fine-tuned-twitter-roberta-base-sentiment-latest

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: 0.3062
• Accuracy: {'accuracy': 0.8868131868131868}
• F1score: {'f1': 0.88247351021472}

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1score
0.461	0.2443	500	0.3381	{'accuracy': 0.8565934065934065}	{'f1': 0.8483856477235431}
0.3702	0.4885	1000	0.3378	{'accuracy': 0.865934065934066}	{'f1': 0.8655309886906097}
0.3574	0.7328	1500	0.2971	{'accuracy': 0.8714285714285714}	{'f1': 0.8709435986031107}
0.3358	0.9770	2000	0.3062	{'accuracy': 0.8868131868131868}	{'f1': 0.88247351021472}

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1