absa-restaurants-bert-base-uncased

This model is a fine-tuned version of google-bert/bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7275
• Accuracy: 0.8790

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: 256
• eval_batch_size: 256
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2396	1.0	23	0.4017	0.8679
0.1873	2.0	46	0.4208	0.8714
0.1489	3.0	69	0.4428	0.8679
0.1123	4.0	92	0.4690	0.8721
0.0867	5.0	115	0.5119	0.8776
0.0672	6.0	138	0.5222	0.8741
0.0753	7.0	161	0.5781	0.8721
0.0732	8.0	184	0.5416	0.8672
0.0589	9.0	207	0.5674	0.8769
0.0507	10.0	230	0.5819	0.8734
0.0466	11.0	253	0.5871	0.8769
0.0452	12.0	276	0.5818	0.8797
0.0444	13.0	299	0.6002	0.8783
0.0376	14.0	322	0.5886	0.8714
0.0315	15.0	345	0.5846	0.8790
0.0295	16.0	368	0.6067	0.8693
0.0292	17.0	391	0.6186	0.8734
0.0226	18.0	414	0.6222	0.8783
0.0305	19.0	437	0.6400	0.8734
0.0245	20.0	460	0.6544	0.8721
0.0241	21.0	483	0.6438	0.8748
0.0208	22.0	506	0.6512	0.8734
0.0225	23.0	529	0.6580	0.8748
0.0215	24.0	552	0.6401	0.8790
0.019	25.0	575	0.6705	0.8679
0.0197	26.0	598	0.6809	0.8769
0.0192	27.0	621	0.6569	0.8748
0.0182	28.0	644	0.6763	0.8748
0.0173	29.0	667	0.6846	0.8790
0.0151	30.0	690	0.7100	0.8776
0.0171	31.0	713	0.6979	0.8748
0.0185	32.0	736	0.6947	0.8762
0.0161	33.0	759	0.6979	0.8755
0.0169	34.0	782	0.7015	0.8741
0.0158	35.0	805	0.7085	0.8804
0.0144	36.0	828	0.7080	0.8776
0.0138	37.0	851	0.7070	0.8797
0.0145	38.0	874	0.7088	0.8776
0.0154	39.0	897	0.7177	0.8755
0.0159	40.0	920	0.7217	0.8776
0.0154	41.0	943	0.7165	0.8762
0.013	42.0	966	0.7222	0.8783
0.0142	43.0	989	0.7190	0.8776
0.0143	44.0	1012	0.7207	0.8762
0.0123	45.0	1035	0.7256	0.8762
0.013	46.0	1058	0.7291	0.8790
0.0135	47.0	1081	0.7276	0.8797
0.012	48.0	1104	0.7286	0.8797
0.0135	49.0	1127	0.7278	0.8783
0.0139	50.0	1150	0.7275	0.8790

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0