N_bert_twitterfin_padding30model

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: 1.0650
• Accuracy: 0.8815

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
0.6032	1.0	597	0.3796	0.8652
0.3371	2.0	1194	0.3746	0.8760
0.2299	3.0	1791	0.4256	0.8848
0.1487	4.0	2388	0.6041	0.8836
0.1052	5.0	2985	0.6433	0.8769
0.0388	6.0	3582	0.6947	0.8874
0.04	7.0	4179	0.7443	0.8894
0.0331	8.0	4776	0.8264	0.8773
0.0226	9.0	5373	0.8278	0.8815
0.0177	10.0	5970	0.9051	0.8832
0.0141	11.0	6567	0.9610	0.8807
0.0136	12.0	7164	0.9300	0.8848
0.0131	13.0	7761	0.9504	0.8823
0.0075	14.0	8358	1.0103	0.8823
0.0035	15.0	8955	1.0177	0.8844
0.0026	16.0	9552	1.0749	0.8819
0.0056	17.0	10149	1.0699	0.8844
0.002	18.0	10746	1.0380	0.8853
0.0036	19.0	11343	1.0700	0.8811
0.0009	20.0	11940	1.0650	0.8815

Framework versions

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