taskA-hallisky-sarcasm-classifier-gpt4-data

This model is a fine-tuned version of hallisky/sarcasm-classifier-gpt4-data on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.9103
• Accuracy: 0.7722
• Precision: 0.5525
• Recall: 0.4697
• F1: 0.5078

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: 4e-06
• 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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.5499	0.7752	500	0.6671	0.7599	0.5307	0.3487	0.4209
0.4267	1.5504	1000	0.6445	0.7664	0.5441	0.4092	0.4671
0.3895	2.3256	1500	0.7337	0.7851	0.6522	0.3026	0.4134
0.3492	3.1008	2000	0.6803	0.7823	0.5830	0.4553	0.5113
0.3251	3.8760	2500	0.7877	0.7621	0.5251	0.5130	0.5190
0.308	4.6512	3000	0.9103	0.7722	0.5525	0.4697	0.5078

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 2.19.2
• Tokenizers 0.19.1