N_bert_twitterfin_padding60model

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: 1.0511
• Accuracy: 0.8911

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
0.5957	1.0	597	0.3652	0.8643
0.322	2.0	1194	0.3316	0.8794
0.2127	3.0	1791	0.4469	0.8802
0.1327	4.0	2388	0.5983	0.8798
0.1008	5.0	2985	0.6930	0.8815
0.0396	6.0	3582	0.7063	0.8827
0.0299	7.0	4179	0.8153	0.8827
0.0214	8.0	4776	0.8951	0.8794
0.023	9.0	5373	0.8829	0.8886
0.0221	10.0	5970	0.8879	0.8874
0.0129	11.0	6567	0.9308	0.8823
0.0079	12.0	7164	0.9553	0.8874
0.012	13.0	7761	0.9391	0.8907
0.0061	14.0	8358	1.0109	0.8894
0.0034	15.0	8955	1.0525	0.8811
0.002	16.0	9552	1.0680	0.8874
0.0023	17.0	10149	1.0690	0.8874
0.0024	18.0	10746	1.0537	0.8874
0.0036	19.0	11343	1.0434	0.8899
0.0024	20.0	11940	1.0511	0.8911

Framework versions

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