ABSA-SentencePair-DAPT-HARDAR-bert-base-Camel-MSA-ru3

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: 0.1055
• Accuracy: 0.9026
• F1: 0.9026
• Precision: 0.9026
• Recall: 0.9026

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: 23
• 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
0.0585	1.0	265	0.0424	0.8497	0.8497	0.8497	0.8497
0.0377	2.0	530	0.0422	0.8861	0.8861	0.8861	0.8861
0.0259	3.0	795	0.0421	0.8819	0.8819	0.8819	0.8819
0.017	4.0	1060	0.0477	0.8899	0.8899	0.8899	0.8899
0.0108	5.0	1325	0.0582	0.8927	0.8927	0.8927	0.8927
0.0085	6.0	1590	0.0650	0.8899	0.8899	0.8899	0.8899
0.0063	7.0	1855	0.0680	0.8922	0.8922	0.8922	0.8922
0.0043	8.0	2120	0.0705	0.9003	0.9003	0.9003	0.9003
0.0042	9.0	2385	0.0711	0.8974	0.8974	0.8974	0.8974
0.003	10.0	2650	0.0773	0.8979	0.8979	0.8979	0.8979
0.0026	11.0	2915	0.0842	0.8965	0.8965	0.8965	0.8965
0.002	12.0	3180	0.0888	0.8979	0.8979	0.8979	0.8979
0.002	13.0	3445	0.0896	0.8970	0.8970	0.8970	0.8970
0.0015	14.0	3710	0.0930	0.9008	0.9008	0.9008	0.9008
0.0012	15.0	3975	0.1008	0.9003	0.9003	0.9003	0.9003
0.0015	16.0	4240	0.0987	0.9017	0.9017	0.9017	0.9017
0.0011	17.0	4505	0.1030	0.9017	0.9017	0.9017	0.9017
0.0011	18.0	4770	0.1051	0.9003	0.9003	0.9003	0.9003
0.0011	19.0	5035	0.1046	0.9036	0.9036	0.9036	0.9036
0.001	20.0	5300	0.1055	0.9026	0.9026	0.9026	0.9026

Framework versions

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