metadata
license: apache-2.0
tags:
- generated_from_trainer
- medical
datasets:
- imagefolder
metrics:
- accuracy
- f1
- recall
- precision
model-index:
- name: vit-large-patch32-384-Hyper_Kvasir_Labeled_Images
results: []
language:
- en
pipeline_tag: image-classification
vit-large-patch32-384-Breast_Histopathology_Images
This model is a fine-tuned version of google/vit-large-patch32-384.
It achieves the following results on the evaluation set:
- Loss: 0.3954
- Accuracy: 0.8202
- F1
- Weighted: 0.8151
- Micro: 0.8202
- Macro: 0.7674
- Recall
- Weighted: 0.8202
- Micro: 0.8202
- Macro: 0.7549
- Precision
- Weighted: 0.8141
- Micro: 0.8202
- Macro: 0.7860
Model description
For more information on how it was created, check out the following link: https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/blob/main/Computer%20Vision/Image%20Classification/Binary%20Classification/Breast%20Histopathology%20Images/Breast_Histopathology_Images_Using_ViT.ipynb
Intended uses & limitations
This model is intended to demonstrate my ability to solve a complex problem using technology.
Training and evaluation data
Dataset Source: https://huggingface.co/datasets/EulerianKnight/breast-histopathology-images-train-test-valid-split
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.005
- train_batch_size: 64
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted F1 | Micro F1 | Macro F1 | Weighted Recall | Micro Recall | Macro Recall | Weighted Precision | Micro Precision | Macro Precision |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.3536 | 1.0 | 649 | 0.3568 | 0.8455 | 0.8411 | 0.8455 | 0.8003 | 0.8455 | 0.8455 | 0.7863 | 0.8411 | 0.8455 | 0.8205 |
| 0.4417 | 2.0 | 1298 | 0.3954 | 0.8202 | 0.8151 | 0.8202 | 0.7674 | 0.8202 | 0.8202 | 0.7549 | 0.8141 | 0.8202 | 0.7860 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3