convnext-tiny-224-finetuned

This model is a fine-tuned version of facebook/convnext-tiny-224 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9272
• Accuracy: 0.6275
• Precision: 0.6426
• Recall: 0.6275
• F1: 0.6068

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
1.281	0.9846	32	1.2165	0.5428	0.5230	0.5428	0.4989
1.0964	2.0	65	1.0549	0.5823	0.5459	0.5823	0.5427
0.9929	2.9846	97	0.9905	0.6169	0.5755	0.6169	0.5848
0.9804	4.0	130	0.9691	0.6131	0.5734	0.6131	0.5867
0.9389	4.9846	162	0.9539	0.6246	0.5874	0.6246	0.6007
0.9078	6.0	195	0.9536	0.6189	0.5910	0.6189	0.5973
0.8741	6.9846	227	0.9333	0.6333	0.5947	0.6333	0.6098
0.8523	8.0	260	0.9322	0.6323	0.5952	0.6323	0.6122
0.8222	8.9846	292	0.9354	0.6198	0.6361	0.6198	0.5992
0.7975	9.8462	320	0.9272	0.6275	0.6426	0.6275	0.6068

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1