N_roberta_twitterfin_padding20model

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

• Loss: 0.9240
• Accuracy: 0.9079

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
0.5487	1.0	597	0.3240	0.8769
0.304	2.0	1194	0.2679	0.9037
0.2471	3.0	1791	0.3491	0.9012
0.1868	4.0	2388	0.4564	0.8999
0.1591	5.0	2985	0.5265	0.8920
0.0871	6.0	3582	0.5935	0.8945
0.0715	7.0	4179	0.6186	0.9020
0.0591	8.0	4776	0.7599	0.8936
0.0458	9.0	5373	0.7605	0.8932
0.0454	10.0	5970	0.7508	0.9070
0.0298	11.0	6567	0.8410	0.8903
0.0187	12.0	7164	0.8508	0.8987
0.0255	13.0	7761	0.8590	0.9024
0.0152	14.0	8358	0.9083	0.9033
0.0125	15.0	8955	0.8954	0.8974
0.0157	16.0	9552	0.9223	0.8970
0.0042	17.0	10149	0.9276	0.9049
0.0087	18.0	10746	0.9098	0.9079
0.004	19.0	11343	0.9121	0.9075
0.0069	20.0	11940	0.9240	0.9079

Framework versions

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