training_sentiment_analysis

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

• Loss: 0.5287
• Accuracy: 0.7977

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: 0.0003
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9299	0.21	200	0.8274	0.6387
0.7793	0.43	400	0.6643	0.7188
0.6574	0.64	600	0.5868	0.7659
0.6132	0.86	800	0.5582	0.7723
0.5791	1.07	1000	0.5516	0.7831
0.554	1.28	1200	0.5187	0.7964
0.5258	1.5	1400	0.5126	0.8034
0.5373	1.71	1600	0.5168	0.8003
0.5266	1.93	1800	0.5284	0.8028
0.5076	2.14	2000	0.5178	0.7977
0.5094	2.36	2200	0.5135	0.8028
0.5032	2.57	2400	0.5023	0.8104
0.5034	2.78	2600	0.5088	0.8047
0.4923	3.0	2800	0.5219	0.7996
0.4934	3.21	3000	0.4905	0.8130
0.4798	3.43	3200	0.4908	0.8098
0.4831	3.64	3400	0.4875	0.8073
0.4707	3.85	3600	0.4986	0.8073
0.4674	4.07	3800	0.5196	0.8104
0.4535	4.28	4000	0.4896	0.8098
0.464	4.5	4200	0.5175	0.8079
0.4715	4.71	4400	0.5002	0.8028
0.468	4.93	4600	0.4883	0.8111
0.4645	5.14	4800	0.5187	0.8041
0.445	5.35	5000	0.4928	0.8066
0.4558	5.57	5200	0.4870	0.8079
0.4405	5.78	5400	0.4985	0.8104
0.4648	6.0	5600	0.4842	0.8060
0.435	6.21	5800	0.4911	0.8117
0.437	6.42	6000	0.4854	0.8085
0.4588	6.64	6200	0.4879	0.8085
0.4342	6.85	6400	0.4922	0.8104
0.4347	7.07	6600	0.4911	0.8142
0.4326	7.28	6800	0.4914	0.8079
0.4267	7.49	7000	0.4917	0.8104
0.4241	7.71	7200	0.4887	0.8136
0.4376	7.92	7400	0.5122	0.8079
0.4323	8.14	7600	0.4909	0.8098
0.4264	8.35	7800	0.4882	0.8142
0.4175	8.57	8000	0.5091	0.8053
0.4228	8.78	8200	0.5060	0.8098
0.4189	8.99	8400	0.4941	0.8092
0.4161	9.21	8600	0.5010	0.8174
0.4078	9.42	8800	0.4949	0.8079
0.4201	9.64	9000	0.5017	0.8073
0.4141	9.85	9200	0.4985	0.8092
0.4132	10.06	9400	0.5032	0.8053
0.4043	10.28	9600	0.5038	0.8130
0.4187	10.49	9800	0.4981	0.8104
0.3827	10.71	10000	0.5126	0.8073
0.4074	10.92	10200	0.5088	0.8073
0.4013	11.13	10400	0.5061	0.8073
0.3888	11.35	10600	0.5013	0.8085
0.3855	11.56	10800	0.4993	0.8060
0.3924	11.78	11000	0.5075	0.8085
0.4046	11.99	11200	0.4999	0.8028
0.3957	12.21	11400	0.5089	0.8034
0.381	12.42	11600	0.5208	0.8073
0.3906	12.63	11800	0.5137	0.8066
0.3734	12.85	12000	0.5183	0.8041
0.3928	13.06	12200	0.5069	0.8066
0.3774	13.28	12400	0.5086	0.8009
0.3892	13.49	12600	0.4967	0.8060
0.372	13.7	12800	0.5043	0.8041
0.388	13.92	13000	0.5095	0.8073
0.3754	14.13	13200	0.5104	0.8022
0.3639	14.35	13400	0.5263	0.7983
0.3795	14.56	13600	0.5146	0.8015
0.3792	14.78	13800	0.5066	0.8041
0.3589	14.99	14000	0.5136	0.8079
0.3624	15.2	14200	0.5237	0.8022
0.3659	15.42	14400	0.5166	0.8060
0.3657	15.63	14600	0.5178	0.8003
0.359	15.85	14800	0.5152	0.7983
0.3677	16.06	15000	0.5212	0.8034
0.3521	16.27	15200	0.5324	0.8003
0.3589	16.49	15400	0.5238	0.8041
0.3695	16.7	15600	0.5113	0.7977
0.3606	16.92	15800	0.5137	0.7983
0.3581	17.13	16000	0.5131	0.7996
0.3488	17.34	16200	0.5270	0.7990
0.3499	17.56	16400	0.5236	0.7964
0.3603	17.77	16600	0.5187	0.8003
0.3578	17.99	16800	0.5224	0.8022
0.3449	18.2	17000	0.5228	0.7990
0.3418	18.42	17200	0.5287	0.8009
0.3334	18.63	17400	0.5322	0.7996
0.3567	18.84	17600	0.5294	0.7983
0.3541	19.06	17800	0.5250	0.8003
0.365	19.27	18000	0.5246	0.7983
0.337	19.49	18200	0.5278	0.7977
0.3301	19.7	18400	0.5283	0.7990
0.3421	19.91	18600	0.5287	0.7977

Framework versions

• PEFT 0.10.0
• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2