Edit model card

SwinV2-Base-30VN-Food

This model is a fine-tuned version of microsoft/swinv2-base-patch4-window12-192-22k on the vuongnhathien/30VNFoods dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4828
  • Accuracy: 0.8629

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: 0.0003
  • train_batch_size: 64
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.8268 1.0 275 0.5937 0.8270
0.5113 2.0 550 0.5267 0.8545
0.331 3.0 825 0.5459 0.8545
0.2273 4.0 1100 0.6090 0.8441
0.1384 5.0 1375 0.6096 0.8736
0.0918 6.0 1650 0.6669 0.8414
0.0616 7.0 1925 0.6487 0.8891
0.0307 8.0 2200 0.6908 0.8787
0.0173 9.0 2475 0.6673 0.8938
0.0109 10.0 2750 0.6488 0.9014

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.1.2
  • Datasets 2.18.0
  • Tokenizers 0.15.2
Downloads last month
0
Safetensors
Model size
86.9M params
Tensor type
F32
·

Finetuned from

Evaluation results