N_distilbert_twitterfin_padding10model

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: 1.1006
• Accuracy: 0.8769

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
0.6443	1.0	597	0.3970	0.8497
0.3705	2.0	1194	0.3491	0.8727
0.2636	3.0	1791	0.3912	0.8710
0.1666	4.0	2388	0.4878	0.8798
0.1335	5.0	2985	0.6620	0.8714
0.0481	6.0	3582	0.6601	0.8731
0.0421	7.0	4179	0.8063	0.8702
0.0344	8.0	4776	0.8370	0.8702
0.0293	9.0	5373	0.8666	0.8689
0.0288	10.0	5970	0.9554	0.8673
0.0165	11.0	6567	0.9338	0.8693
0.0144	12.0	7164	0.9260	0.8756
0.011	13.0	7761	1.0099	0.8752
0.0117	14.0	8358	0.9851	0.8765
0.0071	15.0	8955	1.0191	0.8748
0.0023	16.0	9552	1.0490	0.8748
0.0053	17.0	10149	1.0738	0.8777
0.0023	18.0	10746	1.0958	0.8769
0.0019	19.0	11343	1.1029	0.8773
0.0014	20.0	11940	1.1006	0.8769

Framework versions

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