N_distilbert_twitterfin_padding80model

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: 1.1025
• Accuracy: 0.8748

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
0.6959	1.0	597	0.5037	0.8057
0.4339	2.0	1194	0.3885	0.8601
0.3194	3.0	1791	0.3828	0.8773
0.2262	4.0	2388	0.4689	0.8760
0.1725	5.0	2985	0.5596	0.8505
0.0862	6.0	3582	0.6319	0.8714
0.0624	7.0	4179	0.7186	0.8677
0.0548	8.0	4776	0.8019	0.8668
0.0433	9.0	5373	0.8155	0.8689
0.0338	10.0	5970	0.9369	0.8698
0.0181	11.0	6567	0.9505	0.8685
0.0111	12.0	7164	0.9379	0.8698
0.0244	13.0	7761	0.9528	0.8781
0.012	14.0	8358	0.9441	0.8781
0.016	15.0	8955	1.0228	0.8740
0.0119	16.0	9552	1.0243	0.8740
0.0028	17.0	10149	1.0488	0.8773
0.0055	18.0	10746	1.1024	0.8714
0.0039	19.0	11343	1.1000	0.8740
0.0034	20.0	11940	1.1025	0.8748

Framework versions

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