metadata

base_model: vinai/phobert-base-v2
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - recall
  - precision
model-index:
  - name: cls-comment-phobert-base-v2-v3.2
    results: []

cls-comment-phobert-base-v2-v3.2

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: 0.6534
Accuracy: 0.9222
F1 Score: 0.9110
Recall: 0.9115
Precision: 0.9149

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: 1e-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_ratio: 0.1
training_steps: 4000
label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Score	Recall	Precision
1.8971	0.87	100	1.7443	0.4064	0.0826	0.1429	0.0581
1.5979	1.73	200	1.3562	0.6147	0.2496	0.2794	0.2703
1.2633	2.6	300	1.0631	0.7399	0.4779	0.4720	0.5906
1.0047	3.46	400	0.8834	0.8255	0.5990	0.6173	0.5876
0.8689	4.33	500	0.8139	0.8434	0.6169	0.6459	0.5936
0.7901	5.19	600	0.7564	0.8659	0.6766	0.6790	0.7339
0.7217	6.06	700	0.7114	0.8886	0.8211	0.8018	0.8731
0.6635	6.93	800	0.6800	0.9041	0.8787	0.8775	0.8829
0.6119	7.79	900	0.6681	0.9073	0.8870	0.8844	0.8917
0.5888	8.66	1000	0.6620	0.9130	0.8948	0.8961	0.8975
0.5721	9.52	1100	0.6746	0.9098	0.8931	0.9058	0.8849
0.5521	10.39	1200	0.6434	0.9231	0.9061	0.9038	0.9086
0.5451	11.26	1300	0.6397	0.9231	0.9095	0.9042	0.9162
0.5315	12.12	1400	0.6552	0.9174	0.9044	0.9128	0.8993
0.5182	12.99	1500	0.6483	0.9206	0.9009	0.9063	0.8973
0.5078	13.85	1600	0.6534	0.9222	0.9110	0.9115	0.9149
0.498	14.72	1700	0.6493	0.9255	0.9080	0.9016	0.9152
0.4998	15.58	1800	0.6451	0.9244	0.9073	0.9146	0.9009
0.49	16.45	1900	0.6693	0.9174	0.9018	0.9039	0.9019
0.4882	17.32	2000	0.6501	0.9236	0.9057	0.9101	0.9023
0.4859	18.18	2100	0.6591	0.9258	0.9110	0.9126	0.9113

Framework versions

Transformers 4.38.2
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.15.2