This media is not supported in your browser
VIEW IN TELEGRAM
Self Attention vs Cross Attention by hand โ๏ธ
Resize the matrices yourself ๐ https://byhand.ai/aMisxP
Two attention mechanisms, side by side. Both project X into queries; both compute attention via S = Kแต ร Q and F = V ร A. The only difference is the source of K and V.
Self attention uses X for everything. Q, K, and V all come from projecting X. Each X token attends to every other X token. The score matrix S is square โ 128 ร 128.
Cross attention uses X for queries and a second sequence E for keys and values. Each X token attends to every E token instead. The score matrix S is rectangular โ 64 ร 128.
Notice what's shared and what's not:
X is the same in both โ same 36 ร 128 input.
Q and K share the 16 dimension โ that's what makes the dot product Kแต ร Q valid in either case.
V dimensions are independent: self-attention uses 12, cross-attention uses 12. The choice doesn't depend on which mechanism you're using; it depends on what output dimension your downstream layer expects.
https://t.iss.one/CodeProgrammer
Resize the matrices yourself ๐ https://byhand.ai/aMisxP
Two attention mechanisms, side by side. Both project X into queries; both compute attention via S = Kแต ร Q and F = V ร A. The only difference is the source of K and V.
Self attention uses X for everything. Q, K, and V all come from projecting X. Each X token attends to every other X token. The score matrix S is square โ 128 ร 128.
Cross attention uses X for queries and a second sequence E for keys and values. Each X token attends to every E token instead. The score matrix S is rectangular โ 64 ร 128.
Notice what's shared and what's not:
X is the same in both โ same 36 ร 128 input.
Q and K share the 16 dimension โ that's what makes the dot product Kแต ร Q valid in either case.
V dimensions are independent: self-attention uses 12, cross-attention uses 12. The choice doesn't depend on which mechanism you're using; it depends on what output dimension your downstream layer expects.
https://t.iss.one/CodeProgrammer
โค4
Forwarded from Machine Learning with Python
Follow the Machine Learning with Python channel on WhatsApp: https://whatsapp.com/channel/0029VaC7Weq29753hpcggW2A
GitHub repositories to enhance your Python proficiency:
- Web development with Django โ https://github.com/django/django
- Data Science tools โ https://github.com/rasbt/python-machine-learning-book
- Algorithmic challenges โ https://github.com/TheAlgorithms/Python
- Machine learning recipes โ https://github.com/ageron/handson-ml2
- Testing best practices โ https://github.com/pytest-dev/pytest
- Automation scripts โ https://github.com/soimort/you-get
- Advanced Python concepts โ https://github.com/faif/python-patterns
Bookmark and share
https://t.iss.one/CodeProgrammer๐
- Web development with Django โ https://github.com/django/django
- Data Science tools โ https://github.com/rasbt/python-machine-learning-book
- Algorithmic challenges โ https://github.com/TheAlgorithms/Python
- Machine learning recipes โ https://github.com/ageron/handson-ml2
- Testing best practices โ https://github.com/pytest-dev/pytest
- Automation scripts โ https://github.com/soimort/you-get
- Advanced Python concepts โ https://github.com/faif/python-patterns
Bookmark and share
https://t.iss.one/CodeProgrammer
Please open Telegram to view this post
VIEW IN TELEGRAM
โค3๐ฅ2
Searched 35 free courses, so you don't have to! ๐โจ
Here are the 35 best free courses: ๐
