N_bert_twitterfin_padding80model

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: 1.0050
• Accuracy: 0.8894

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
0.6072	1.0	597	0.3449	0.8748
0.3281	2.0	1194	0.3080	0.8878
0.2384	3.0	1791	0.3908	0.8894
0.1552	4.0	2388	0.5590	0.8719
0.1206	5.0	2985	0.6288	0.8802
0.0501	6.0	3582	0.6952	0.8874
0.0357	7.0	4179	0.7691	0.8827
0.0314	8.0	4776	0.8138	0.8844
0.0293	9.0	5373	0.8231	0.8886
0.0199	10.0	5970	0.8076	0.8890
0.0191	11.0	6567	0.8359	0.8903
0.0071	12.0	7164	0.8779	0.8857
0.0105	13.0	7761	0.9540	0.8874
0.0068	14.0	8358	0.9292	0.8890
0.0094	15.0	8955	0.9410	0.8903
0.0065	16.0	9552	0.9804	0.8857
0.0066	17.0	10149	0.9936	0.8878
0.0025	18.0	10746	1.0035	0.8915
0.0038	19.0	11343	1.0173	0.8878
0.0024	20.0	11940	1.0050	0.8894

Framework versions

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