N_roberta_twitterfin_padding30model

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

• Loss: 0.8882
• Accuracy: 0.9075

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
0.5609	1.0	597	0.3012	0.8890
0.3047	2.0	1194	0.2728	0.9049
0.2576	3.0	1791	0.3331	0.8961
0.1824	4.0	2388	0.4308	0.8995
0.168	5.0	2985	0.5599	0.8957
0.087	6.0	3582	0.5452	0.9012
0.0765	7.0	4179	0.6220	0.9016
0.0673	8.0	4776	0.6395	0.9008
0.0652	9.0	5373	0.7767	0.8915
0.0477	10.0	5970	0.7780	0.8974
0.0307	11.0	6567	0.7124	0.9070
0.026	12.0	7164	0.7456	0.9049
0.0304	13.0	7761	0.8278	0.9037
0.0197	14.0	8358	0.8793	0.9041
0.0103	15.0	8955	0.8116	0.9079
0.01	16.0	9552	0.8631	0.9062
0.0086	17.0	10149	0.8748	0.9058
0.0103	18.0	10746	0.8648	0.9100
0.0057	19.0	11343	0.9243	0.9008
0.0074	20.0	11940	0.8882	0.9075

Framework versions

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