N_bert_twitterfin_padding40model

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: 0.9437
• Accuracy: 0.8915

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
0.6129	1.0	597	0.3541	0.8719
0.3267	2.0	1194	0.3151	0.8882
0.2206	3.0	1791	0.3821	0.8874
0.1488	4.0	2388	0.5206	0.8827
0.0968	5.0	2985	0.5811	0.8936
0.0418	6.0	3582	0.7151	0.8827
0.0347	7.0	4179	0.7350	0.8874
0.0216	8.0	4776	0.7844	0.8861
0.03	9.0	5373	0.7782	0.8823
0.0232	10.0	5970	0.7842	0.8894
0.0167	11.0	6567	0.7931	0.8920
0.0106	12.0	7164	0.8861	0.8899
0.0138	13.0	7761	0.8340	0.8915
0.0095	14.0	8358	0.9040	0.8832
0.0061	15.0	8955	0.8930	0.8861
0.0041	16.0	9552	0.9084	0.8894
0.0031	17.0	10149	0.9436	0.8869
0.0027	18.0	10746	0.9364	0.8924
0.0014	19.0	11343	0.9439	0.8907
0.0023	20.0	11940	0.9437	0.8915

Framework versions

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