accuracy: 0.6268
• F1 Weighted: 0.6262
• F1 Macro: 0.5950

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-06
• train_batch_size: 16
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 0.1
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Weighted	F1 Macro
1.9559	1.0	466	1.9700	0.0700	0.0210	0.0316
1.6338	2.0	932	1.7810	0.3673	0.3709	0.3010
1.3372	3.0	1398	1.4718	0.5364	0.5332	0.4835
1.1957	4.0	1864	1.3979	0.6006	0.5944	0.5470
1.1208	5.0	2330	1.4322	0.5816	0.5779	0.5336
1.0224	6.0	2796	1.3638	0.6108	0.6093	0.5765
1.0143	7.0	3262	1.4218	0.6166	0.6133	0.5753
0.9404	8.0	3728	1.4227	0.6079	0.6040	0.5637
0.8748	9.0	4194	1.4276	0.6297	0.6231	0.5846
0.8368	10.0	4660	1.4311	0.6195	0.6218	0.5886
0.8524	11.0	5126	1.4132	0.6385	0.6341	0.6010
0.7606	12.0	5592	1.4652	0.6254	0.6244	0.5959
0.7555	13.0	6058	1.4591	0.6312	0.6295	0.5993
0.7664	14.0	6524	1.4833	0.6312	0.6290	0.6041
0.7142	15.0	6990	1.4986	0.6195	0.6162	0.5851
0.7380	16.0	7456	1.4939	0.6297	0.6269	0.6002
0.7724	17.0	7922	1.5013	0.6254	0.6248	0.5983
0.6518	18.0	8388	1.5146	0.6239	0.6211	0.5935
0.7474	19.0	8854	1.5110	0.6268	0.6262	0.5950

Framework versions

• Transformers 5.10.2
• Pytorch 2.11.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2