Edit model card

my_awesome_food_model

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the food101 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8834
  • Accuracy: 0.9

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: 3e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 64
  • 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
3.6073 0.99 62 3.3725 0.818
2.2956 2.0 125 2.1579 0.854
1.7042 2.99 187 1.6201 0.887
1.3278 4.0 250 1.3513 0.89
1.1314 4.99 312 1.1549 0.908
1.007 6.0 375 1.0737 0.889
0.905 6.99 437 0.9600 0.906
0.8227 8.0 500 0.9113 0.912
0.7948 8.99 562 0.8908 0.909
0.7598 9.92 620 0.8834 0.9

Framework versions

  • Transformers 4.33.3
  • Pytorch 2.0.1+cu117
  • Datasets 2.14.5
  • Tokenizers 0.13.3
Downloads last month
6
Inference API
Drag image file here or click to browse from your device
This model can be loaded on Inference API (serverless).

Finetuned from

Dataset used to train gchabcou/my_awesome_food_model

Evaluation results