ceva

This model is a fine-tuned version of dianamihalache27/Twroberta-baseB_5epoch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1932
• Accuracy: 0.7714
• F1: 0.2892
• Precision: 0.2435
• Recall: 0.3579
• Precision Sarcastic: 0.3372
• Recall Sarcastic: 0.4833
• F1 Sarcastic: 0.3973

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: 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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	Precision Sarcastic	Recall Sarcastic	F1 Sarcastic
No log	1.0	217	0.1618	0.7893	0.2947	0.2611	0.3395	0.3458	0.4611	0.3952
No log	2.0	434	0.1815	0.7629	0.3104	0.2560	0.4059	0.3299	0.5278	0.4060
0.0559	3.0	651	0.1762	0.8	0.2957	0.2991	0.3173	0.3721	0.4444	0.4051
0.0559	4.0	868	0.1811	0.7636	0.2933	0.2418	0.3727	0.3297	0.5111	0.4009
0.0245	5.0	1085	0.1932	0.7714	0.2892	0.2435	0.3579	0.3372	0.4833	0.3973

Framework versions

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