N_distilbert_twitterfin_padding40model

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: 1.1236
• Accuracy: 0.8752

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
0.6494	1.0	597	0.4836	0.8132
0.385	2.0	1194	0.3568	0.8752
0.2703	3.0	1791	0.3817	0.8781
0.1881	4.0	2388	0.5134	0.8802
0.1451	5.0	2985	0.6251	0.8735
0.0661	6.0	3582	0.7005	0.8723
0.0549	7.0	4179	0.7862	0.8731
0.043	8.0	4776	0.8363	0.8677
0.0351	9.0	5373	0.8813	0.8702
0.0232	10.0	5970	1.0102	0.8773
0.0186	11.0	6567	0.9771	0.8702
0.0228	12.0	7164	0.9386	0.8740
0.017	13.0	7761	1.0307	0.8710
0.0127	14.0	8358	1.0443	0.8744
0.0067	15.0	8955	1.0813	0.8681
0.0048	16.0	9552	1.1306	0.8685
0.0042	17.0	10149	1.1365	0.8731
0.0029	18.0	10746	1.1268	0.8744
0.0041	19.0	11343	1.1145	0.8740
0.0017	20.0	11940	1.1236	0.8752

Framework versions

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