1. Data Science: Machine Learning ๐ค
Link: https://lnkd.in/gUNVYgGB
2. Introduction to computer science ๐ป
Link: https://lnkd.in/gR66-htH
3. Introduction to programming with scratch ๐งฉ
Link: https://lnkd.in/gBDUf_Wx
4. Computer science for business professionals ๐ผ
Link: https://lnkd.in/g8gQ6N-H
5. How to conduct and write a literature review ๐
Link: https://lnkd.in/gsh63GET
6. Software Construction ๐
Link: https://lnkd.in/ghtwpNFJ
7. Machine Learning with Python: from linear models to deep learning ๐๐ง
Link: https://lnkd.in/g_T7tAdm
8. Startup Success: How to launch a technology company in 6 steps ๐
Link: https://lnkd.in/gN3-_Utz
9. Data analysis: statistical modeling and computation in applications ๐
Link: https://lnkd.in/gCeihcZN
10. The art and science of searching in systematic reviews ๐
Link: https://lnkd.in/giFW5q4y
11. Introduction to conducting systematic review ๐
Link: https://lnkd.in/g6EEgCkW
12. Introduction to computer science and programming using python ๐ฅ
Link: https://lnkd.in/gwhMpWck
13. Introduction to computational thinking and data science ๐ก
Link: https://lnkd.in/gfjuDp5y
14. Becoming an Entrepreneur ๐ธ
Link: https://lnkd.in/gqkYmVAW
15. High-dimensional data analysis ๐
Link: https://lnkd.in/gv9RV9Zc
16. Statistics and R ๐
Link: https://lnkd.in/gUY3jd8v
17. Conduct a literature review ๐
Link: https://lnkd.in/g4au3w2j
18. Systematic Literature Review: An Introduction ๐ง
Link: https://lnkd.in/gVwGAzzY
19. Introduction to systematic review and meta-analysis ๐งฎ
Link: https://lnkd.in/gnpN9ivf
20. Creating a systematic literature review โ๏ธ
Link: https://lnkd.in/gbevCuy6
21. Systematic reviews and meta-analysis ๐
Link: https://lnkd.in/ggnNeX5j
22. Research methodologies ๐ต๏ธโโ๏ธ
Link: https://lnkd.in/gqh3VKCC
23. Quantitative and Qualitative research for beginners ๐๐ฌ
Link: https://shorturl.at/uNT58
24. Writing case studies: science of delivery ๐
Link: https://shorturl.at/ejnMY
25. research methodology: complete research project blueprint ๐บ
Link: https://lnkd.in/gFU8Nbrv
26. How to write a successful research paper ๐
Link: https://lnkd.in/g-ni3u5q
27. Research proposal bootcamp: how to write a research proposal ๐โโ๏ธ
Link: https://lnkd.in/gNRitBwX
28. Understanding technology ๐ฑ
Link: https://lnkd.in/gfjUnHfd
29. Introduction to artificial intelligence with Python ๐ค๐
Link: https://lnkd.in/gygaeAcY
30. Introduction to programming with Python ๐ป
Link: https://lnkd.in/gAdyf6xR
31. Web programming with Python and JavaScript ๐
Link: https://lnkd.in/g_i5-SeG
32. Understanding Research methods ๐ฌ
Link: https://lnkd.in/g-xBFj4v
33. How to write and publish a scientific paper ๐ข
Link: https://lnkd.in/giwTe2is
34. Introduction to systematic review and meta-analysis ๐
Link: https://lnkd.in/gnpN9ivf
35. Research for impact ๐
Link: https://lnkd.in/gRsWsUsq
Here are the 35 best free courses: ๐
1. Data Science: Machine Learning ๐ค
Link: https://lnkd.in/gUNVYgGB
2. Introduction to computer science ๐ป
Link: https://lnkd.in/gR66-htH
3. Introduction to programming with scratch ๐งฉ
Link: https://lnkd.in/gBDUf_Wx
4. Computer science for business professionals ๐ผ
Link: https://lnkd.in/g8gQ6N-H
5. How to conduct and write a literature review ๐
Link: https://lnkd.in/gsh63GET
6. Software Construction ๐
Link: https://lnkd.in/ghtwpNFJ
7. Machine Learning with Python: from linear models to deep learning ๐๐ง
Link: https://lnkd.in/g_T7tAdm
8. Startup Success: How to launch a technology company in 6 steps ๐
