cita_test

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: 0.3031
• Accuracy: 0.9183
• F1: 0.9029
• Precision: 0.9113
• Recall: 0.8958

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: 4
• total_train_batch_size: 256
• 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	Precision	Recall
0.6745	2.1277	100	0.2197	0.926	0.9110	0.9256	0.8996
0.2806	4.2553	200	0.2333	0.9293	0.9140	0.9363	0.8979
0.2806	6.3830	300	0.2366	0.916	0.9027	0.8998	0.9058
0.1669	8.5106	400	0.2277	0.9283	0.9149	0.9229	0.9080
0.1669	10.6383	500	0.2593	0.922	0.9067	0.9183	0.8972
0.1132	12.7660	600	0.2683	0.9243	0.9099	0.9191	0.9022
0.1132	14.8936	700	0.2796	0.9203	0.9051	0.9148	0.8969
0.0818	17.0213	800	0.2948	0.9217	0.9064	0.9175	0.8973
0.0818	19.1489	900	0.3031	0.9183	0.9029	0.9113	0.8958

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0