GDGoC Comsats Attock ML/DL Fellowship: Week 1 Recap

It’s been an exciting week for Developer Groups on Campus(GDGoC) Comsats Attock! As a Community Engagement Lead, I’m proud to see the launch of our ML/DL fellowship, led by Nimra Waqar. While the concepts we covered are not new to me, it's a welcome refresher; seeing the enthusiasm and learning progress of our fellows has been incredibly rewarding.

This fellowship is a significant step in enhancing our core team’s capabilities, and it’s been fantastic to see everyone, including myself, engage so passionately.

Starting with the Basics: AI, ML, DL, and Data Science:

Our journey began with demystifying the core terminologies. We watched a fantastic video by Krish Naik and read an informative article ‘Data Science vs. Artificial Intelligence vs. Machine Learning vs. Deep Learning’, which really helped clarify the distinctions between Artificial Intelligence (AI), Machine Learning (ML), Deep Learning (DL), and Data Science.

To recap, we learned that:

• AI represents the broad concept of machines mimicking human intelligence.
• ML is a subset of AI where machines learn from data without explicit programming.
• DL is a specialized area of ML that uses neural networks with multiple layers for complex data analysis.
• Data Science is the multidisciplinary field that extracts knowledge from data.

It’s a classic nested structure, with AI encompassing ML, which in turn encompasses DL. Data Science provides the tools and techniques to navigate them all.

Python from First Principles: No Libraries Allowed!

Task two was a deep dive into Python fundamentals, requiring us to build functions from scratch — no libraries! This was a great exercise in solidifying our understanding of the underlying math. We tackled:

• Calculating the mean, median, and mode, is a classic statistics refresher.
• Computing variance and the standard deviation is vital for understanding data spread.
• Determining Euclidean distance is crucial for measuring data point similarity.
• Implementing the sigmoid function: A cornerstone of neural networks.

The main focus of the first week was not to understand these concepts in depth(we will cover that later) but to know how to write mathematical formulas in Python. This hands-on approach reinforced the importance of the mathematical foundations in ML/DL.

Mathematical Muscle: MSE, RMSE, Cosine Similarity, and More

Task three pushed us further, requiring us to implement:

• Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) are essential for regression model evaluation.
• Cosine similarity is a key metric for vector similarity.
• Linear regression is a fundamental predictive modeling technique.
• The softmax function: Used for multiclass classification.

softmax

Doing this without libraries was a bit challenging but valuable exercise, ensuring we truly understood the underlying formulas and their implementation.

Kaggle and Practical Application

We also explored Kaggle, completing the first exercise of the “Intro to Programming” course. The interactive exercises were a fun and engaging way to apply our knowledge.

All my work of fellowship is available on my GitHub repository and will be updated regularly: ML/DL fellowship by GDGoC Comsats Attock

Reflections and Endorsement

While the first week served as a refresher for me, I’m incredibly impressed with the progress of our fellows and the excellent start Nimra has provided. This fellowship is a testament to the dedication of our team and the eagerness of our community to learn.

Key takeaways:

• This fellowship is a great step to upscale our core team skills.
• Understanding the math behind ML/DL is just as important as coding skills.
• Python is an indispensable tool.
• Platforms like Kaggle make learning engaging and accessible.
• GitHub is important to keep learnings open source, collaborative learning, and organized project management.
• The fellowship tasks are designed to create a strong mathematical and programmatic base.

I’m excited for the upcoming weeks and the continued growth of our community. Stay tuned for more updates on our ML/DL journey!

Amaan signing off!