---
license: apache-2.0
base_model: google/vit-huge-patch14-224-in21k
tags:
- image-classification
- vision
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: vit-huge-patch14-224-in21k-finetuned-galaxy10-decals
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# vit-huge-patch14-224-in21k-finetuned-galaxy10-decals

This model is a fine-tuned version of [google/vit-huge-patch14-224-in21k](https://huggingface.co/google/vit-huge-patch14-224-in21k) on the matthieulel/galaxy10_decals dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4927
- Accuracy: 0.8523
- Precision: 0.8538
- Recall: 0.8523
- F1: 0.8489

## 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: 0.0001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 30

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
| 1.7563        | 0.99  | 62   | 1.6011          | 0.5096   | 0.4694    | 0.5096 | 0.4415 |
| 1.0516        | 2.0   | 125  | 0.9115          | 0.7661   | 0.7679    | 0.7661 | 0.7525 |
| 0.8551        | 2.99  | 187  | 0.7590          | 0.7706   | 0.7860    | 0.7706 | 0.7653 |
| 0.6701        | 4.0   | 250  | 0.6253          | 0.8095   | 0.8013    | 0.8095 | 0.7985 |
| 0.6112        | 4.99  | 312  | 0.6058          | 0.8095   | 0.8120    | 0.8095 | 0.8083 |
| 0.6109        | 6.0   | 375  | 0.5428          | 0.8292   | 0.8353    | 0.8292 | 0.8196 |
| 0.5643        | 6.99  | 437  | 0.5230          | 0.8343   | 0.8350    | 0.8343 | 0.8332 |
| 0.5204        | 8.0   | 500  | 0.5010          | 0.8365   | 0.8391    | 0.8365 | 0.8344 |
| 0.4918        | 8.99  | 562  | 0.5000          | 0.8365   | 0.8419    | 0.8365 | 0.8348 |
| 0.4673        | 10.0  | 625  | 0.4949          | 0.8410   | 0.8394    | 0.8410 | 0.8371 |
| 0.4569        | 10.99 | 687  | 0.4803          | 0.8467   | 0.8451    | 0.8467 | 0.8446 |
| 0.4164        | 12.0  | 750  | 0.5012          | 0.8326   | 0.8314    | 0.8326 | 0.8295 |
| 0.424         | 12.99 | 812  | 0.4940          | 0.8410   | 0.8454    | 0.8410 | 0.8382 |
| 0.4045        | 14.0  | 875  | 0.4927          | 0.8523   | 0.8538    | 0.8523 | 0.8489 |
| 0.3651        | 14.99 | 937  | 0.4809          | 0.8416   | 0.8396    | 0.8416 | 0.8403 |
| 0.3512        | 16.0  | 1000 | 0.4955          | 0.8331   | 0.8306    | 0.8331 | 0.8307 |
| 0.2922        | 16.99 | 1062 | 0.5103          | 0.8399   | 0.8357    | 0.8399 | 0.8359 |
| 0.3212        | 18.0  | 1125 | 0.5197          | 0.8439   | 0.8408    | 0.8439 | 0.8412 |
| 0.3171        | 18.99 | 1187 | 0.5253          | 0.8348   | 0.8335    | 0.8348 | 0.8335 |
| 0.2896        | 20.0  | 1250 | 0.5303          | 0.8467   | 0.8456    | 0.8467 | 0.8438 |
| 0.271         | 20.99 | 1312 | 0.5571          | 0.8393   | 0.8391    | 0.8393 | 0.8366 |
| 0.2996        | 22.0  | 1375 | 0.5468          | 0.8422   | 0.8411    | 0.8422 | 0.8404 |
| 0.2663        | 22.99 | 1437 | 0.5620          | 0.8405   | 0.8393    | 0.8405 | 0.8393 |
| 0.2513        | 24.0  | 1500 | 0.5338          | 0.8467   | 0.8448    | 0.8467 | 0.8450 |
| 0.2453        | 24.99 | 1562 | 0.5562          | 0.8484   | 0.8452    | 0.8484 | 0.8446 |
| 0.2237        | 26.0  | 1625 | 0.5619          | 0.8467   | 0.8450    | 0.8467 | 0.8442 |
| 0.2296        | 26.99 | 1687 | 0.5751          | 0.8484   | 0.8496    | 0.8484 | 0.8464 |
| 0.2479        | 28.0  | 1750 | 0.5782          | 0.8461   | 0.8441    | 0.8461 | 0.8431 |
| 0.2207        | 28.99 | 1812 | 0.5746          | 0.8410   | 0.8381    | 0.8410 | 0.8378 |
| 0.2125        | 29.76 | 1860 | 0.5754          | 0.8416   | 0.8393    | 0.8416 | 0.8383 |


### Framework versions

- Transformers 4.37.2
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.15.1