bert-base-uncased-sst-2-32-13-30

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

• Loss: 0.5572
• Accuracy: 0.75

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: 1.5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 5
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	2	0.6997	0.4375
No log	2.0	4	0.6973	0.4375
No log	3.0	6	0.6912	0.5781
No log	4.0	8	0.6876	0.5
0.6783	5.0	10	0.6843	0.5312
0.6783	6.0	12	0.6800	0.5781
0.6783	7.0	14	0.6738	0.5938
0.6783	8.0	16	0.6662	0.6562
0.6783	9.0	18	0.6573	0.6562
0.5945	10.0	20	0.6496	0.7031
0.5945	11.0	22	0.6427	0.7188
0.5945	12.0	24	0.6343	0.7188
0.5945	13.0	26	0.6270	0.7031
0.5945	14.0	28	0.6218	0.6875
0.4805	15.0	30	0.6166	0.6875
0.4805	16.0	32	0.6110	0.7188
0.4805	17.0	34	0.6046	0.7344
0.4805	18.0	36	0.5972	0.7344
0.4805	19.0	38	0.5895	0.7344
0.3522	20.0	40	0.5823	0.75
0.3522	21.0	42	0.5767	0.7344
0.3522	22.0	44	0.5708	0.7344
0.3522	23.0	46	0.5667	0.7344
0.3522	24.0	48	0.5637	0.7344
0.2697	25.0	50	0.5616	0.7344
0.2697	26.0	52	0.5603	0.7344
0.2697	27.0	54	0.5592	0.7344
0.2697	28.0	56	0.5582	0.75
0.2697	29.0	58	0.5574	0.75
0.2363	30.0	60	0.5572	0.75

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.4.0
• Tokenizers 0.13.3