Edit model card

vit-base-patch16-224-food101

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

  • Loss: 0.3856
  • Accuracy: 0.8971

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Script

"cmd_list": [
        "python",
        "run_image_classification.py",
        "--model_name_or_path",
        "eslamxm/vit-base-food101",
        "--dataset_name",
        "food101",
        "--output_dir",
        "<output_dir>",
        "--overwrite_output_dir",
        "--remove_unused_columns",
        "False",
        "--do_train",
        "--do_eval",
        "--optim",
        "adamw_torch",
        "--learning_rate",
        "6e-05",
        "--num_train_epochs",
        "3",
        "--dataloader_num_workers",
        "10",
        "--per_device_train_batch_size",
        "64",
        "--gradient_accumulation_steps",
        "2",
        "--per_device_eval_batch_size",
        "128",
        "--logging_strategy",
        "steps",
        "--logging_steps",
        "10",
        "--evaluation_strategy",
        "steps",
        "--eval_steps",
        "500",
        "--save_steps",
        "500",
        "--evaluation_strategy",
        "epoch",
        "--save_strategy",
        "epoch",
        "--load_best_model_at_end",
        "False",
        "--save_total_limit",
        "1",
        "--seed",
        "42",
        "--fp16"
    ]

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 6e-05
  • train_batch_size: 64
  • eval_batch_size: 128
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.3687 1.0 592 0.4044 0.8889
0.3422 2.0 1184 0.3911 0.8953
0.3808 3.0 1776 0.3856 0.8971

Framework versions

  • Transformers 4.27.4
  • Pytorch 1.13.1
  • Datasets 2.11.0
  • Tokenizers 0.13.3
Downloads last month
21

Dataset used to train yujiepan/vit-base-patch16-224-food101

Evaluation results