N_bert_twitterfin_padding50model

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: 1.0004
• Accuracy: 0.8874

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
0.6211	1.0	597	0.3962	0.8492
0.3341	2.0	1194	0.3131	0.8911
0.2233	3.0	1791	0.4254	0.8874
0.1535	4.0	2388	0.6356	0.8819
0.1104	5.0	2985	0.6353	0.8886
0.0362	6.0	3582	0.7047	0.8886
0.0337	7.0	4179	0.7146	0.8865
0.02	8.0	4776	0.7171	0.8869
0.0271	9.0	5373	0.7534	0.8907
0.0173	10.0	5970	0.8021	0.8949
0.0148	11.0	6567	0.8200	0.8894
0.0073	12.0	7164	0.9640	0.8823
0.0082	13.0	7761	0.9143	0.8823
0.0093	14.0	8358	0.9854	0.8827
0.0058	15.0	8955	0.9301	0.8911
0.0036	16.0	9552	0.9559	0.8844
0.003	17.0	10149	0.9667	0.8915
0.0019	18.0	10746	0.9877	0.8915
0.0023	19.0	11343	0.9900	0.8878
0.0027	20.0	11940	1.0004	0.8874

Framework versions

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