Adekeye Olayemi

MSc AI | Learning ML & Documenting The Journey | Decoding patterns, one algorithm at a time 🙂

[Adekeye Olayemi @Pro_lynk]

@Timi_moses24 @dammiedammie35 😂😂😂😭

[Adekeye Olayemi @Pro_lynk]

@vtechthedev 1. All rows where column has a value.
2. Same answer as the first question

NOT (column IS NULL) is basically the same as WHERE column IS NOT NULL.

[Adekeye Olayemi @Pro_lynk]

@Ademola_Odumesi This is where I regularly order my pounded yam from 😅.

[Adekeye Olayemi @Pro_lynk]

@berrytom02 What’s your estimated timeline for reaching your target weight ?

[Jino Rohit @jino_rohit]

miss the days when all you needed was a kaggle/colab T4 GPU. now you need a 8xH100 to just load a model.

[Adekeye Olayemi @Pro_lynk]

@ManusAI Impressive

[Adekeye Olayemi @Pro_lynk]

@jayhemz 😂😂😂😭

[3rdSon @3rdsonx]

Thank God for free GPUs provided by google colab, kaggle, etc
It’s really saving lives mehnnn

[Adekeye Olayemi @Pro_lynk]

@ebukagaus 😂😂😂

[Adekeye Olayemi @Pro_lynk]

@ArmanHezarkhani AIE

[Brian Armstrong @brian_armstrong]

One of my favorite lessons I’ve learnt from working with smart people:

Action produces information. If you’re unsure of what to do, just do anything, even if it’s the wrong thing. This will give you information about what you should actually be doing.

Sounds simple on the surface - the hard part is making it part of your every day working process.

[Adekeye Olayemi @Pro_lynk]

🔍 Understanding Regression in Machine Learning (A view of the deep dive I did today).

One of the first and most fundamental concepts in ML is regression.
It’s all about predicting continuous outcomes and finding relationships between variables.
At its core, regression answers a simple question:
👉 How does one variable change when another changes?
💡 Types of Regression
Linear Regression: fits a straight line, ideal for simple relationships.

Multiple Linear Regression: models several independent variables.

Polynomial Regression: captures non-linear trends by adding higher-degree terms.

Regularized Regression (Ridge, Lasso, ElasticNet): helps prevent overfitting by penalizing large coefficients.

🔢 Mathematical Core:
The model tries to minimize the loss function, often the Mean Squared Error (MSE)
The smaller this error, the better your line (or curve) fits the data.

Practical Tip:
Always visualize residuals (the differences between predicted and actual values). Patterns in residuals often signal bias, variance issues, or model misfit.
Next, I’ll explore how regression connects to classification through concepts like logistic regression and decision boundaries.
What regression challenges have you encountered in your ML journey: data quality, multicollinearity, or model bias?

#MachineLearning #Regression #DataScience #Statistics #AI #MLRoadmap #LinearRegression #LearningJourney

[ℏεsam @Hesamation]

Stanford just dropped a 1:47 hour masterclass about LLM training on YouTube. this is the 4th lecture in their new Fall 2025 series of transformers and LLM.

this is the new go-to course on LLMs imo, and you need to take it seriously.

subjects in this video:
→ Pretraining
→ Scaling laws, Chinchilla law
→ Training optimizations overview
→ Data parallelism with ZeRO
→ Model parallelism
→ Flash Attention
→ Quantization
→ Mixed precision training
→ Supervised finetuning
→ Instruction tuning
→ Parameter-efficient finetuning with LoRA
→ QLoRA

[Adekeye Olayemi @Pro_lynk]

STEP 1:

Mathematical Foundations for Machine Learning

The deeper I go into AI, the more I realize that mathematics isn’t optional it’s the language of ML.

This first stage of my roadmap is all about building a solid foundation in the math concepts that power every algorithm, from linear regression to neural networks.

I’ll be covering:
🔹 Linear Algebra – vectors, matrices, transformations
🔹 Calculus – gradients, optimization, backpropagation
🔹 Probability & Statistics – understanding data distributions and uncertainty
🔹 Discrete Mathematics & Logic – structure and reasoning in algorithms

My goal isn’t to just “do math” but to understand why it matters in the learning process.

Over the coming days, I’ll share:
Practical ways to connect these topics to ML
Resources I’m using

Small exercises and visual examples I create along the way

🧠 To those working in ML:
What’s one mathematical concept you wish you had mastered earlier in your journey?

#MachineLearning #Mathematics #AI #LearningJourney #DeepLearning #DataScience #MLRoadmap

[Adekeye Olayemi @Pro_lynk]

A transformation is only linear if it has this two properties :

1. All lines must remain lines without getting curved.
2. The origin must remain fixed in place.

[Adekeye Olayemi @Pro_lynk]

I’ll be breaking each of these areas into smaller, digestible topics and sharing my learning process along the way.

#MachineLearning #ArtificialIntelligence #DeepLearning #MLRoadmap #DataScience #MLOps #AICommunity #LearningJourney

[Adekeye Olayemi @Pro_lynk]

I’m mastering Machine Learning from scratch and documenting every step✨

Here’s how I’m structuring my ML journey

1️⃣ Mathematics for Machine Learning
2️⃣ Programming & Software Engineering Essentials
3️⃣ Data Handling & Databases
4️⃣ Core Machine Learning Theory & Practice

[Adekeye Olayemi @Pro_lynk]

5️⃣ Deep Learning and Neural Networks
6️⃣ Specialized Machine Learning Domains
7️⃣ Model Deployment and MLOps
8️⃣ Cloud Computing and Distributed Systems
9️⃣ Ethics, Security, and Responsible AI
🔟 Portfolio and Career Development

[Adekeye Olayemi @Pro_lynk]

@bl_airee @coO_Ooi @_tea_anna I would love to have those informations too. Thank you. @_tea_anna