N_bert_twitterfin_padding20model

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: 1.0103
• Accuracy: 0.8920

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: 16
• eval_batch_size: 16
• seed: 42
• 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
0.6026	1.0	597	0.3849	0.8589
0.3307	2.0	1194	0.3351	0.8832
0.2306	3.0	1791	0.4305	0.8865
0.1415	4.0	2388	0.5673	0.8827
0.1018	5.0	2985	0.6632	0.8794
0.0396	6.0	3582	0.7322	0.8819
0.0367	7.0	4179	0.7720	0.8874
0.0253	8.0	4776	0.8155	0.8836
0.0281	9.0	5373	0.8304	0.8853
0.0246	10.0	5970	0.8940	0.8882
0.0091	11.0	6567	1.0241	0.8823
0.0102	12.0	7164	0.9821	0.8874
0.0192	13.0	7761	1.0144	0.8765
0.0064	14.0	8358	1.0386	0.8861
0.0033	15.0	8955	0.9737	0.8907
0.0029	16.0	9552	1.0372	0.8890
0.002	17.0	10149	1.0022	0.8928
0.0016	18.0	10746	1.0081	0.8894
0.0017	19.0	11343	1.0171	0.8915
0.0024	20.0	11940	1.0103	0.8920

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3