N_bert_twitterfin_padding100model

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: 0.9657
• Accuracy: 0.8915

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
0.6297	1.0	597	0.3593	0.8723
0.3274	2.0	1194	0.3075	0.8865
0.2191	3.0	1791	0.4139	0.8819
0.1377	4.0	2388	0.5998	0.8731
0.1034	5.0	2985	0.6520	0.8823
0.0389	6.0	3582	0.6765	0.8844
0.0351	7.0	4179	0.7896	0.8790
0.0227	8.0	4776	0.7827	0.8865
0.026	9.0	5373	0.7999	0.8844
0.022	10.0	5970	0.8195	0.8890
0.0073	11.0	6567	0.8964	0.8773
0.0104	12.0	7164	0.8724	0.8865
0.0124	13.0	7761	0.8707	0.8915
0.0071	14.0	8358	0.9058	0.8928
0.0049	15.0	8955	0.9455	0.8857
0.0027	16.0	9552	0.9714	0.8844
0.0017	17.0	10149	0.9661	0.8886
0.0017	18.0	10746	0.9660	0.8928
0.0012	19.0	11343	0.9602	0.8911
0.0006	20.0	11940	0.9657	0.8915

Framework versions

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