---
library_name: transformers
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: vit-base-patch16-224-in21k-finetuned-inaturalist
  results:
  - task:
      name: Image Classification
      type: image-classification
    dataset:
      name: imagefolder
      type: imagefolder
      config: default
      split: validation
      args: default
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.9483333333333334
---

<!-- 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-base-patch16-224-in21k-finetuned-inaturalist

This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co./google/vit-base-patch16-224-in21k) on the imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0056
- Accuracy: 0.9483

## 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: 128
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 512
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 100

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| No log        | 0.8   | 2    | 3.2007          | 0.0333   |
| No log        | 2.0   | 5    | 3.1889          | 0.0444   |
| No log        | 2.8   | 7    | 3.1747          | 0.0639   |
| 3.1888        | 4.0   | 10   | 3.1442          | 0.1097   |
| 3.1888        | 4.8   | 12   | 3.1183          | 0.1458   |
| 3.1888        | 6.0   | 15   | 3.0710          | 0.2194   |
| 3.1888        | 6.8   | 17   | 3.0331          | 0.3042   |
| 3.0673        | 8.0   | 20   | 2.9627          | 0.4389   |
| 3.0673        | 8.8   | 22   | 2.9109          | 0.4944   |
| 3.0673        | 10.0  | 25   | 2.8360          | 0.5764   |
| 3.0673        | 10.8  | 27   | 2.7809          | 0.6056   |
| 2.8151        | 12.0  | 30   | 2.6958          | 0.6542   |
| 2.8151        | 12.8  | 32   | 2.6401          | 0.6764   |
| 2.8151        | 14.0  | 35   | 2.5583          | 0.6944   |
| 2.8151        | 14.8  | 37   | 2.5034          | 0.7083   |
| 2.5143        | 16.0  | 40   | 2.4202          | 0.7347   |
| 2.5143        | 16.8  | 42   | 2.3662          | 0.7375   |
| 2.5143        | 18.0  | 45   | 2.2884          | 0.7444   |
| 2.5143        | 18.8  | 47   | 2.2374          | 0.7569   |
| 2.2236        | 20.0  | 50   | 2.1632          | 0.7778   |
| 2.2236        | 20.8  | 52   | 2.1175          | 0.7833   |
| 2.2236        | 22.0  | 55   | 2.0528          | 0.7931   |
| 2.2236        | 22.8  | 57   | 2.0099          | 0.7958   |
| 1.9677        | 24.0  | 60   | 1.9488          | 0.8014   |
| 1.9677        | 24.8  | 62   | 1.9113          | 0.8097   |
| 1.9677        | 26.0  | 65   | 1.8582          | 0.8139   |
| 1.9677        | 26.8  | 67   | 1.8242          | 0.8139   |
| 1.7467        | 28.0  | 70   | 1.7740          | 0.8111   |
| 1.7467        | 28.8  | 72   | 1.7458          | 0.8056   |
| 1.7467        | 30.0  | 75   | 1.7013          | 0.8181   |
| 1.7467        | 30.8  | 77   | 1.6714          | 0.8194   |
| 1.5765        | 32.0  | 80   | 1.6316          | 0.8264   |
| 1.5765        | 32.8  | 82   | 1.6083          | 0.8236   |
| 1.5765        | 34.0  | 85   | 1.5738          | 0.8292   |
| 1.5765        | 34.8  | 87   | 1.5531          | 0.8347   |
| 1.4431        | 36.0  | 90   | 1.5228          | 0.8431   |
| 1.4431        | 36.8  | 92   | 1.5046          | 0.8444   |
| 1.4431        | 38.0  | 95   | 1.4780          | 0.8472   |
| 1.4431        | 38.8  | 97   | 1.4608          | 0.8458   |
| 1.3049        | 40.0  | 100  | 1.4357          | 0.8458   |
| 1.3049        | 40.8  | 102  | 1.4188          | 0.85     |
| 1.3049        | 42.0  | 105  | 1.3949          | 0.8528   |
| 1.3049        | 42.8  | 107  | 1.3808          | 0.8528   |
| 1.2312        | 44.0  | 110  | 1.3636          | 0.8458   |
| 1.2312        | 44.8  | 112  | 1.3513          | 0.8486   |
| 1.2312        | 46.0  | 115  | 1.3329          | 0.8528   |
| 1.2312        | 46.8  | 117  | 1.3193          | 0.8528   |
| 1.1368        | 48.0  | 120  | 1.3025          | 0.8528   |
| 1.1368        | 48.8  | 122  | 1.2945          | 0.8542   |
| 1.1368        | 50.0  | 125  | 1.2820          | 0.8528   |
| 1.1368        | 50.8  | 127  | 1.2705          | 0.8569   |
| 1.0821        | 52.0  | 130  | 1.2616          | 0.8583   |
| 1.0821        | 52.8  | 132  | 1.2545          | 0.8556   |
| 1.0821        | 54.0  | 135  | 1.2423          | 0.8542   |
| 1.0821        | 54.8  | 137  | 1.2332          | 0.8597   |
| 1.0232        | 56.0  | 140  | 1.2210          | 0.8639   |
| 1.0232        | 56.8  | 142  | 1.2161          | 0.8625   |
| 1.0232        | 58.0  | 145  | 1.2094          | 0.8569   |
| 1.0232        | 58.8  | 147  | 1.2057          | 0.8542   |
| 0.9814        | 60.0  | 150  | 1.1973          | 0.85     |
| 0.9814        | 60.8  | 152  | 1.1919          | 0.8486   |
| 0.9814        | 62.0  | 155  | 1.1825          | 0.8625   |
| 0.9814        | 62.8  | 157  | 1.1799          | 0.8597   |
| 0.9415        | 64.0  | 160  | 1.1716          | 0.8597   |
| 0.9415        | 64.8  | 162  | 1.1665          | 0.8625   |
| 0.9415        | 66.0  | 165  | 1.1611          | 0.8639   |
| 0.9415        | 66.8  | 167  | 1.1600          | 0.8625   |
| 0.9135        | 68.0  | 170  | 1.1577          | 0.8639   |
| 0.9135        | 68.8  | 172  | 1.1547          | 0.8639   |
| 0.9135        | 70.0  | 175  | 1.1493          | 0.8639   |
| 0.9135        | 70.8  | 177  | 1.1464          | 0.8611   |
| 0.8946        | 72.0  | 180  | 1.1423          | 0.8556   |
| 0.8946        | 72.8  | 182  | 1.1402          | 0.8611   |
| 0.8946        | 74.0  | 185  | 1.1375          | 0.8583   |
| 0.8946        | 74.8  | 187  | 1.1360          | 0.8597   |
| 0.8866        | 76.0  | 190  | 1.1344          | 0.8625   |
| 0.8866        | 76.8  | 192  | 1.1334          | 0.8639   |
| 0.8866        | 78.0  | 195  | 1.1324          | 0.8639   |
| 0.8866        | 78.8  | 197  | 1.1320          | 0.8639   |
| 0.8798        | 80.0  | 200  | 1.1319          | 0.8639   |


### Framework versions

- Transformers 4.45.2
- Pytorch 2.4.1+cu121
- Datasets 2.21.0
- Tokenizers 0.20.1