N_distilbert_twitterfin_padding20model

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

• Loss: 1.1313
• Accuracy: 0.8777

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
0.6575	1.0	597	0.4498	0.8233
0.3836	2.0	1194	0.3575	0.8731
0.2801	3.0	1791	0.3797	0.8790
0.1862	4.0	2388	0.5389	0.8727
0.1449	5.0	2985	0.6156	0.8735
0.0625	6.0	3582	0.6428	0.8668
0.0556	7.0	4179	0.7536	0.8698
0.0483	8.0	4776	0.7399	0.8802
0.0389	9.0	5373	0.9035	0.8639
0.0277	10.0	5970	0.9541	0.8685
0.0248	11.0	6567	0.9428	0.8752
0.0135	12.0	7164	0.9988	0.8723
0.0164	13.0	7761	1.0564	0.8735
0.0104	14.0	8358	1.1117	0.8698
0.0081	15.0	8955	1.1022	0.8706
0.0102	16.0	9552	1.1030	0.8706
0.0051	17.0	10149	1.0971	0.8760
0.0046	18.0	10746	1.1281	0.8769
0.0021	19.0	11343	1.1408	0.8769
0.0019	20.0	11940	1.1313	0.8777

Framework versions

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