68382
Learn Machine Learning with hands-on Python tutorials, real-world code examples, and clear explanations for researchers and developers. Admin: @HusseinSheikho || @Hussein_Sheikho
Build a Large Language Model from Scratch! 🚀
This repository provides code examples for developing, pretraining, and fine-tuning a Large Language Model (LLM) from the ground up. It serves as the official codebase for the book "Build a Large Language Model (From Scratch)." 📘
Notebook examples are included for each chapter:
Chapter 1: Understanding Large Language Models 🧠
Chapter 2: Working with Text Data 📝
Chapter 3: Coding Attention Mechanisms ⚙️
Chapter 4: Implementing a GPT Model from Scratch 🏗
Chapter 5: Pretraining on Unlabeled Data 📊
Chapter 6: Fine-tuning for Text Classification 🏷
Chapter 7: Fine-tuning to Follow Instructions 🗣
Repository: https://github.com/rasbt/LLMs-from-scratch 🔗
🔥2026 New IT Certification Prep Kit – Free!
SPOTO cover: #Python #AI #Cisco #PMI #Fortinet #AWS #Azure #Excel #CompTIA #ITIL #Cloud + more
✅ Grab yours free kit now:
• Free Courses (Python, Excel, Cyber Security, Cisco, SQL, ITIL, PMP, AWS)
👉 https://bit.ly/3Ogtn3i
• IT Certs E-book
👉 https://bit.ly/41KZlru
• IT Exams Skill Test
👉 https://bit.ly/4ve6ZbC
• Free AI Materials & Support Tools
👉 https://bit.ly/4vagTuw
• Free Cloud Study Guide
👉 https://bit.ly/4c3BZCh
💬 Need exam help? Contact admin: wa.link/w6cems
✅ Join our IT community: get free study materials, exam tips & peer support
https://chat.whatsapp.com/BiazIVo5RxfKENBv10F444
This channels is for Programmers, Coders, Software Engineers.
0️⃣ Python
1️⃣ Data Science
2️⃣ Machine Learning
3️⃣ Data Visualization
4️⃣ Artificial Intelligence
5️⃣ Data Analysis
6️⃣ Statistics
7️⃣ Deep Learning
8️⃣ programming Languages
✅ /channel/addlist/8_rRW2scgfRhOTc0
✅ /channel/Codeprogrammer
🚀 $0.15/GB - PROXYFOG.COM — SCALE WITHOUT LIMITS
💎 Premium Residential & Mobile Proxies
🌍 60M+ Real IPs — 195 Countries (🇺🇸 USA Included)
💰 Prices as low as $0.15/GB
🎯 Instant & Precise Country Targeting
🔄 Sticky Sessions + Fresh IP on Every Request
♾️ Balance Never Expires
⚡ Built for Arbitrage. Automation. Scraping. Scaling.
⚡ Fast. Stable. High-Performance Infrastructure.
👉 Website: https://tglink.io/7d5c3d9fd92485
📩 Telegram: /channel/proxyfog?utm_source=telegain&utm_medium=cpp&utm_campaign=s1&utm_content=codeprogrammer&utm_term=
Start today. Scale without limits. 🚀
The most complete list of video courses on Computer Science on the internet.
cs-video-courses — 78K+ stars.
MIT.
Stanford University.
University of California, Berkeley.
Harvard University.
Carnegie Mellon University.
Indian Institutes of Technology.
Princeton University.
California Institute of Technology.
Everything is free. All lectures are in video format. Everything is collected in one repository.
Topics:
→ Data structures and algorithms
→ Operating systems
→ Distributed systems
→ Database systems
→ Computer networks
→ Machine learning
→ Deep learning
→ Natural language processing (NLP)
→ Computer vision
→ Computer graphics
→ Security
→ Quantum computing
→ Robotics
→ Blockchain
From beginner level (CS50) to advanced (6.824 Distributed Systems).
