PhoBertLexical-finetuned_70KURL

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: 0.1508
• Accuracy: 0.9565
• F1: 0.9565

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
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2150
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.4651	200	0.4128	0.8255	0.7984
No log	0.9302	400	0.2414	0.9038	0.9057
No log	1.3953	600	0.1894	0.9289	0.9294
No log	1.8605	800	0.1859	0.9293	0.9307
0.3236	2.3256	1000	0.1580	0.9423	0.9430
0.3236	2.7907	1200	0.1532	0.9405	0.9413
0.3236	3.2558	1400	0.1412	0.9478	0.9483
0.3236	3.7209	1600	0.1345	0.9514	0.9517
0.1539	4.1860	1800	0.1399	0.9498	0.9505
0.1539	4.6512	2000	0.1382	0.9504	0.9511
0.1539	5.1163	2200	0.1457	0.9492	0.9499
0.1539	5.5814	2400	0.1343	0.9543	0.9543
0.1146	6.0465	2600	0.1382	0.9573	0.9576
0.1146	6.5116	2800	0.1483	0.9518	0.9524
0.1146	6.9767	3000	0.1392	0.9501	0.9508
0.1146	7.4419	3200	0.1388	0.9500	0.9507
0.1146	7.9070	3400	0.1508	0.9565	0.9565

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 2.19.1
• Tokenizers 0.19.1