Data Analytics
Don't forget to try it; it's free and includes most AI models.
Forwarded from Machine Learning with Python
Follow the Machine Learning with Python channel on WhatsApp: https://whatsapp.com/channel/0029VaC7Weq29753hpcggW2A
WhatsApp.com
Machine Learning with Python
Channel β’ 7.9K followers β’ Learn Machine Learning with hands-on Python tutorials, real-world code examples, and clear explanations for researchers and developers.
Forwarded from Machine Learning with Python
Iβm Eng. Hussein Sheikho
Promote your ad across all our listed channels for only $45!
Your ad will be published for 20 days across all our channels,
plus it will be pinned for 7 days
Want your tech channel to grow fast?
You can add your channel to our promo folder for just $20/month β
average growth rate 2000+ subscribers/month
Our Share folder (our channels)
https://t.me/addlist/8_rRW2scgfRhOTc0
Please open Telegram to view this post
VIEW IN TELEGRAM
Telegram
ENG. Hussein Sheikho
Experience in Deep Learning and Computer Vision + Python Project
π1
π©π»βπ» Stop saving dozens of different Claude guides that you'll never actually read! This list contains only the resources that are truly useful for real-world projects.
Please open Telegram to view this post
VIEW IN TELEGRAM
β€3
π Comprehensive Course on Hugging Face Transformers
A comprehensive course for those who want to understand transformers and LLMs.
It covers the architecture of transformers, model training and optimization, working with text sequences, and language modeling. RNNs and LSTMs for NLP tasks are also discussed separately.
βοΈ Link to the course
https://www.youtube.com/playlist?list=PLOj3JD_j8uXEsxFZGcRyjI_ZawoSvCfti
A comprehensive course for those who want to understand transformers and LLMs.
It covers the architecture of transformers, model training and optimization, working with text sequences, and language modeling. RNNs and LSTMs for NLP tasks are also discussed separately.
βοΈ Link to the course
https://www.youtube.com/playlist?list=PLOj3JD_j8uXEsxFZGcRyjI_ZawoSvCfti
β€3
Personal AI assistant in 5 minutes
No code. No card. Free π³
Works in Telegram, WhatsApp, or Discord β just send it tasks by voice or text. It gets things done, not just tells you how to do them.
β Create your personal AI assistant here β
getamplify.team
No code. No card. Free π³
Works in Telegram, WhatsApp, or Discord β just send it tasks by voice or text. It gets things done, not just tells you how to do them.
β’ reads and sends emails
β’ creates and edits Google Sheets
β’ uploads files to Google Drive
β’ works in Notion
β’ sends reminders
β’ generates PDFs, images, and videos
β’ actually makes life and work easier
β Create your personal AI assistant here β
getamplify.team
β€1
If you're interested in learning how to train a large language model from scratch, I recommend checking this out β
François Chollet, co-founder of the ARC Prize and creator of Keras, recently suggested a path for learning about LLMs from the ground up.
He said that if you're 17 years old or any age, and you want to learn how to create LLMs from scratch, simply read chapters 15 and 16 of his book Deep Learning with Python.
I quickly skimmed through them, and these two chapters are definitely worth saving.
Chapter 15 starts with the most basic language models and gradually progresses to:
Character-level Language Model β Seq2Seq β Attention β QKV β Scaled Dot-Product Attention β Multi-Head Attention β Self-Attention β Transformer
There's even a dedicated section explaining why Dot-Product Attention works in the first place.
He himself says that this is one of the best explanations of the topic.
The author doesn't just stop at formulas like
Chapter 16 is even more practical. It directly shows how to train a mini-GPT from scratch.
It explains how to take nearly 1 billion tokens from C4, create a SentencePiece vocabulary of 32,000 tokens, train a mini-GPT with 41 million parameters, 8 layers, 8 attention heads, and a hidden state size of 512, and then build a data pipeline, implement weight tying, learning rate warmup, pre-training, and generation with temperature and top-k.
From the tokenizer, data pipeline, causal attention, weight tying, and learning rate warmup to pre-training, greedy decoding, temperature, and top-k sampling. In essence, you are guided step-by-step through the entire process of training a GPT model.
You don't even need an expensive server for this.
