hushem_1x_deit_tiny_rms_lr0001_fold2

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: 1.7300
  • Accuracy: 0.6889

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: 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 3.0363 0.2444
2.6477 2.0 12 1.6954 0.2444
2.6477 3.0 18 1.4980 0.2444
1.482 4.0 24 1.3454 0.3556
1.4166 5.0 30 1.3094 0.4
1.4166 6.0 36 1.6095 0.2444
1.3414 7.0 42 1.9023 0.2444
1.3414 8.0 48 1.3957 0.2222
1.2396 9.0 54 1.1738 0.4
1.2068 10.0 60 1.2312 0.4889
1.2068 11.0 66 1.0903 0.6
0.9263 12.0 72 0.9211 0.5778
0.9263 13.0 78 1.1912 0.4444
0.8539 14.0 84 1.2631 0.5333
0.6672 15.0 90 1.2596 0.5111
0.6672 16.0 96 1.3999 0.4889
0.5299 17.0 102 1.2988 0.5556
0.5299 18.0 108 1.3328 0.5333
0.3853 19.0 114 1.0485 0.6222
0.332 20.0 120 1.1428 0.5778
0.332 21.0 126 1.0486 0.6444
0.1829 22.0 132 1.0866 0.6667
0.1829 23.0 138 1.7727 0.5778
0.111 24.0 144 1.2950 0.6889
0.0444 25.0 150 1.4579 0.7111
0.0444 26.0 156 1.4269 0.6889
0.0017 27.0 162 1.4804 0.6889
0.0017 28.0 168 1.5281 0.6889
0.0007 29.0 174 1.5658 0.6667
0.0005 30.0 180 1.5943 0.6667
0.0005 31.0 186 1.6212 0.6667
0.0004 32.0 192 1.6444 0.6667
0.0004 33.0 198 1.6608 0.6667
0.0003 34.0 204 1.6759 0.6667
0.0003 35.0 210 1.6896 0.6667
0.0003 36.0 216 1.7018 0.6667
0.0003 37.0 222 1.7108 0.6889
0.0003 38.0 228 1.7185 0.6889
0.0003 39.0 234 1.7237 0.6889
0.0002 40.0 240 1.7275 0.6889
0.0002 41.0 246 1.7295 0.6889
0.0003 42.0 252 1.7300 0.6889
0.0003 43.0 258 1.7300 0.6889
0.0002 44.0 264 1.7300 0.6889
0.0002 45.0 270 1.7300 0.6889
0.0002 46.0 276 1.7300 0.6889
0.0002 47.0 282 1.7300 0.6889
0.0002 48.0 288 1.7300 0.6889
0.0002 49.0 294 1.7300 0.6889
0.0002 50.0 300 1.7300 0.6889

Framework versions

  • Transformers 4.35.0
  • Pytorch 2.1.0+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1
Downloads last month
24
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_rms_lr0001_fold2

Finetuned
(363)
this model

Evaluation results