Hugging Face's logo

bryanzhou008
/
vit-base-patch16-224-in21k-finetuned-inaturalist

Image Classification
Transformers
Safetensors
vit
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
vit-base-patch16-224-in21k-finetuned-inaturalist / README.md
bryanzhou008's picture
bryanzhou008
Model save
9f604a4 verified
preview code
|
raw
history blame contribute delete
7.29 kB
metadata
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: 0.8541666666666666

vit-base-patch16-224-in21k-finetuned-inaturalist

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: 0.7703
  • Accuracy: 0.8542

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 0.8421 4 3.1793 0.0347
No log 1.8947 9 3.1647 0.0486
3.1648 2.9474 14 3.1382 0.0944
3.1648 4.0 19 3.0995 0.1556
3.0817 4.8421 23 3.0555 0.2639
3.0817 5.8947 28 2.9849 0.3889
2.9167 6.9474 33 2.8932 0.5139
2.9167 8.0 38 2.7775 0.5972
2.6682 8.8421 42 2.6706 0.6528
2.6682 9.8947 47 2.5233 0.7069
2.3659 10.9474 52 2.3859 0.7375
2.3659 12.0 57 2.2546 0.75
2.079 12.8421 61 2.1531 0.7528
2.079 13.8947 66 2.0372 0.75
1.828 14.9474 71 1.9339 0.7597
1.828 16.0 76 1.8403 0.7694
1.6253 16.8421 80 1.7733 0.7764
1.6253 17.8947 85 1.6914 0.7903
1.4502 18.9474 90 1.6153 0.7875
1.4502 20.0 95 1.5510 0.7986
1.4502 20.8421 99 1.5016 0.8
1.2959 21.8947 104 1.4454 0.8222
1.2959 22.9474 109 1.3912 0.8181
1.1802 24.0 114 1.3390 0.8333
1.1802 24.8421 118 1.2995 0.8333
1.0629 25.8947 123 1.2707 0.8389
1.0629 26.9474 128 1.2335 0.8361
0.9801 28.0 133 1.1975 0.8444
0.9801 28.8421 137 1.1672 0.8389
0.9076 29.8947 142 1.1338 0.8444
0.9076 30.9474 147 1.1137 0.8472
0.8349 32.0 152 1.0855 0.8528
0.8349 32.8421 156 1.0717 0.8542
0.7782 33.8947 161 1.0483 0.8514
0.7782 34.9474 166 1.0352 0.85
0.7208 36.0 171 1.0202 0.8556
0.7208 36.8421 175 0.9994 0.8486
0.6708 37.8947 180 0.9814 0.8556
0.6708 38.9474 185 0.9691 0.8542
0.6303 40.0 190 0.9599 0.8486
0.6303 40.8421 194 0.9422 0.8472
0.6303 41.8947 199 0.9278 0.8486
0.6018 42.9474 204 0.9172 0.8528
0.6018 44.0 209 0.9093 0.8514
0.5622 44.8421 213 0.9030 0.8583
0.5622 45.8947 218 0.8972 0.8625
0.5474 46.9474 223 0.8859 0.8569
0.5474 48.0 228 0.8858 0.8653
0.5254 48.8421 232 0.8779 0.8556
0.5254 49.8947 237 0.8635 0.8569
0.5036 50.9474 242 0.8563 0.8611
0.5036 52.0 247 0.8613 0.8542
0.4855 52.8421 251 0.8546 0.8625
0.4855 53.8947 256 0.8469 0.8597
0.4697 54.9474 261 0.8327 0.8528
0.4697 56.0 266 0.8268 0.8597
0.4482 56.8421 270 0.8188 0.8556
0.4482 57.8947 275 0.8171 0.8653
0.4436 58.9474 280 0.8133 0.8486
0.4436 60.0 285 0.8070 0.8639
0.4436 60.8421 289 0.7986 0.8542
0.4211 61.8947 294 0.7937 0.8597
0.4211 62.9474 299 0.7908 0.8611
0.4228 64.0 304 0.7952 0.8625
0.4228 64.8421 308 0.8010 0.8514
0.4046 65.8947 313 0.7975 0.8472
0.4046 66.9474 318 0.7927 0.8417
0.4048 68.0 323 0.7880 0.8556
0.4048 68.8421 327 0.7860 0.8514
0.3925 69.8947 332 0.7899 0.8403
0.3925 70.9474 337 0.7883 0.8417
0.3936 72.0 342 0.7885 0.8417
0.3936 72.8421 346 0.7874 0.8361
0.3985 73.8947 351 0.7832 0.8417
0.3985 74.9474 356 0.7787 0.8514
0.3849 76.0 361 0.7753 0.8486
0.3849 76.8421 365 0.7746 0.8514
0.3796 77.8947 370 0.7736 0.8542
0.3796 78.9474 375 0.7731 0.8528
0.3717 80.0 380 0.7715 0.8556
0.3717 80.8421 384 0.7709 0.8556
0.3717 81.8947 389 0.7706 0.8569
0.3802 82.9474 394 0.7704 0.8556
0.3802 84.0 399 0.7704 0.8542
0.3782 84.2105 400 0.7703 0.8542

Framework versions

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