🐦 Follow me on Twitter • 📧 Feedback

TorchLeet is a curated set of PyTorch practice problems, inspired by LeetCode-style challenges, designed to enhance your skills in deep learning and PyTorch.

Table of Contents

• TorchLeet
  • Table of Contents
  • Question Set
    • 🟢Easy
    • 🟡Medium
    • 🔴Hard
  • Getting Started
    • 1. Install Dependencies
    • 2. Structure
    • 3. How to Use
• Contribution
• Authors:

Question Set

🟢Easy

1. Implement linear regression (Solution)
2. Write a custom Dataset and Dataloader to load from a CSV file (Solution)
3. Write a custom activation function (Simple) (Solution)
4. Implement Custom Loss Function (Huber Loss) (Solution)
5. Implement a Deep Neural Network (Solution)
6. Visualize Training Progress with TensorBoard in PyTorch (Solution)
7. Save and Load Your PyTorch Model (Solution)

🟡Medium

1. Implement an LSTM (Solution)
2. Implement a CNN on CIFAR-10 (Solution)
3. Implement an RNN (Solution)
4. Use torchvision.transforms to apply data augmentation (Solution)
5. Add a benchmark to your PyTorch code (Solution)
6. Train an autoencoder for anomaly detection (Solution)

🔴Hard

1. Write a custom Autograd function for activation (SILU) (Solution)
2. Write a Neural Style Transfer
3. Write a Transformer (Solution)
4. Write a GAN (Solution)
5. Write Sequence-to-Sequence with Attention (Solution)
6. Quantize your language model (Solution)
7. [Enable distributed training in pytorch (DistributedDataParallel)]
8. [Work with Sparse Tensors]
9. Implement Mixed Precision Training using torch.cuda.amp (Solution)
10. Add GradCam/SHAP to explain the model. (Solution)

What's cool? 🚀

• Diverse Questions: Covers beginner to advanced PyTorch concepts (e.g., tensors, autograd, CNNs, GANs, and more).
• Guided Learning: Includes incomplete code blocks (... and #TODO) for hands-on practice along with Answers

Getting Started

1. Install Dependencies

• Install pytorch: Install pytorch locally
• Some problems need other packages. Install as needed.

2. Structure

• <E/M/H><ID>/: Easy/Medium/Hard along with the question ID.
• <E/M/H><ID>/qname.ipynb: The question file with incomplete code blocks.
• <E/M/H><ID>/qname_SOLN.ipynb: The corresponding solution file.

3. How to Use

• Navigate to questions/ and pick a problem
• Fill in the missing code blocks (...) and address the #TODO comments.
• Test your solution and compare it with the corresponding file in solutions/.

Happy Learning! 🚀

Contribution

Feel free to contribute by adding new questions or improving existing ones. Ensure that new problems are well-documented and follow the project structure.

Authors

Chandrahas Aroori 💻 AI/ML Dev

Caslow Chien 💻 Developer

Stargazers over time