Edit model card

hushem_1x_deit_tiny_rms_001_fold3

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: 1.1536
  • Accuracy: 0.4186

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: 0.001
  • 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 3.8148 0.2558
4.0682 2.0 12 1.5106 0.2558
4.0682 3.0 18 2.1015 0.2558
1.8697 4.0 24 2.1521 0.2558
1.6204 5.0 30 2.0540 0.2558
1.6204 6.0 36 1.4487 0.2558
1.5557 7.0 42 1.5322 0.2326
1.5557 8.0 48 1.6480 0.2558
1.5276 9.0 54 1.5085 0.2558
1.4446 10.0 60 1.3921 0.2558
1.4446 11.0 66 1.5703 0.2558
1.4728 12.0 72 1.3608 0.2791
1.4728 13.0 78 1.4250 0.3488
1.3652 14.0 84 1.4495 0.2558
1.3593 15.0 90 1.4182 0.3023
1.3593 16.0 96 1.5418 0.3023
1.2943 17.0 102 1.4454 0.3256
1.2943 18.0 108 1.5941 0.3721
1.2915 19.0 114 1.4889 0.2558
1.2591 20.0 120 1.3804 0.3488
1.2591 21.0 126 1.8125 0.2558
1.2263 22.0 132 1.4098 0.3023
1.2263 23.0 138 1.4818 0.2558
1.1885 24.0 144 1.4257 0.3721
1.1814 25.0 150 1.4317 0.3023
1.1814 26.0 156 1.3854 0.3488
1.1163 27.0 162 1.9054 0.3256
1.1163 28.0 168 1.3109 0.3488
1.0609 29.0 174 1.3896 0.3488
1.1038 30.0 180 1.3466 0.3256
1.1038 31.0 186 1.3101 0.3256
1.0099 32.0 192 1.2865 0.3721
1.0099 33.0 198 1.2846 0.3721
1.0297 34.0 204 1.2587 0.4186
0.964 35.0 210 1.2832 0.3953
0.964 36.0 216 1.1929 0.3721
0.9335 37.0 222 1.2162 0.3953
0.9335 38.0 228 1.1906 0.4419
0.8668 39.0 234 1.1859 0.4186
0.8296 40.0 240 1.1516 0.4884
0.8296 41.0 246 1.1577 0.4651
0.8332 42.0 252 1.1536 0.4186
0.8332 43.0 258 1.1536 0.4186
0.8289 44.0 264 1.1536 0.4186
0.8217 45.0 270 1.1536 0.4186
0.8217 46.0 276 1.1536 0.4186
0.8205 47.0 282 1.1536 0.4186
0.8205 48.0 288 1.1536 0.4186
0.8548 49.0 294 1.1536 0.4186
0.8042 50.0 300 1.1536 0.4186

Framework versions

  • Transformers 4.35.0
  • Pytorch 2.1.0+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1
Downloads last month
10
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_001_fold3

Finetuned
(363)
this model

Evaluation results