convnextv2-tiny-22k-224-finetuned-piid

This model is a fine-tuned version of facebook/convnextv2-tiny-22k-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.6118
• Accuracy: 0.7854

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2083	0.98	20	1.0137	0.6027
0.6826	2.0	41	0.6901	0.6895
0.5161	2.98	61	0.6377	0.7078
0.4475	4.0	82	0.5423	0.7215
0.4325	4.98	102	0.5165	0.7671
0.3433	6.0	123	0.5916	0.7763
0.2677	6.98	143	0.5866	0.7534
0.2498	8.0	164	0.5146	0.7900
0.2387	8.98	184	0.5631	0.7580
0.2132	10.0	205	0.5320	0.7991
0.2178	10.98	225	0.5833	0.7854
0.1474	12.0	246	0.5902	0.7900
0.1627	12.98	266	0.6142	0.7808
0.1651	14.0	287	0.6063	0.7808
0.158	14.98	307	0.6130	0.7808
0.126	16.0	328	0.6647	0.7671
0.0821	16.98	348	0.5972	0.7808
0.1062	18.0	369	0.5975	0.7945
0.1031	18.98	389	0.6129	0.7808
0.1268	19.51	400	0.6118	0.7854

Framework versions

• Transformers 4.33.3
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3