The curriculum is free. 🤙
https://github.com/Developer-Y/cs-video-courses
/channel/CodeProgrammer ⚡️
Save & Share & Like 🏃♀️
When have you ever needed to add a mathematical description for your function in Python, but found that it takes too much time?
Non-programmers can't easily read Python's logic. However, manually converting it to LaTeX is slow and quickly becomes outdated as the code changes.
latexify_py solves this problem with a single decorator, generating LaTeX directly from your function, so that the mathematics remains readable and always synchronized with the code.
Main features:
• Three decorators for different outputs: expressions, full equations, or pseudocode
• Displays the rendered LaTeX directly in Jupyter cells
• Functions continue to work normally when called
In addition, latexify_py is open source. Install it using pip install latexify-py
An article about 3 tools that convert Python code to LaTeX: https://bit.ly/3Pw89yP
Run this code: https://bit.ly/4bW2ycE
/channel/CodeProgrammer
Follow the Machine Learning with Python channel on WhatsApp: https://whatsapp.com/channel/0029VaC7Weq29753hpcggW2A
Читать полностью…
🐼 Cheat Sheet on Data Wrangling — for everyone who works with Pandas
Everything you need is collected in one file: creating and merging DataFrames, filtering, grouping, handling missing values, and visualization.
It's convenient when you need to quickly refresh your syntax and don't want to dig into the documentation.
The cheat sheet in good quality
https://pandas.pydata.org/Pandas_Cheat_Sheet.pdf
tags: #useful
For more please ❤️
➡ /channel/CodeProgrammer
🚀 New Edge for Polymarket Traders: Oracle Lag Sniper
A high-performance, open-source strategy repo is making waves right now among serious Polymarket users: the Oracle Lag Sniper.
📈 Why it’s worth your attention:
• Exploits oracle timing inefficiencies
• Built for fast execution & precise entries
• Fully open-source, inspect, modify, and run it yourself
🔗 Check out the repo here:
Oracle Lag Sniper GitHub
Want more early signals like this, plus private insights and rising strategies to stay ahead of the curve? Subscribe to Polymarket Analytics for exclusive access:
Polymarket Analytics Pricing
📊 Don’t just follow the market, get ahead of it.
💾 TensorTonic — a platform with 200+ algorithms and tasks for ML developers
What's inside:
▶️ Analysis of research and step-by-step reproduction of model architectures;
▶️ Explanation of topics and concepts with interactive visualizations;
▶️ A progress and achievement system — what would we do without gamification.
A great option to hone your ML skills in the evening ⚡
https://www.tensortonic.com/
🗂 Harvard has released a textbook on ML systems — from autograd to production
Not just another "what is a neural network" course — this is about how to build combat-ready ML systems around models.
What's inside:
▶️ Building autograd, optimizers, attention, and mini-PyTorch from scratch;
▶️ Batches, computational accuracy, architectures, and training;
▶️ Performance optimization, hardware acceleration, and benchmarking.
You can read the book and the code for free right now.
✈️ Link to GitHub
https://github.com/harvard-edge/cs249r_book
Visualizing the complexity of algorithms
/channel/CodeProgrammer ✅
📱 TorchCode — a PyTorch training tool for preparing for ML interviews
40 tasks for implementing operators and architectures that are actually asked in interviews. Automatic checking, hints, and reference solutions — all in the browser without installation.
If you're preparing for an ML interview, it's useful to go through at least half of them.
Link: https://github.com/duoan/TorchCode
tags: #useful #pytorch
/channel/CodeProgrammer ✅
𝐕𝐢𝐬𝐮𝐚𝐥 𝐛𝐥𝐨𝐠 on Vision Transformers is live.
https://vizuaranewsletter.com/p/vision-transformers?r=5b5pyd&utm_campaign=post&utm_medium=web
Learn how ViT works from the ground up, and fine-tune one on a real classification dataset.
