metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- f1
model-index:
- name: vit-base-patch16-224-in21k
results:
- task:
name: Image Classification
type: image-classification
dataset:
name: imagefolder
type: imagefolder
config: default
split: train
args: default
metrics:
- name: F1
type: f1
value: 0.960503161050642
vit-base-patch16-224-in21k
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: 0.0377
- F1: 0.9605
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: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.1855 | 0.99 | 53 | 0.1819 | 0.4851 |
| 0.1147 | 1.99 | 107 | 0.1140 | 0.7505 |
| 0.1075 | 3.0 | 161 | 0.0932 | 0.8654 |
| 0.0755 | 4.0 | 215 | 0.0684 | 0.9268 |
| 0.0605 | 4.99 | 268 | 0.0584 | 0.9294 |
| 0.0475 | 5.99 | 322 | 0.0436 | 0.9550 |
| 0.0442 | 7.0 | 376 | 0.0503 | 0.9367 |
| 0.0464 | 8.0 | 430 | 0.0398 | 0.9599 |
| 0.0267 | 8.99 | 483 | 0.0445 | 0.9423 |
| 0.0374 | 9.86 | 530 | 0.0377 | 0.9605 |
Framework versions
- Transformers 4.37.2
- Pytorch 1.12.1+cu102
- Datasets 2.16.1
- Tokenizers 0.15.1