The official notes state that the entire example can be run on a free T4 in Google Colab. Training takes about 6 hours. On an A100, it takes just over an hour.
Further in the book, Gemma, SFT, RLHF, RAG, and multimodal models are discussed.
So, if someone asks me:
"I've never trained a large model before. Where do I start?"
These two chapters can really be a great starting point.
The third edition of Deep Learning with Python is currently available for free online, and all the accompanying notebooks are fully open-source. You can simply download them into Colab and run them.
In 2026, it won't be necessary to immediately dive into a hundred research papers to learn about LLMs.
If you train a GPT model with 41 million parameters yourself, from data preparation to text generation, many concepts will naturally fall into place.
Link: https://deeplearningwithpython.io/
François Chollet, co-founder of the ARC Prize and creator of Keras, recently suggested a path for learning about LLMs from the ground up.
He said that if you're 17 years old or any age, and you want to learn how to create LLMs from scratch, simply read chapters 15 and 16 of his book Deep Learning with Python.
I quickly skimmed through them, and these two chapters are definitely worth saving.
Chapter 15 starts with the most basic language models and gradually progresses to:
Character-level Language Model β Seq2Seq β Attention β QKV β Scaled Dot-Product Attention β Multi-Head Attention β Self-Attention β Transformer
There's even a dedicated section explaining why Dot-Product Attention works in the first place.
He himself says that this is one of the best explanations of the topic.
The author doesn't just stop at formulas like
QKα΅ / βd. He starts with Word2Vec and embedding spaces, and then explains how the Transformer layer by layer, through Attention, gradually transforms the relationships between tokens into distance relationships in a vector space.Chapter 16 is even more practical. It directly shows how to train a mini-GPT from scratch.
It explains how to take nearly 1 billion tokens from C4, create a SentencePiece vocabulary of 32,000 tokens, train a mini-GPT with 41 million parameters, 8 layers, 8 attention heads, and a hidden state size of 512, and then build a data pipeline, implement weight tying, learning rate warmup, pre-training, and generation with temperature and top-k.
From the tokenizer, data pipeline, causal attention, weight tying, and learning rate warmup to pre-training, greedy decoding, temperature, and top-k sampling. In essence, you are guided step-by-step through the entire process of training a GPT model.
You don't even need an expensive server for this.
The official notes state that the entire example can be run on a free T4 in Google Colab. Training takes about 6 hours. On an A100, it takes just over an hour.
Further in the book, Gemma, SFT, RLHF, RAG, and multimodal models are discussed.
So, if someone asks me:
"I've never trained a large model before. Where do I start?"
These two chapters can really be a great starting point.
The third edition of Deep Learning with Python is currently available for free online, and all the accompanying notebooks are fully open-source. You can simply download them into Colab and run them.
In 2026, it won't be necessary to immediately dive into a hundred research papers to learn about LLMs.
If you train a GPT model with 41 million parameters yourself, from data preparation to text generation, many concepts will naturally fall into place.
Link: https://deeplearningwithpython.io/
β€4
Personal AI assistant in 5 minutes
No code. No card. Free π³
Works in Telegram, WhatsApp, or Discord β just send it tasks by voice or text. It gets things done, not just tells you how to do them.
β Create your personal AI assistant here β
getamplify.team
No code. No card. Free π³
Works in Telegram, WhatsApp, or Discord β just send it tasks by voice or text. It gets things done, not just tells you how to do them.
β’ reads and sends emails
β’ creates and edits Google Sheets
β’ uploads files to Google Drive
β’ works in Notion
β’ sends reminders
β’ generates PDFs, images, and videos
β’ actually makes life and work easier
β Create your personal AI assistant here β
getamplify.team
π₯2β€1
Data Analytics
Personal AI assistant in 5 minutes No code. No card. Free π³ Works in Telegram, WhatsApp, or Discord β just send it tasks by voice or text. It gets things done, not just tells you how to do them. β’ reads and sends emails β’ creates and edits Google Sheetsβ¦
A unique experience, I recommend you try it.
We found an open-source course covering Transformers, LoRA, RAG, prompts, model editing, and other key topics.