CNNs process images through small sliding filters. Each filter only sees a tiny local region, and the model has to stack many layers before distant parts of an image can even talk to each other.
Vision Transformers threw that whole approach out.
ViT chops an image into patches, treats each patch like a token, and runs self-attention across the full sequence.
Every patch can attend to every other patch from the very first layer. No stacking required.
That global view from layer one is what made ViT surpass CNNs on large-scale benchmarks.
𝐖𝐡𝐚𝐭 𝐭𝐡𝐞 𝐛𝐥𝐨𝐠 𝐜𝐨𝐯𝐞𝐫𝐬:
- Introduction to Vision Transformers and comparison with CNNs
- Adapting transformers to images: patch embeddings and flattening
- Positional encodings in Vision Transformers
- Encoder-only structure for classification
- Benefits and drawbacks of ViT
- Real-world applications of Vision Transformers
- Hands-on: fine-tuning ViT for image classification
The Image below shows
Self-attention connects every pixel to every other pixel at once. Convolution only sees a small local window. That's why ViT captures things CNNs miss, like the optical illusion painting where distant patches form a hidden face.
The architecture is simple. Split image into patches, flatten them into embeddings (like words in a sentence), run them through a Transformer encoder, and the class token collects info from all patches for the final prediction. Patch in, class out.
Inside attention: each patch (query) compares itself to all other patches (keys), softmax gives attention weights, and the weighted sum of values produces a new representation aware of the full image, visualizes what the CLS token actually attends to through attention heatmaps.
The second half of the blog is hands-on code. I fine-tuned ViT-Base from google (86M params) on the Oxford-IIIT Pet dataset, 37 breeds, ~7,400 images.
𝐁𝐥𝐨𝐠 𝐋𝐢𝐧𝐤
https://vizuaranewsletter.com/p/vision-transformers?r=5b5pyd&utm_campaign=post&utm_medium=web
Coursera scholarship
I will choose the first 30 people
You must register with this bot
Selection for those who want to become a certified Claude architect
Useful resources for preparation in one place 👇
• Registration for certification: https://anthropic.skilljar.com/claude-certified-architect-foundations-access-request
• Training (13 free courses):
https://anthropic.skilljar.com
• Cookbook (examples and practices):
https://github.com/anthropics/anthropic-cookbook
• Exam guide:
https://share.google/0eqIbebzRMUt8KTc8
• Practice questions:
http://claudecertifications.com
• MCP documentation:
http://modelcontextprotocol.io
API documentation:
http://docs.anthropic.com
Useful playbook:
https://drive.google.com/file/d/1luC0rnrET4tDYtS7xe5jUxMDZA-4qNf-/view
📊 Data Science Cheat Sheets
📦 596.3 MB | 👍 5.5K | ⬇️ 73.4K
📡 @DATASETS1
Please note that the permanent subscription to our Premium channel will be permanently closed in five days.
The cost of a permanent subscription to our premium channel is $35.
The Premium channel contains thousands of books and courses available for free as direct downloadable Telegram files.
Contact me @HusseinSheikho
💻 Collection of cheat sheets on SQL
I've gathered for you short and understandable cheat sheets on the main topics:
▶️ Basics of the SQL language;
▶️ JOINs with clear examples;
▶️ Window functions;
▶️ SQL for data analysis.
An excellent set to refresh your knowledge before a job interview or quickly recall the syntax.
tags: #sql #useful
/channel/DataAnalyticsX
🗂 20 free MIT courses — the entire Computer Science base in one place
#MIT has made courses in key CS areas publicly available. #Python, #algorithms, #ML, neural networks, #OS, #databases, #mathematics — all can be completed for free directly on #YouTube.