Link: https://lnkd.in/gN3-_Utz
9. Data analysis: statistical modeling and computation in applications ๐
Link: https://lnkd.in/gCeihcZN
10. The art and science of searching in systematic reviews ๐
Link: https://lnkd.in/giFW5q4y
11. Introduction to conducting systematic review ๐
Link: https://lnkd.in/g6EEgCkW
12. Introduction to computer science and programming using python ๐ฅ
Link: https://lnkd.in/gwhMpWck
13. Introduction to computational thinking and data science ๐ก
Link: https://lnkd.in/gfjuDp5y
14. Becoming an Entrepreneur ๐ธ
Link: https://lnkd.in/gqkYmVAW
15. High-dimensional data analysis ๐
Link: https://lnkd.in/gv9RV9Zc
16. Statistics and R ๐
Link: https://lnkd.in/gUY3jd8v
17. Conduct a literature review ๐
Link: https://lnkd.in/g4au3w2j
18. Systematic Literature Review: An Introduction ๐ง
Link: https://lnkd.in/gVwGAzzY
19. Introduction to systematic review and meta-analysis ๐งฎ
Link: https://lnkd.in/gnpN9ivf
20. Creating a systematic literature review โ๏ธ
Link: https://lnkd.in/gbevCuy6
21. Systematic reviews and meta-analysis ๐
Link: https://lnkd.in/ggnNeX5j
22. Research methodologies ๐ต๏ธโโ๏ธ
Link: https://lnkd.in/gqh3VKCC
23. Quantitative and Qualitative research for beginners ๐๐ฌ
Link: https://shorturl.at/uNT58
24. Writing case studies: science of delivery ๐
Link: https://shorturl.at/ejnMY
25. research methodology: complete research project blueprint ๐บ
Link: https://lnkd.in/gFU8Nbrv
26. How to write a successful research paper ๐
Link: https://lnkd.in/g-ni3u5q
27. Research proposal bootcamp: how to write a research proposal ๐โโ๏ธ
Link: https://lnkd.in/gNRitBwX
28. Understanding technology ๐ฑ
Link: https://lnkd.in/gfjUnHfd
29. Introduction to artificial intelligence with Python ๐ค๐
Link: https://lnkd.in/gygaeAcY
30. Introduction to programming with Python ๐ป
Link: https://lnkd.in/gAdyf6xR
31. Web programming with Python and JavaScript ๐
Link: https://lnkd.in/g_i5-SeG
32. Understanding Research methods ๐ฌ
Link: https://lnkd.in/g-xBFj4v
33. How to write and publish a scientific paper ๐ข
Link: https://lnkd.in/giwTe2is
34. Introduction to systematic review and meta-analysis ๐
Link: https://lnkd.in/gnpN9ivf
35. Research for impact ๐
Link: https://lnkd.in/gRsWsUsq
โค5
Register for the FREE Python Demo Session!
๐ Date: 30 April 2026
โฐ Time: 7:30 PM
๐ Zoom Link: https://us06web.zoom.us/meeting/register/HSOTmzzpTkGIGm9C9oGbaA
Everyone is welcome!
https://t.iss.one/CodeProgrammer
๐ Date: 30 April 2026
โฐ Time: 7:30 PM
๐ Zoom Link: https://us06web.zoom.us/meeting/register/HSOTmzzpTkGIGm9C9oGbaA
Everyone is welcome!
https://t.iss.one/CodeProgrammer
โค6
Softmax vs Hardmax by hand โ๏ธ ~ interactive calculator ๐ https://byhand.ai/vhUJDH
Softmax turns a set of raw scores (z) into a probability distribution (Y) over choices (a, b, c, d, e). Instead of just saying which option is best, it tells us how likely each option is to be chosen. In this example, most of the probability mass is concentrated on c, while the other options are still possible but clearly less likely. That's the point of softmax: it converts relative scores into meaningful, comparable probabilities that sum to 100%.
Think of a raffle. Hardmax is when the person who bought the most tickets always wins the prize โ the top score takes it, every time. Softmax is when everyone's chance is proportional to the tickets they hold: even if I bought just one ticket, I may still get lucky. Who knows. That's the psychology of softmax.
