convnext-tiny-224-finetuned-biopsy

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

  • Loss: 0.0688
  • Accuracy: 0.9816

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: 5e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 128
  • 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
1.2488 1.0 42 1.1914 0.5360
0.9265 2.0 84 0.8634 0.5544
0.5701 3.0 126 0.4834 0.8543
0.4041 4.0 168 0.2996 0.9213
0.2747 5.0 210 0.2743 0.9146
0.2518 6.0 252 0.1826 0.9497
0.2363 7.0 294 0.1731 0.9497
0.1782 8.0 336 0.1870 0.9363
0.2122 9.0 378 0.1327 0.9615
0.1856 10.0 420 0.2082 0.9313
0.1736 11.0 462 0.1306 0.9564
0.1423 12.0 504 0.0989 0.9732
0.1296 13.0 546 0.0949 0.9732
0.1158 14.0 588 0.1084 0.9631
0.1383 15.0 630 0.0865 0.9715
0.1384 16.0 672 0.0879 0.9715
0.0924 17.0 714 0.0758 0.9782
0.0966 18.0 756 0.0866 0.9682
0.1324 19.0 798 0.0876 0.9715
0.0995 20.0 840 0.0990 0.9648
0.083 21.0 882 0.0911 0.9698
0.082 22.0 924 0.0816 0.9799
0.1038 23.0 966 0.1453 0.9430
0.0751 24.0 1008 0.0877 0.9732
0.0733 25.0 1050 0.0878 0.9682
0.0813 26.0 1092 0.0688 0.9816
0.0788 27.0 1134 0.0732 0.9782
0.0617 28.0 1176 0.0722 0.9749
0.0568 29.0 1218 0.0883 0.9648
0.0701 30.0 1260 0.0703 0.9765
0.0535 31.0 1302 0.0792 0.9782
0.0716 32.0 1344 0.0684 0.9799
0.0419 33.0 1386 0.0666 0.9816
0.054 34.0 1428 0.0768 0.9749
0.0332 35.0 1470 0.0717 0.9799
0.0524 36.0 1512 0.1067 0.9715
0.0372 37.0 1554 0.0604 0.9816
0.0692 38.0 1596 0.0579 0.9799
0.038 39.0 1638 0.0824 0.9732
0.0524 40.0 1680 0.0635 0.9765
0.0429 41.0 1722 0.0644 0.9816
0.0705 42.0 1764 0.0747 0.9765
0.0325 43.0 1806 0.0685 0.9816
0.0446 44.0 1848 0.0683 0.9782
0.0439 45.0 1890 0.0707 0.9749
0.0346 46.0 1932 0.0642 0.9782
0.0504 47.0 1974 0.0654 0.9799
0.0379 48.0 2016 0.0651 0.9765
0.0433 49.0 2058 0.0654 0.9765
0.0337 50.0 2100 0.0655 0.9765

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.5.0+cu121
  • Datasets 3.0.2
  • Tokenizers 0.19.1
Downloads last month
48
Safetensors
Model size
27.8M 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 MakAIHealthLab/convnext-tiny-224-finetuned-biopsy

Finetuned
(36)
this model