hushem_1x_deit_tiny_sgd_lr00001_fold5

This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

  • Loss: 1.6421
  • Accuracy: 0.1220

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: 1e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 50

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 1.0 6 1.6490 0.1220
1.6062 2.0 12 1.6487 0.1220
1.6062 3.0 18 1.6483 0.1220
1.6229 4.0 24 1.6480 0.1220
1.5995 5.0 30 1.6477 0.1220
1.5995 6.0 36 1.6474 0.1220
1.5906 7.0 42 1.6470 0.1220
1.5906 8.0 48 1.6468 0.1220
1.609 9.0 54 1.6465 0.1220
1.6018 10.0 60 1.6462 0.1220
1.6018 11.0 66 1.6459 0.1220
1.5944 12.0 72 1.6457 0.1220
1.5944 13.0 78 1.6454 0.1220
1.6013 14.0 84 1.6452 0.1220
1.5987 15.0 90 1.6449 0.1220
1.5987 16.0 96 1.6447 0.1220
1.5899 17.0 102 1.6445 0.1220
1.5899 18.0 108 1.6443 0.1220
1.626 19.0 114 1.6441 0.1220
1.5972 20.0 120 1.6439 0.1220
1.5972 21.0 126 1.6437 0.1220
1.5649 22.0 132 1.6436 0.1220
1.5649 23.0 138 1.6434 0.1220
1.6699 24.0 144 1.6433 0.1220
1.5696 25.0 150 1.6431 0.1220
1.5696 26.0 156 1.6430 0.1220
1.5743 27.0 162 1.6429 0.1220
1.5743 28.0 168 1.6427 0.1220
1.6236 29.0 174 1.6426 0.1220
1.5936 30.0 180 1.6426 0.1220
1.5936 31.0 186 1.6425 0.1220
1.5875 32.0 192 1.6424 0.1220
1.5875 33.0 198 1.6423 0.1220
1.6171 34.0 204 1.6423 0.1220
1.5897 35.0 210 1.6422 0.1220
1.5897 36.0 216 1.6422 0.1220
1.5725 37.0 222 1.6421 0.1220
1.5725 38.0 228 1.6421 0.1220
1.6227 39.0 234 1.6421 0.1220
1.5924 40.0 240 1.6421 0.1220
1.5924 41.0 246 1.6421 0.1220
1.5811 42.0 252 1.6421 0.1220
1.5811 43.0 258 1.6421 0.1220
1.6072 44.0 264 1.6421 0.1220
1.5938 45.0 270 1.6421 0.1220
1.5938 46.0 276 1.6421 0.1220
1.6243 47.0 282 1.6421 0.1220
1.6243 48.0 288 1.6421 0.1220
1.5633 49.0 294 1.6421 0.1220
1.6091 50.0 300 1.6421 0.1220

Framework versions

  • Transformers 4.35.0
  • Pytorch 2.1.0+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1
Downloads last month
20
Safetensors
Model size
5.53M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for hkivancoral/hushem_1x_deit_tiny_sgd_lr00001_fold5

Finetuned
(363)
this model

Evaluation results