▶️ Introduction to Python Programming
▶️ Data Structures and Algorithms
▶️ Mathematics for Computer Science
▶️ Machine Learning
▶️ Deep Learning
▶️ Artificial Intelligence
▶️ Machine Learning in Healthcare
▶️ Database Management Systems
▶️ Operating Systems
▶️ One-Variable Calculus
▶️ Many-Variable Calculus
▶️ Introduction to Probability Theory
▶️ Statistics
▶️ Probability Theory and Statistics
▶️ Linear Algebra
▶️ Matrix Calculus for Machine Learning
▶️ Java Programming
▶️ Design and Analysis of Algorithms
▶️ Advanced Data Structures
▶️ Introduction to Computational Thinking
tags: #courses
➡/channel/python53
RAG won't work in 2026 if you're still using old approaches.
Yes, many companies are still failing with RAG — not because they're doing it wrong, but because they're stuck on outdated techniques.
Here's what usually happens: most companies start with a chatbot / chat app when talking about AI implementation. And here RAG becomes key — to connect their data via a database and enable the chat app to retrieve relevant documents.
But today, RAG is no longer limited to just chats. The applications of RAG are practically limitless, and that's a good thing.
RAG still remains the foundation for everything you build on LLMs and AI agents. The only thing that's changed is the RAG techniques themselves. The old approach no longer works — more advanced techniques are needed, what's now called advanced RAG.
The essence of RAG is to enrich the system with your data via a database so it can find relevant documents or their parts. The results are simple and often "okay", especially if the documents are well-structured and there aren't many of them.
But when the documents are unstructured and it's important to get not just accurate documents but also the right context, advanced techniques come into play:
- query decomposition
- metadata enrichment
- hybrid indexing
- reranking
- context fusion
These approaches allow the RAG system to find and generate more accurate and contextually relevant answers.
Therefore, advanced RAG is important. RAG isn't dead and can't die. Just use smarter techniques.
🚀 $0.15/GB - PROXYFOG.COM — SCALE WITHOUT LIMITS
💎 Premium Residential & Mobile Proxies
🌍 60M+ Real IPs — 195 Countries (🇺🇸 USA Included)
💰 Prices as low as $0.15/GB
🎯 Instant & Precise Country Targeting
🔄 Sticky Sessions + Fresh IP on Every Request
♾️ Balance Never Expires
⚡ Built for Arbitrage. Automation. Scraping. Scaling.
⚡ Fast. Stable. High-Performance Infrastructure.
👉 Website: https://tglink.io/13a3b748098cf2
📩 Telegram: /channel/proxyfog?utm_source=telegain&utm_medium=cpp&utm_campaign=s1&utm_content=codeprogrammer&utm_term=
Start today. Scale without limits. 🚀
Classical filters & convolution: The heart of computer vision
Before Deep Learning exploded onto the scene, traditional computer vision centered on filters. Filters were small, hand-engineered matrices that you convolved with an image to detect specific features like edges, corners, or textures. In this article, we will dive into the details of classical filters and convolution operation - how they work, why they matter, and how to implement them.
