bert_twitterfin_padding0model

This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0341
• Accuracy: 0.8861

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
0.5848	1.0	597	0.3765	0.8601
0.339	2.0	1194	0.3377	0.8807
0.2321	3.0	1791	0.4471	0.8794
0.1389	4.0	2388	0.6145	0.8807
0.102	5.0	2985	0.6765	0.8807
0.0398	6.0	3582	0.7212	0.8832
0.0344	7.0	4179	0.7934	0.8848
0.0252	8.0	4776	0.9231	0.8660
0.0248	9.0	5373	0.8171	0.8807
0.0184	10.0	5970	0.8674	0.8853
0.0154	11.0	6567	0.9038	0.8861
0.0102	12.0	7164	0.9355	0.8911
0.0086	13.0	7761	1.0081	0.8765
0.0062	14.0	8358	0.9375	0.8886
0.007	15.0	8955	0.9815	0.8874
0.0038	16.0	9552	1.0139	0.8827
0.0022	17.0	10149	0.9909	0.8903
0.0019	18.0	10746	1.0112	0.8886
0.0004	19.0	11343	1.0231	0.8878
0.0009	20.0	11940	1.0341	0.8861

Framework versions

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