search

UTSAVS26 / PyVerse

PyVerse is an open-source collection of diverse Python projects, tools, and scripts, ranging from beginner to advanced, across various domains like machine learning, web development, and automation.
https://sites.google.com/view/pyverse-python-universe/
MIT License
70 stars 205 forks source link

[Code Addition Request]: Add a new project #919

Closed Dharun235 closed 2 weeks ago

Dharun235 commented 2 weeks ago

Have you completed your first issue?

Guidelines

Latest Merged PR Link

https://github.com/UTSAVS26/PyVerse/pull/900

Project Description

Semi-Supervised Image Classification

The main idea is to use a consistency regularization approach to leverage the unlabeled data for training. The process involves:

  1. Training a neural network on a small labeled dataset.

  2. Adding consistency regularization to utilize the large unlabeled dataset by enforcing that the model's predictions for augmented versions of the same image should be similar.

    Use the CIFAR-10 dataset, which consists of 10 different classes of images. Split the dataset into labeled and unlabeled subsets. For example, use only 10% of the dataset as labeled and the remaining 90% as unlabeled. The model will be trained to make similar predictions for different augmented versions of the same image (e.g., using random cropping, flipping, etc.).

Full Name

Dharun Kumar

Participant Role

Contributor in GSSOC'EXTD and Hacktoberfest

github-actions[bot] commented 2 weeks ago

🙌 Thank you for bringing this issue to our attention! We appreciate your input and will investigate it as soon as possible.

Feel free to join our community on Discord to discuss more!

UTSAVS26 commented 2 weeks ago

@Dharun235 raise 1 issue at a time.

github-actions[bot] commented 2 weeks ago

✅ This issue has been closed. Thank you for your contribution! If you have any further questions or issues, feel free to join our community on Discord to discuss more!

Dharun235 commented 2 weeks ago

@UTSAVS26 Okay. Thanks.