maitrechen / medgan-reslite

一个用于肺炎图像分类的轻量级ResNet18-SAM模型实现，采用SH-DCGAN生成少类样本数据，解决了数据不平衡的问题，同时结合剪枝策略实现轻量化！MedGAN-ResLite-V2 Released! Stay tuned!❤

Hi there 👋

Welcome to my coding haven where pixels turn into magic!

• 🔭 Currently working on: My graduation project titled "Study on Few-shot Enhancement for Pneumonia Image Classification and Lightweight Research". It's expected to be completed around May, after which I'll update the MedGAN-ResLite project.

• 🌱 I’m currently learning:

  • C++
  • Generative Adversarial Networks (GANs)
  • Medical Image Processing

• 🤔 I’m looking for help with:

  • Tricks for Improving DCGAN networks
  • Introducing residual connections in DCGAN networks effectively
  • How to Balancing the generator and discriminator when training

• 💬 Ask me about:

  • Image classification
  • Image generation
  • OpenVINO model deployment
  • Model pruning

• 📫 How to reach me:

  • If you're my classmate, you can find me directly at the dormitory~~
  • If you're my junior, we might meet on campus.
  • If we haven't met before, you can reach me on my [Zhihu] or [CSDN] profiles.

Feel free to join me in this electrifying escapade as we decode the universe's secrets, one line of code at a time! 💻✨