N_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: 1.0856
• 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
0.6668	1.0	597	0.4299	0.8379
0.3864	2.0	1194	0.3746	0.8639
0.2774	3.0	1791	0.4084	0.8698
0.1897	4.0	2388	0.4874	0.8693
0.14	5.0	2985	0.7174	0.8442
0.0629	6.0	3582	0.6829	0.8740
0.0623	7.0	4179	0.7725	0.8790
0.0403	8.0	4776	0.7908	0.8752
0.0384	9.0	5373	0.8619	0.8693
0.0304	10.0	5970	0.8196	0.8765
0.0183	11.0	6567	0.9165	0.8731
0.0111	12.0	7164	0.9440	0.8756
0.0153	13.0	7761	0.9888	0.8689
0.0129	14.0	8358	0.9551	0.8760
0.0098	15.0	8955	1.0042	0.8731
0.0059	16.0	9552	1.0226	0.8710
0.0076	17.0	10149	1.0608	0.8744
0.0031	18.0	10746	1.0613	0.8786
0.0034	19.0	11343	1.0824	0.8786
0.0021	20.0	11940	1.0856	0.8777

Framework versions

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