Augusto777's picture
Model save
9444427 verified
metadata
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: vit-base-patch16-224-dmae-va-U5-100bc
    results: []

vit-base-patch16-224-dmae-va-U5-100bc

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5346
  • Accuracy: 0.85

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: 5e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.05
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 0.9 7 0.5017 0.8667
0.3168 1.94 15 0.5970 0.8
0.2613 2.97 23 0.5442 0.8167
0.222 4.0 31 0.7156 0.7667
0.222 4.9 38 0.5175 0.85
0.1783 5.94 46 0.6035 0.8167
0.168 6.97 54 0.5045 0.85
0.1456 8.0 62 0.4923 0.85
0.1456 8.9 69 0.5346 0.85
0.1236 9.03 70 0.5346 0.85

Framework versions

  • Transformers 4.36.2
  • Pytorch 2.1.2+cu118
  • Datasets 2.16.1
  • Tokenizers 0.15.0