roberta-best-finetuned-hopes-fears

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

• Loss: 0.3593
• Accuracy: 0.9434
• F1 Weighted: 0.9453
• Precision Fears: 0.7053
• Recall Fears: 0.8171
• F1 Fears: 0.7571
• Precision Hopes: 0.7458
• Recall Hopes: 0.88
• F1 Hopes: 0.8073
• Precision Neither: 0.9795
• Recall Neither: 0.9579
• F1 Neither: 0.9685

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: 1e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 600
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Weighted	Precision Fears	Recall Fears	F1 Fears	Precision Hopes	Recall Hopes	F1 Hopes	Precision Neither	Recall Neither	F1 Neither
No log	1.0	214	0.7739	0.8930	0.8651	0.4776	0.2602	0.3368	0.0	0.0	0.0	0.9129	0.9876	0.9488
0.8895	2.0	428	0.2800	0.8960	0.9087	0.4736	0.9106	0.6231	0.7417	0.89	0.8091	0.9893	0.8949	0.9397
0.2905	3.0	642	0.3252	0.9492	0.9496	0.7879	0.7398	0.7631	0.7143	0.95	0.8155	0.9759	0.9691	0.9725
0.2905	4.0	856	0.2671	0.9281	0.9340	0.5813	0.8862	0.7021	0.8018	0.89	0.8436	0.9869	0.9335	0.9595
0.1741	5.0	1070	0.3593	0.9434	0.9453	0.7053	0.8171	0.7571	0.7458	0.88	0.8073	0.9795	0.9579	0.9685

Framework versions

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