search

Google-Developer-Student-Club-CCOEW / AIML-2023-GDSC-CUMMINS-X-GDSC-MMCOE

Welcome to the Hacktoberfest 2023 AI-ML Cohort for Cummins College and MMCOE students! To request issue assignment, create a pull request, providing: 1. Full Name 🧑‍🎓 2.Email 📧 3.College ID (RNO) 🔢 4.Branch of Study.📚 5. Year 📆 .The Cummins College and MMCOE students' PRs will be considered only. Thank you!
1 stars 6 forks source link

Create a Neural Network for Traffic Sign Recognition #12

Open Gdsc-Cummins opened 9 months ago

Gdsc-Cummins commented 9 months ago

Hacktoberfest Open Source Contribution - Traffic Sign Recognition Project

Issue: Create a Neural Network for Traffic Sign Recognition

Description:

We are seeking contributors to develop a neural network model for identifying traffic signs in photographs. The dataset provided consists of 43 subdirectories, numbered 0 through 42, with each subdirectory representing a different category of road sign. Your task is to create a neural network that can accurately classify these traffic signs.

Requirements:

  1. Utilize the OpenCV-Python module (cv2) to read each image as a numpy.ndarray (a numpy multidimensional array).

  2. Resize each image to a common size so that they can be passed into the neural network.

  3. Use TensorFlow or PyTorch to define the architecture of the neural network.

  4. Implement the load_data() function for loading the dataset and preprocessing the images.

  5. Develop the get_model() function to define the neural network architecture.

  6. Train the model using the provided dataset and validate its performance.

Documentation:

In a separate file called README.md, document your experimentation process. Describe what you tried, what worked well, what didn't work well, and any observations you made during the development of the neural network model. This documentation will help others understand your thought process and provide insights into your approach.

Reference Code and Dataset:

You can find reference code and the dataset in the "gtsrb" folder inside the "level-hard" folder of our GitHub repository. Please refer to these resources to get started.

Note: The pull request with the best solution, as determined by our community and project maintainers, will be merged at the end of October. Make sure to submit your contributions and participate in Hacktoberfest for a chance to have your work incorporated into the project.

Thank you for your interest in contributing to our traffic sign recognition project. Your efforts will contribute to safer road navigation and benefit the community. Happy coding!

PrasannaBG commented 8 months ago

I like to contribute to this issue. Assign me this issue

parthsolanke commented 8 months ago

@PrasannaBG you can start contributing without being assigned to this issue.