This is how a language model chooses its next word. Each time a word appears in the training data, it earns a ticket. Hardmax would always speak the word with the most tickets โ the same safe choice, over and over. Softmax gives every word a chance proportional to its tickets, so less common words can still be spoken. The word with the most tickets still has the highest chance of winning โ just not 100%. That's what lets the model surprise us with its creativity (and also its hallucinations) instead of repeating itself.
https://t.iss.one/CodeProgrammer๐ฑ
Softmax turns a set of raw scores (z) into a probability distribution (Y) over choices (a, b, c, d, e). Instead of just saying which option is best, it tells us how likely each option is to be chosen. In this example, most of the probability mass is concentrated on c, while the other options are still possible but clearly less likely. That's the point of softmax: it converts relative scores into meaningful, comparable probabilities that sum to 100%.
Think of a raffle. Hardmax is when the person who bought the most tickets always wins the prize โ the top score takes it, every time. Softmax is when everyone's chance is proportional to the tickets they hold: even if I bought just one ticket, I may still get lucky. Who knows. That's the psychology of softmax.
This is how a language model chooses its next word. Each time a word appears in the training data, it earns a ticket. Hardmax would always speak the word with the most tickets โ the same safe choice, over and over. Softmax gives every word a chance proportional to its tickets, so less common words can still be spoken. The word with the most tickets still has the highest chance of winning โ just not 100%. That's what lets the model surprise us with its creativity (and also its hallucinations) instead of repeating itself.
https://t.iss.one/CodeProgrammer
Please open Telegram to view this post
VIEW IN TELEGRAM
โค5
๐ก Level Up Your IT Career in 2026 โ For FREE
Areas covered: #Python #AI #Cisco #PMP #Fortinet #AWS #Azure #Excel #CompTIA #ITIL #Cloud + more
๐ Download each free resource here:
โข Free Courses (Python, Excel, Cyber Security, Cisco, SQL, ITIL, PMP, AWS)
๐https://bit.ly/4ejSFbz
โข IT Certs E-book
๐ https://bit.ly/42y8owh
โข IT Exams Skill Test
๐ https://bit.ly/42kp7Dv
โข Free AI Materials & Support Tools
๐ https://bit.ly/3QEfWek
โข Free Cloud Study Guide
๐https://bit.ly/4u8Zb9r
๐ฒ Need exam help? Contact admin: wa.link/40f942
๐ฌ Join our study group (free tips & support): https://chat.whatsapp.com/K3n7OYEXgT1CHGylN6fM5a
Areas covered: #Python #AI #Cisco #PMP #Fortinet #AWS #Azure #Excel #CompTIA #ITIL #Cloud + more
๐ Download each free resource here:
โข Free Courses (Python, Excel, Cyber Security, Cisco, SQL, ITIL, PMP, AWS)
๐https://bit.ly/4ejSFbz
โข IT Certs E-book
๐ https://bit.ly/42y8owh
โข IT Exams Skill Test
๐ https://bit.ly/42kp7Dv
โข Free AI Materials & Support Tools
๐ https://bit.ly/3QEfWek
โข Free Cloud Study Guide
๐https://bit.ly/4u8Zb9r
๐ฒ Need exam help? Contact admin: wa.link/40f942
๐ฌ Join our study group (free tips & support): https://chat.whatsapp.com/K3n7OYEXgT1CHGylN6fM5a
โค4
A good selection for those who want to improve their skills in practice, rather than just reading theory:
tags: #ML #DataScience #DataAnalysis
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM
โค2๐ฏ1
Awesome DataScience โ a structured list of open-source data, datasets, libraries, and tutorials for solving real-world problems. ๐ ๏ธ
It's useful for both beginners and those already familiar with the field โ you'll find something new here.
โ๏ธ Link to GitHub: https://github.com/academic/awesome-datascience
tags: #DataScientist
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM
๐งฎ $40/day ร 30 days = $1,200/month.
That's what my students average.
From their phone. In 10 minutes a day.
No degree needed.
No investment knowledge required.
Just Copy & Paste my moves.
I'm Tania, and this is real.
๐ Join for Free, Click here
#ad๐ข InsideAd
That's what my students average.
From their phone. In 10 minutes a day.
No degree needed.
No investment knowledge required.
Just Copy & Paste my moves.
I'm Tania, and this is real.
๐ Join for Free, Click here
#ad
Please open Telegram to view this post
VIEW IN TELEGRAM