metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- generated_from_trainer
datasets:
- imagefolder
model-index:
- name: attraction-classifier
results: []
attraction-classifier
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.5429
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: 16
- eval_batch_size: 16
- seed: 69
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.7599 | 0.11 | 20 | 0.7105 |
0.6535 | 0.22 | 40 | 0.6561 |
0.6191 | 0.33 | 60 | 0.6045 |
0.7219 | 0.44 | 80 | 0.5976 |
0.62 | 0.55 | 100 | 0.5781 |
0.5249 | 0.66 | 120 | 0.5743 |
0.5934 | 0.77 | 140 | 0.5684 |
0.4143 | 0.88 | 160 | 0.5522 |
0.4472 | 0.99 | 180 | 0.5645 |
0.3507 | 1.1 | 200 | 0.5784 |
0.4165 | 1.21 | 220 | 0.5796 |
0.2822 | 1.32 | 240 | 0.5867 |
0.4765 | 1.43 | 260 | 0.5446 |
0.4557 | 1.54 | 280 | 0.5739 |
0.4087 | 1.65 | 300 | 0.5292 |
0.4181 | 1.76 | 320 | 0.5182 |
0.6616 | 1.87 | 340 | 0.5328 |
0.48 | 1.98 | 360 | 0.5282 |
0.2159 | 2.09 | 380 | 0.5457 |
0.5038 | 2.2 | 400 | 0.5832 |
0.6434 | 2.31 | 420 | 0.5667 |
0.4566 | 2.42 | 440 | 0.5128 |
0.2972 | 2.53 | 460 | 0.5475 |
0.5094 | 2.64 | 480 | 0.5536 |
0.4807 | 2.75 | 500 | 0.5512 |
0.3666 | 2.86 | 520 | 0.5128 |
0.38 | 2.97 | 540 | 0.5499 |
0.3496 | 3.08 | 560 | 0.5129 |
0.3503 | 3.19 | 580 | 0.5480 |
0.4435 | 3.3 | 600 | 0.5918 |
0.2163 | 3.41 | 620 | 0.5558 |
0.4316 | 3.52 | 640 | 0.5429 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.0.1+cu117
- Datasets 2.15.0
- Tokenizers 0.15.0