N_roberta_twitterfin_padding10model

This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8928
• Accuracy: 0.9041

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
0.5357	1.0	597	0.3076	0.8899
0.2981	2.0	1194	0.2893	0.8920
0.256	3.0	1791	0.3286	0.8924
0.1801	4.0	2388	0.4796	0.8978
0.1503	5.0	2985	0.5024	0.9008
0.0803	6.0	3582	0.5620	0.8974
0.0729	7.0	4179	0.7456	0.8869
0.0554	8.0	4776	0.7191	0.8936
0.056	9.0	5373	0.6453	0.9070
0.0429	10.0	5970	0.7056	0.9028
0.0291	11.0	6567	0.7841	0.8982
0.025	12.0	7164	0.8934	0.8941
0.0176	13.0	7761	0.7528	0.9058
0.0191	14.0	8358	0.8226	0.8987
0.0201	15.0	8955	0.8367	0.9003
0.0135	16.0	9552	0.8616	0.9037
0.0068	17.0	10149	0.8865	0.9016
0.0089	18.0	10746	0.8481	0.9070
0.0054	19.0	11343	0.8897	0.9045
0.0075	20.0	11940	0.8928	0.9041

Framework versions

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