Adib_Reza

Hey folks! I'll be hosting an online session tomorrow on Neural Networks, where I'll demonstrate the simplicity behind these fascinating structures. I'll show you how analyzing a simple function is the key to understanding the complex structure of a Neural Network, and I'll break down how Backpropagation works. In fact, we will create a Neuron, layers of Neurons, and lastly, an MLP where these layers team up and work their magic together.

I will take the session tomorrow(Friday 26th) from 8:00 pm to 9:00 pm. Don't miss out!

Registration link: https://docs.google.com/forms/d/e/1FAIpQLScBdrCmUTY-V3oE1kWXWu1C3-uFiJ706Wud34PVgcim6tjbRw/viewform?usp=sf_link

একাডেমিক প্রেসারের মধ্যে নন একাডেমিক প্রজেক্ট করাটা একটা গিলটি প্লেজার আসলে। পেজে পোস্ট করার মতো কোনো কন্টেন্ট বানাতে একটু কষ্ট হচ্ছে তাই ভাবলাম আমি যেটা করছি সেটাই বলি, মন্দ কী?

মেশিন লার্নিং এর এন্ড-টু-এন্ড একটা প্রজেক্ট করলাম। বেশ ইন্টারেস্টিং ছিল। সিম্পল সিম্পল জায়গায় আটকায় গেছিলাম বাট মাথার এককোটি চুল ছিঁড়ে হইলেও সলভ করেছি। আ হেয়ারলেস ভিক্টরি। AWS ক্লাউডে ডিপ্লয় করেছি, Elastic Beanstalk আর Codepipeline ব্যবহার করেছি, তারপর Docker, Kubernetes ও কেমন করে ইউজ করতে হয়, সেটা মোটামুটি শিখেছি। অনেক কিছু একটু বুঝতে সমস্যা হয়েছে তবে চ্যাটজিপিটি স্যার সুন্দর মতো বুঝায় দিয়েছে। এখন নিজে থেকে একটা ইমপ্লিমেন্ট করা লাগবে বাট ফাইনালের পরে সেটা।

লিংক দিচ্ছি, ইন্টারেস্টেড থাকলে করে ফেলো:https://youtube.com/playlist?list=PLZoTAELRMXVPS-dOaVbAux22vzqdgoGhG&si=S6sXsjzPcv7q5aiI

Github link: https://github.com/AdibReza/end-to-end-ML-project

GitHub - AdibReza/end-to-end-ML-project Contribute to AdibReza/end-to-end-ML-project development by creating an account on GitHub.

I wish I could visualize the nulls inside my brain just like this. Need to handle those asap

I have finally understood how to teach machine learning in the most effective way.

My plan is to initially concentrate on demonstrating how to handle various cases, such as Null values, Label Encoding, Underfitting, and Overfitting, Data Normalization, and so on. We will learn when to use each method since one technique might be useful in one dataset but not in another. We will understand these concepts with our hearts.

After that, we will dive into hands-on projects that showcase the practical application of these handling techniques we have just learned. We will frequently jump between theory and projects, which will allow you to learn by understanding the necessity of these techniques rather than memorizing the Machine Learning concepts.

Join me on this journey guys. It will be fun. Let's feel what ML has to offer.

Funny thing about opening a page is, I constantly think about what content I should post next. And this makes me wanna learn more stuff. Lazy me has finally started organizing projects and planning to do sessions.

Motivation: +10
Urge to start a series: -20 (dhet)

Know the different branches of Machine Learning. Here's a simplified and short map of it.

Supervised ML:
In supervised learning, the algorithm is trained on a labeled dataset, where each input is paired with its output. The goal is to learn a mapping from inputs to outputs, understanding the pattern. The objective is to make predictions on new, unseen data. 🔄

Unsupervised ML:
Unsupervised learning teaches a model without using labeled data. It figures out patterns and relationships in the data, grouping similar things together. This helps the model make predictions when faced with new, unlabeled information. 🧩

Reinforcement Learning:
Reinforcement learning is like teaching a computer through trial and error. The computer learns by getting rewards for good actions and punishments for mistakes. This helps it figure out the best way to achieve a goal in a given situation. 🎮

Ensemble Learning:
Ensemble learning is like teamwork for models. Instead of relying on just one model, it combines the opinions of several models to make better predictions. It's like having a group discussion where everyone's input is considered, resulting in a more accurate and reliable decision. 🤝

Neural Network & Deep Learning:
Neural networks are like digital brains inspired by how our brains work. They're made up of connected nodes that learn from data to make predictions.
Deep learning is like using a super-smart brain with many layers of nodes. It's excellent for figuring out complex patterns in things like images or language. 🧠💡

Trust me, once you dive into learning these different branches, you'll discover the incredible possibilities they offer for handling and understanding data. 🚀

Join a live stream with the amazing Andrew Ng and learn to build your custom computer vision model in just one hour! Here's what you'll gain from this quick session:

1. Identify and scope vision applications.
2. Choose a vision project type/model.
3. Apply Data-Centric AI.
4. Deploy a Computer Vision model.

Date: November 6th
Time: 10:00 AM - 11:00 AM Pacific Time
Dhaka Time: 12:00 AM - 1:00 PM

Don't miss out! Register now using this: https://landing.ai/building-computer-vision-applications/?fbclid=IwAR1_0GTFkvkU7s8uhtMwEK_OB3LeR8rPKhcPsUAbHiwVtCB4jBUiEPhNze8

Get ready to enhance your computer vision skills with a true expert. See you there!

Done with the ML basics? Start doing lots of projects! Gaining hands-on experience through projects is a great way to solidify your understanding of machine learning and develop practical skills.

After completing at least 10 to 15 projects, revisit some ML mathematical theories so that you can visualize what you're learning. You have multiple options: you can read books, watch YouTube videos, or even take a course on Coursera.

"Siddharthan" is one of my favorite YouTube channels for learning and practicing Machine Learning. He offers multiple playlists, including one named "Machine Learning Projects." I recommend starting with that one.

🔗 Here are the links:

Machine Learning Projects: https://youtube.com/playlist?list=PLfFghEzKVmjvuSA67LszN1dZ-Dd_pkus6&si=H18D__uchlkxLr9S

Mathematics for ML: https://www.coursera.org/specializations/mathematics-for-machine-learning-and-data-science?specialization=jhu-data-science