After each chapter, you can immediately access the original sources β the authors have compiled papers and collections from arXiv.
https://github.com/ZJU-LLMs/Foundations-of-LLMs
Please open Telegram to view this post
VIEW IN TELEGRAM
β€4
π 10 GitHub repositories that are gaining the most traction right now
Here's a fresh selection of projects that are experiencing significant growth and active discussion.
1. OpenAI Codex
An OpenAI coding agent for the terminal and development automation.
https://github.com/openai/codex
2. God's Eye View
OSINT and open geospatial data on an interactive 3D globe: aviation, satellites, and other sources.
https://github.com/bilawalsidhu/gods-eye-view
3. awesome-gpt-image-2
A large library of prompts, examples, and templates for GPT Image 2.
https://github.com/freestylefly/awesome-gpt-image-2
4. mattpocock/skills
Ready-made skills and workflows for coding agents that can be reused instead of using large prompts.
https://github.com/mattpocock/skills
5. Archify
Builds an interactive map of the codebase architecture: sequence, data flow, lifecycle, and component relationships.
https://github.com/tt-a1i/archify
6. Tailcat
A tool from Tailscale for directly connecting machines via WireGuard and NAT traversal.
https://github.com/tailscale/tailcat
7. Prime Agent
A self-improving coding agent for long, autonomous tasks.
https://github.com/PrimeIntellect-ai/prime-agent
8. Semantica
A graph-native infrastructure for storing context, relationships, and data provenance in AI systems.
https://github.com/semantica-agi/semantica
9. Cursor Plugins
The official repository of Cursor plugins: orchestration, code review, continual learning, and other agent functions.
https://github.com/cursor/plugins
10. AnyDoc
A tool from Firecrawl for converting documents into structured content for AI and RAG.
https://github.com/firecrawl/anydoc
π Looking at the overall trend for the week:
AI agents β agent skills β memory and context β visualization of architecture β tools for autonomous work with the environment
πΎ Save this for yourself so you don't lose it!
#GitHub #AI #DevTools #OpenSource #Coding #TechTrends
β¨ Join Best TG Channels https://t.me/addlist/0f6vfFbEMdAwODBk
βοΈ Join Our WhatsApp Channel https://whatsapp.com/channel/0029VaC7Weq29753hpcggW2A
Here's a fresh selection of projects that are experiencing significant growth and active discussion.
1. OpenAI Codex
An OpenAI coding agent for the terminal and development automation.
https://github.com/openai/codex
2. God's Eye View
OSINT and open geospatial data on an interactive 3D globe: aviation, satellites, and other sources.
https://github.com/bilawalsidhu/gods-eye-view
3. awesome-gpt-image-2
A large library of prompts, examples, and templates for GPT Image 2.
https://github.com/freestylefly/awesome-gpt-image-2
4. mattpocock/skills
Ready-made skills and workflows for coding agents that can be reused instead of using large prompts.
https://github.com/mattpocock/skills
5. Archify
Builds an interactive map of the codebase architecture: sequence, data flow, lifecycle, and component relationships.
https://github.com/tt-a1i/archify
6. Tailcat
A tool from Tailscale for directly connecting machines via WireGuard and NAT traversal.
https://github.com/tailscale/tailcat
7. Prime Agent
A self-improving coding agent for long, autonomous tasks.
https://github.com/PrimeIntellect-ai/prime-agent
8. Semantica
A graph-native infrastructure for storing context, relationships, and data provenance in AI systems.
https://github.com/semantica-agi/semantica
9. Cursor Plugins
The official repository of Cursor plugins: orchestration, code review, continual learning, and other agent functions.
https://github.com/cursor/plugins
10. AnyDoc
A tool from Firecrawl for converting documents into structured content for AI and RAG.
https://github.com/firecrawl/anydoc
π Looking at the overall trend for the week:
AI agents β agent skills β memory and context β visualization of architecture β tools for autonomous work with the environment
πΎ Save this for yourself so you don't lose it!
#GitHub #AI #DevTools #OpenSource #Coding #TechTrends
β¨ Join Best TG Channels https://t.me/addlist/0f6vfFbEMdAwODBk
βοΈ Join Our WhatsApp Channel https://whatsapp.com/channel/0029VaC7Weq29753hpcggW2A
GitHub
GitHub - openai/codex: Lightweight coding agent that runs in your terminal
Lightweight coding agent that runs in your terminal - openai/codex
β€2