Intro
The HEp-2 cell image classification model is a deep learning model designed specifically for the cell image classification task, using deep convolutional neural network techniques. The training data for the model comes from the HEp-2 cell image dataset, which originated from the cell image classification competition at the 2014 International Pattern Recognition Conference. The dataset consists of images categorized into a training set (8,701 images), a validation set (2,175 images), and a test set (2,720 images). In addition, a .csv file is provided containing two columns of data: the first column is the image ID, which matches the name of the image in the three datasets; the second column is the category of the cell image. The model draws on the classical structure of AlexNet and is based on a deep convolutional neural network, including components such as convolutional, pooling, and fully connected layers, with powerful image feature learning capabilities. Its main training goal is to acquire discriminative features for the HEp-2 cell image classification task in order to improve the classification performance on both validation and test sets. The model is trained to efficiently capture critical information in HEp-2 cell images in order to accurately classify the images.
Demo
https://huggingface.co/spaces/Genius-Society/HEp2
Usage
from modelscope import snapshot_download
model_dir = snapshot_download("Genius-Society/HEp2")
Dataset
https://huggingface.co/datasets/Genius-Society/HEp2
Training curve
| Accuracy |  |
|---|---|
| Loss |  |
Mirror
https://www.modelscope.cn/models/Genius-Society/HEp2
Reference
[1] https://github.com/Genius-Society/medical_image_computing/tree/hep2