roberta_twitterfin_padding60model

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

• Loss: 0.8326
• Accuracy: 0.9091

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
0.608	1.0	597	0.3054	0.8819
0.3336	2.0	1194	0.2933	0.8941
0.2761	3.0	1791	0.3703	0.8953
0.2026	4.0	2388	0.4144	0.9028
0.1762	5.0	2985	0.5646	0.8765
0.1003	6.0	3582	0.5252	0.9008
0.0839	7.0	4179	0.7081	0.8915
0.0602	8.0	4776	0.7266	0.8874
0.0565	9.0	5373	0.6926	0.9003
0.054	10.0	5970	0.7251	0.9075
0.0295	11.0	6567	0.7756	0.9012
0.0273	12.0	7164	0.7419	0.9083
0.0267	13.0	7761	0.7496	0.9066
0.0285	14.0	8358	0.8201	0.8978
0.0163	15.0	8955	0.8301	0.8987
0.0132	16.0	9552	0.7921	0.9079
0.0111	17.0	10149	0.7980	0.9104
0.004	18.0	10746	0.8342	0.9095
0.0041	19.0	11343	0.8707	0.9054
0.0074	20.0	11940	0.8326	0.9091

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1
• Datasets 2.12.0
• Tokenizers 0.13.3