Hugging Face's logo

rshrott
/
ryan03302024

Image Classification
Transformers
TensorBoard
Safetensors
vit
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
ryan03302024 / README.md
rshrott's picture
rshrott
🍻 cheers
e9867c0 verified
preview code
|
raw history blame contribute delete
No virus
9.64 kB
metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - image-classification
  - generated_from_trainer
model-index:
  - name: ryan03302024
    results: []

ryan03302024

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the properties dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2044
  • Ordinal Mae: 0.4324
  • Ordinal Accuracy: 0.6648
  • Na Accuracy: 0.8333

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: 16
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Ordinal Mae Ordinal Accuracy Na Accuracy
0.3617 0.04 100 0.3133 0.8466 0.4174 0.8763
0.2891 0.07 200 0.3022 0.7739 0.5045 0.6993
0.3163 0.11 300 0.2615 0.6167 0.5597 0.7904
0.2781 0.14 400 0.2598 0.5433 0.5936 0.8299
0.2731 0.18 500 0.2613 0.5651 0.5566 0.8454
0.2926 0.22 600 0.2734 0.5305 0.5494 0.9089
0.2686 0.25 700 0.2362 0.4853 0.6250 0.7887
0.2715 0.29 800 0.2454 0.4914 0.6255 0.7715
0.2459 0.32 900 0.2452 0.4763 0.6072 0.7801
0.2033 0.36 1000 0.2365 0.4967 0.6106 0.7663
0.2234 0.4 1100 0.2299 0.4947 0.6180 0.8677
0.2035 0.43 1200 0.2314 0.4744 0.6309 0.7835
0.2277 0.47 1300 0.2389 0.4649 0.6435 0.7302
0.2535 0.5 1400 0.2259 0.4509 0.6021 0.8247
0.2209 0.54 1500 0.2369 0.4507 0.6363 0.7577
0.2007 0.58 1600 0.2161 0.4272 0.6540 0.8316
0.2013 0.61 1700 0.2433 0.4326 0.6129 0.7732
0.1999 0.65 1800 0.2227 0.4460 0.6553 0.8247
0.2157 0.69 1900 0.2134 0.4728 0.6363 0.8162
0.2154 0.72 2000 0.2239 0.4734 0.5787 0.8574
0.2169 0.76 2100 0.2394 0.4392 0.6255 0.8849
0.2719 0.79 2200 0.2283 0.4324 0.6229 0.8780
0.2244 0.83 2300 0.2140 0.4483 0.6314 0.8729
0.2072 0.87 2400 0.2198 0.4330 0.6440 0.8213
0.1754 0.9 2500 0.2099 0.4198 0.6712 0.8419
0.1773 0.94 2600 0.2053 0.4105 0.6586 0.8643
0.2378 0.97 2700 0.2044 0.4324 0.6648 0.8333
0.1295 1.01 2800 0.2044 0.4016 0.6843 0.8247
0.1126 1.05 2900 0.2302 0.4025 0.6805 0.7577
0.1262 1.08 3000 0.2205 0.4017 0.6517 0.8093
0.1104 1.12 3100 0.2117 0.3931 0.6779 0.8454
0.1657 1.15 3200 0.2174 0.3890 0.6666 0.8591
0.1186 1.19 3300 0.2299 0.4013 0.6622 0.8058
0.1304 1.23 3400 0.2176 0.3801 0.6902 0.8110
0.1081 1.26 3500 0.2330 0.3867 0.6643 0.8316
0.1281 1.3 3600 0.2320 0.3954 0.6902 0.7680
0.1192 1.33 3700 0.2312 0.4109 0.6769 0.7990
0.1029 1.37 3800 0.2195 0.3870 0.6820 0.8024
0.1159 1.41 3900 0.2200 0.3860 0.6812 0.7904
0.1159 1.44 4000 0.2159 0.3712 0.6982 0.7990
0.107 1.48 4100 0.2262 0.3757 0.6905 0.8213
0.1262 1.51 4200 0.2291 0.3841 0.6835 0.8247
0.1437 1.55 4300 0.2311 0.3751 0.6923 0.8007
0.0916 1.59 4400 0.2343 0.3743 0.6792 0.8660
0.1266 1.62 4500 0.2251 0.3724 0.6861 0.8505
0.1185 1.66 4600 0.2242 0.3666 0.6902 0.8265
0.1037 1.69 4700 0.2219 0.3700 0.6846 0.8522
0.1264 1.73 4800 0.2211 0.3677 0.6892 0.8351
0.1404 1.77 4900 0.2206 0.3718 0.6946 0.7938
0.1238 1.8 5000 0.2098 0.3723 0.6948 0.8265
0.0868 1.84 5100 0.2089 0.3574 0.7025 0.8144
0.0828 1.88 5200 0.2204 0.3680 0.7031 0.7818
0.0986 1.91 5300 0.2126 0.3543 0.6982 0.8127
0.0869 1.95 5400 0.2247 0.3532 0.7108 0.8076
0.1006 1.98 5500 0.2268 0.3637 0.7028 0.8162
0.0639 2.02 5600 0.2252 0.3479 0.7069 0.8110
0.0569 2.06 5700 0.2315 0.3399 0.7167 0.8076
0.0626 2.09 5800 0.2304 0.3481 0.7028 0.8127
0.0502 2.13 5900 0.2381 0.3624 0.6954 0.8093
0.0541 2.16 6000 0.2298 0.3405 0.7159 0.8110
0.0671 2.2 6100 0.2432 0.3529 0.7031 0.7990
0.0672 2.24 6200 0.2431 0.3361 0.7195 0.7715
0.0446 2.27 6300 0.2447 0.3401 0.7141 0.7938
0.0424 2.31 6400 0.2426 0.3485 0.7018 0.8162
0.0386 2.34 6500 0.2488 0.3387 0.7123 0.8127
0.0736 2.38 6600 0.2454 0.3382 0.7054 0.8316
0.0421 2.42 6700 0.2513 0.3394 0.7120 0.8316
0.0607 2.45 6800 0.2546 0.3370 0.7092 0.8265
0.0517 2.49 6900 0.2594 0.3376 0.7082 0.8299
0.062 2.52 7000 0.2533 0.3369 0.7105 0.8110
0.0664 2.56 7100 0.2534 0.3329 0.7185 0.8024
0.0389 2.6 7200 0.2470 0.3288 0.7259 0.8093
0.0671 2.63 7300 0.2516 0.3294 0.7159 0.8041
0.0416 2.67 7400 0.2507 0.3307 0.7133 0.8058
0.0541 2.7 7500 0.2529 0.3355 0.7110 0.8058
0.0374 2.74 7600 0.2530 0.3315 0.7149 0.8110
0.04 2.78 7700 0.2520 0.3290 0.7167 0.8076
0.0507 2.81 7800 0.2555 0.3297 0.7105 0.8127
0.0379 2.85 7900 0.2531 0.3274 0.7162 0.8127
0.0736 2.88 8000 0.2526 0.3279 0.7164 0.8196
0.0589 2.92 8100 0.2522 0.3267 0.7144 0.8162
0.0449 2.96 8200 0.2521 0.3272 0.7149 0.8162
0.0498 2.99 8300 0.2520 0.3265 0.7167 0.8144

Framework versions

  • Transformers 4.39.1
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2