More: https://www.vizuaranewsletter.com/p/classical-filters-and-convolution
📱 Python enthusiasts, this is for you — 15 BEST REPOSITORIES on GitHub for learning Python
▶️ Awesome Python — https://github.com/vinta/awesome-python
— the largest and most authoritative collection of frameworks, libraries, and resources for Python — a must-save
▶️ TheAlgorithms/Python — https://github.com/TheAlgorithms/Python
— a huge collection of algorithms and data structures written in Python
▶️ Project-Based-Learning — https://github.com/practical-tutorials/project-based-learning
— learning Python (and not only) through real projects
▶️ Real Python Guide — https://github.com/realpython/python-guide
— a high-quality guide to the Python ecosystem, tools, and best practices
▶️ Materials from Real Python — https://github.com/realpython/materials
— a collection of code and projects for Real Python articles and courses
▶️ Learn Python — https://github.com/trekhleb/learn-python
— a reference with explanations, examples, and exercises
▶️ Learn Python 3 — https://github.com/jerry-git/learn-python3
— a convenient guide to modern Python 3 with tasks
▶️ Python Reference — https://github.com/rasbt/python_reference
— cheat sheets, scripts, and useful tips from one of the most respected Python authors
▶️ 30-Days-Of-Python — https://github.com/Asabeneh/30-Days-Of-Python
— a 30-day challenge: from syntax to more complex topics
▶️ Python Programming Exercises — https://github.com/zhiwehu/Python-programming-exercises
— 100+ Python tasks with answers
▶️ Coding Problems — https://github.com/MTrajK/coding-problems
— tasks on algorithms and data structures, including for preparation for interviews
▶️ Projects — https://github.com/karan/Projects
— a list of ideas for pet projects (not just Python). Great for practice
▶️ 100-Days-Of-ML-Code — https://github.com/Avik-Jain/100-Days-Of-ML-Code
— machine learning in Python in the format of a challenge
▶️ 30-Seconds-of-Python — https://github.com/30-seconds/30-seconds-of-python
— useful snippets and tricks for everyday tasks
▶️ Geekcomputers/Python — https://github.com/geekcomputers/Python
— various scripts: from working with the network to automation tasks
React ♥️ for more posts like this 💛
📱 A huge cheat sheet for Python, Django, Plotly, Matplotlib, Pygame
Many topics are covered inside:
🔸 All basic constructs: variables, conditions, loops, lists, dictionaries, functions, and classes — with clear examples;
🔸 Working with files, exceptions, and data input — understandable even for beginners;
🔸 #Django, #Pygame, #Matplotlib, and #Plotly — brief instructions on how to get started with each of the frameworks;
🔸 Tips on #Git, project structure, and unit testing.
/channel/CodeProgrammer ❤️
⚡️ Colorizing old black-and-white videos and "bringing faces to life" for FREE
SVFR — a full-fledged framework for restoring faces in videos.
It can:
💬 BFR — improve blurry faces.
💬 Colorization — colorize black-and-white videos.
💬 Inpainting — redraw damaged areas.
💬 and combine all of this in one pass.
Essentially, the model takes old or damaged videos and makes them "as if they were shot yesterday". And it's free and open-source.
⚙️ Installation locally:
1. Create an environment
conda create -n svfr python=3.9 -y
conda activate svfr
pip install torch==2.2.2 torchvision==0.17.2 torchaudio==2.2.2
pip install -r requirements.txt
conda install git-lfs
git lfs install
git clone https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt models/stable-video-diffusion-img2vid-xt
python infer.py \
--config config/infer.yaml \
--task_ids 0 \
--input_path input.mp4 \
--output_dir results/ \
--crop_face_region
0 — face enhancement1 — colorization2 — redrawing damage
Follow the Machine Learning with Python channel on WhatsApp: https://whatsapp.com/channel/0029VaC7Weq29753hpcggW2A
Читать полностью…
Do you want to understand the methods used to train LLMs?
The training of large language models (LLMs) is based on various approaches that help models understand and generate text.
Each method shapes the learning process in its own way - from predicting the next word to classifying entire sentences or labeling entities.
Here are 4 common methods of training LLMs in simple language 👇
1. Causal Language Modeling
Predicts the next word in a sequence based on the previous ones. Helps the model master the natural flow of speech and the structure of sentences.
Analogy: how to finish a sentence for another person by guessing the next word.
2. Masked Language Modeling
Learns by guessing the missing words in a sentence based on the surrounding context. Improves the overall understanding of language.
Analogy: how to solve tasks with missing words.
3. Text Classification Modeling
Determines the general class of a sentence (for example, tone or topic) by comparing predictions with actual labels.
Analogy: how to sort letters into folders "Work", "Personal", or "Promotions".
4. Token Classification Modeling
Assigns labels to each word or subword - for example, highlights names, places, or dates in the text.
Analogy: how to highlight words with different colors - names in blue, places in green, dates in yellow.
These methods form the basis of modern LLMs, and each of them plays a role in making AI smarter and more useful.
/channel/CodeProgrammer