README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - cifar10
metrics:
  - accuracy
model-index:
  - name: sagemaker-ViT-CIFAR10
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: cifar10
          type: cifar10
          config: plain_text
          split: test[:2000]
          args: plain_text
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.972

sagemaker-ViT-CIFAR10

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the cifar10 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2966
• Accuracy: 0.972

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	313	1.4582	0.9325
1.6494	2.0	626	0.4472	0.9665
1.6494	3.0	939	0.2966	0.972

Framework versions

• Transformers 4.26.0
• Pytorch 1.13.1+cu117
• Datasets 2.9.0
• Tokenizers 0.13.2