distilbert_twitterfin_padding50model

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: 1.1611
• Accuracy: 0.8731

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
0.6399	1.0	597	0.4406	0.8308
0.3771	2.0	1194	0.3532	0.8710
0.259	3.0	1791	0.4253	0.8752
0.1749	4.0	2388	0.5259	0.8677
0.1332	5.0	2985	0.6621	0.8719
0.0625	6.0	3582	0.6391	0.8756
0.0514	7.0	4179	0.8550	0.8656
0.044	8.0	4776	0.8022	0.8744
0.0324	9.0	5373	0.8844	0.8740
0.0267	10.0	5970	0.8932	0.8735
0.0153	11.0	6567	0.9704	0.8748
0.0143	12.0	7164	1.0869	0.8518
0.0143	13.0	7761	1.0678	0.8698
0.0122	14.0	8358	1.0570	0.8698
0.0119	15.0	8955	1.0582	0.8731
0.0058	16.0	9552	1.1039	0.8668
0.0043	17.0	10149	1.1211	0.8744
0.0024	18.0	10746	1.1487	0.8685
0.0034	19.0	11343	1.1457	0.8719
0.0015	20.0	11940	1.1611	0.8731

Framework versions

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