hushem_1x_deit_tiny_sgd_lr00001_fold1

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: 1.6574
  • Accuracy: 0.2667

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 1.6633 0.2667
1.6088 2.0 12 1.6630 0.2667
1.6088 3.0 18 1.6627 0.2667
1.5763 4.0 24 1.6624 0.2667
1.6076 5.0 30 1.6621 0.2667
1.6076 6.0 36 1.6618 0.2667
1.5951 7.0 42 1.6616 0.2667
1.5951 8.0 48 1.6613 0.2667
1.5898 9.0 54 1.6611 0.2667
1.5905 10.0 60 1.6609 0.2667
1.5905 11.0 66 1.6606 0.2667
1.5785 12.0 72 1.6604 0.2667
1.5785 13.0 78 1.6602 0.2667
1.623 14.0 84 1.6600 0.2667
1.5698 15.0 90 1.6598 0.2667
1.5698 16.0 96 1.6596 0.2667
1.5831 17.0 102 1.6594 0.2667
1.5831 18.0 108 1.6593 0.2667
1.6234 19.0 114 1.6591 0.2667
1.605 20.0 120 1.6589 0.2667
1.605 21.0 126 1.6588 0.2667
1.6023 22.0 132 1.6586 0.2667
1.6023 23.0 138 1.6585 0.2667
1.5903 24.0 144 1.6584 0.2667
1.5877 25.0 150 1.6583 0.2667
1.5877 26.0 156 1.6582 0.2667
1.5697 27.0 162 1.6581 0.2667
1.5697 28.0 168 1.6580 0.2667
1.6252 29.0 174 1.6579 0.2667
1.6032 30.0 180 1.6578 0.2667
1.6032 31.0 186 1.6577 0.2667
1.6035 32.0 192 1.6577 0.2667
1.6035 33.0 198 1.6576 0.2667
1.5747 34.0 204 1.6575 0.2667
1.5966 35.0 210 1.6575 0.2667
1.5966 36.0 216 1.6575 0.2667
1.5685 37.0 222 1.6574 0.2667
1.5685 38.0 228 1.6574 0.2667
1.5973 39.0 234 1.6574 0.2667
1.5951 40.0 240 1.6574 0.2667
1.5951 41.0 246 1.6574 0.2667
1.5959 42.0 252 1.6574 0.2667
1.5959 43.0 258 1.6574 0.2667
1.6121 44.0 264 1.6574 0.2667
1.5823 45.0 270 1.6574 0.2667
1.5823 46.0 276 1.6574 0.2667
1.5911 47.0 282 1.6574 0.2667
1.5911 48.0 288 1.6574 0.2667
1.6171 49.0 294 1.6574 0.2667
1.5945 50.0 300 1.6574 0.2667

Framework versions

  • Transformers 4.35.0
  • Pytorch 2.1.0+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1
Downloads last month
21
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_fold1

Finetuned
(363)
this model

Evaluation results