Edit model card

convnext-nano-15ep

This model is a fine-tuned version of facebook/convnextv2-nano-22k-384 on the vuongnhathien/30VNFoods dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4761
  • Accuracy: 0.9081

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: cosine
  • num_epochs: 15

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.5814 1.0 275 0.5148 0.8477
0.2972 2.0 550 0.4967 0.8557
0.1871 3.0 825 0.4887 0.8716
0.1205 4.0 1100 0.5173 0.8688
0.0732 5.0 1375 0.4979 0.8815
0.0443 6.0 1650 0.5483 0.8815
0.0392 7.0 1925 0.5512 0.8835
0.018 8.0 2200 0.5102 0.8946
0.0043 9.0 2475 0.5423 0.8954
0.0087 10.0 2750 0.4903 0.9105
0.0035 11.0 3025 0.4855 0.9082
0.0022 12.0 3300 0.4874 0.9074
0.0019 13.0 3575 0.4858 0.9082
0.0018 14.0 3850 0.4857 0.9082
0.0018 15.0 4125 0.4859 0.9082

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
15M params
Tensor type
F32
·

Finetuned from

Evaluation results