novo

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

• Loss: 0.8419
• Accuracy: 0.8926
• F1: 0.6997
• Recall: 0.6785
• Precision: 0.7256

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Recall	Precision
No log	1.0	247	0.3108	0.9007	0.5703	0.5571	0.8927
No log	2.0	494	0.3131	0.8926	0.6777	0.6337	0.7649
0.3289	3.0	741	0.5054	0.8896	0.6971	0.7116	0.6857
0.3289	4.0	988	0.6063	0.8987	0.7195	0.7089	0.7309
0.1268	5.0	1235	0.7100	0.8967	0.6993	0.6836	0.7173
0.1268	6.0	1482	0.7683	0.8977	0.7039	0.6861	0.7257
0.0346	7.0	1729	0.8041	0.8936	0.7012	0.6701	0.7415
0.0346	8.0	1976	0.8185	0.8956	0.6988	0.6815	0.7191
0.0107	9.0	2223	0.8280	0.8926	0.6943	0.6739	0.7185
0.0107	10.0	2470	0.8419	0.8926	0.6997	0.6785	0.7256

Framework versions

• Transformers 4.33.1
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3