distilbert_twitterfin_padding70model

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: 1.0730
• Accuracy: 0.8765

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
0.6887	1.0	597	0.4586	0.8262
0.4254	2.0	1194	0.3631	0.8647
0.3042	3.0	1791	0.3923	0.8769
0.2115	4.0	2388	0.5038	0.8693
0.1643	5.0	2985	0.5552	0.8798
0.0827	6.0	3582	0.6608	0.8735
0.0656	7.0	4179	0.7668	0.8660
0.0523	8.0	4776	0.7806	0.8685
0.0474	9.0	5373	0.8615	0.8668
0.031	10.0	5970	0.9038	0.8714
0.0232	11.0	6567	0.9269	0.8693
0.0195	12.0	7164	0.9192	0.8723
0.0275	13.0	7761	0.9875	0.8685
0.0171	14.0	8358	1.0308	0.8714
0.0129	15.0	8955	1.0227	0.8744
0.0052	16.0	9552	1.0471	0.8685
0.0076	17.0	10149	1.0448	0.8769
0.0064	18.0	10746	1.0537	0.8769
0.0078	19.0	11343	1.0615	0.8744
0.0034	20.0	11940	1.0730	0.8765

Framework versions

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