roberta-base-clickbait-task1-20-epoch-post_title

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: 2.5980
• Accuracy: 0.705

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
No log	1.0	200	0.7472	0.7
No log	2.0	400	0.7596	0.685
0.7443	3.0	600	0.7719	0.7275
0.7443	4.0	800	0.9436	0.7125
0.31	5.0	1000	1.0868	0.71
0.31	6.0	1200	1.5539	0.715
0.31	7.0	1400	1.9772	0.67
0.0987	8.0	1600	1.9892	0.7
0.0987	9.0	1800	2.1162	0.7
0.0445	10.0	2000	2.1434	0.7225
0.0445	11.0	2200	2.2596	0.7075
0.0445	12.0	2400	2.3839	0.705
0.0181	13.0	2600	2.4400	0.7025
0.0181	14.0	2800	2.4389	0.7075
0.0122	15.0	3000	2.5178	0.71
0.0122	16.0	3200	2.5055	0.7075
0.0122	17.0	3400	2.5805	0.7
0.0074	18.0	3600	2.5856	0.7075
0.0074	19.0	3800	2.5949	0.705
0.0052	20.0	4000	2.5980	0.705

Framework versions

• Transformers 4.44.0.dev0
• Pytorch 2.4.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1