cvt-21-384-22k-finetuned-LeafBack

This model is a fine-tuned version of microsoft/cvt-21-384-22k on the imagefolder dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0832
  • Accuracy: 0.9740

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: 2e-05
  • train_batch_size: 100
  • eval_batch_size: 50
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 60

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 1.0 7 0.6551 0.6364
No log 2.0 14 0.5215 0.7532
No log 3.0 21 0.4506 0.7922
No log 4.0 28 0.4032 0.8182
No log 5.0 35 0.3416 0.8442
No log 6.0 42 0.2768 0.9091
No log 7.0 49 0.3287 0.8571
No log 8.0 56 0.2419 0.9091
No log 9.0 63 0.2510 0.9091
No log 10.0 70 0.2100 0.8961
No log 11.0 77 0.2691 0.8961
No log 12.0 84 0.2717 0.8961
No log 13.0 91 0.1412 0.9481
No log 14.0 98 0.1541 0.9351
No log 15.0 105 0.2572 0.8701
No log 16.0 112 0.1843 0.8961
No log 17.0 119 0.1376 0.9481
No log 18.0 126 0.1632 0.9351
No log 19.0 133 0.1901 0.9091
No log 20.0 140 0.1495 0.9221
No log 21.0 147 0.1710 0.9221
No log 22.0 154 0.1906 0.9351
No log 23.0 161 0.1192 0.9610
No log 24.0 168 0.2000 0.9091
No log 25.0 175 0.1611 0.9221
No log 26.0 182 0.1074 0.9610
No log 27.0 189 0.1080 0.9481
No log 28.0 196 0.1405 0.9351
No log 29.0 203 0.0890 0.9610
No log 30.0 210 0.0777 0.9740
No log 31.0 217 0.0636 0.9740
No log 32.0 224 0.0709 0.9740
No log 33.0 231 0.0671 0.9740
No log 34.0 238 0.1055 0.9610
No log 35.0 245 0.1366 0.9610
No log 36.0 252 0.1070 0.9610
No log 37.0 259 0.0890 0.9481
No log 38.0 266 0.0876 0.9481
No log 39.0 273 0.0838 0.9481
No log 40.0 280 0.1502 0.9610
No log 41.0 287 0.1241 0.9610
No log 42.0 294 0.1043 0.9740
No log 43.0 301 0.1126 0.9740
No log 44.0 308 0.1283 0.9610
No log 45.0 315 0.1123 0.9610
No log 46.0 322 0.1198 0.9610
No log 47.0 329 0.1434 0.9610
No log 48.0 336 0.1340 0.9610
No log 49.0 343 0.1073 0.9610
No log 50.0 350 0.1112 0.9610
No log 51.0 357 0.1035 0.9610
No log 52.0 364 0.0972 0.9610
No log 53.0 371 0.0960 0.9610
No log 54.0 378 0.0881 0.9610
No log 55.0 385 0.0797 0.9610
No log 56.0 392 0.0773 0.9740
No log 57.0 399 0.0736 0.9740
No log 58.0 406 0.0794 0.9740
No log 59.0 413 0.0841 0.9740
No log 60.0 420 0.0832 0.9740

Framework versions

  • Transformers 4.38.1
  • Pytorch 1.10.0+cu111
  • Datasets 2.17.1
  • Tokenizers 0.15.2
Downloads last month
19
Safetensors
Model size
31.3M 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 b07611031/cvt-21-384-22k-finetuned-LeafBack

Finetuned
(6)
this model

Evaluation results