ABSA-SentencePair-DAPT-HARDAR-bert-base-Camel-MSA-ru1

This model is a fine-tuned version of salohnana2018/CAMEL-BERT-MSA-domianAdaption-Single-ABSA-HARD on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1067
• Accuracy: 0.8993
• F1: 0.8993
• Precision: 0.8993
• Recall: 0.8993

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-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 25
• 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
0.0591	1.0	265	0.0553	0.7925	0.7925	0.7925	0.7925
0.0352	2.0	530	0.0372	0.8932	0.8932	0.8932	0.8932
0.0229	3.0	795	0.0469	0.8866	0.8866	0.8866	0.8866
0.0146	4.0	1060	0.0533	0.8960	0.8960	0.8960	0.8960
0.0101	5.0	1325	0.0581	0.8970	0.8970	0.8970	0.8970
0.0074	6.0	1590	0.0631	0.8828	0.8828	0.8828	0.8828
0.0053	7.0	1855	0.0658	0.8823	0.8823	0.8823	0.8823
0.0051	8.0	2120	0.0723	0.8974	0.8974	0.8974	0.8974
0.0038	9.0	2385	0.0794	0.8913	0.8913	0.8913	0.8913
0.0028	10.0	2650	0.0755	0.8871	0.8871	0.8871	0.8871
0.0026	11.0	2915	0.0811	0.8894	0.8894	0.8894	0.8894
0.0019	12.0	3180	0.0853	0.8951	0.8951	0.8951	0.8951
0.0022	13.0	3445	0.0924	0.8861	0.8861	0.8861	0.8861
0.0018	14.0	3710	0.0898	0.8946	0.8946	0.8946	0.8946
0.0012	15.0	3975	0.0916	0.8856	0.8856	0.8856	0.8856
0.0013	16.0	4240	0.0999	0.8956	0.8956	0.8956	0.8956
0.0013	17.0	4505	0.1019	0.8922	0.8922	0.8922	0.8922
0.001	18.0	4770	0.1025	0.8979	0.8979	0.8979	0.8979
0.001	19.0	5035	0.1061	0.8998	0.8998	0.8998	0.8998
0.001	20.0	5300	0.1067	0.8993	0.8993	0.8993	0.8993

Framework versions

• Transformers 4.38.1
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2