SQuirreL

399 posts

SQuirreL banner
SQuirreL

SQuirreL

@IndianInExile

I give as good as I get

Katılım Mayıs 2020
14 Takip Edilen3 Takipçiler
SQuirreL
SQuirreL@IndianInExile·
@TensorTonic does it also covers models like lgbm & XGboost?
English
0
0
0
45
TensorTonic
TensorTonic@TensorTonic·
All the math you need to understand Machine Learning: Linear Algebra > Vectors, dot products, norms > Matrix multiplication and transpose > Eigenvalues and eigenvectors > Singular Value Decomposition (SVD) > Matrix inverses and pseudo-inverses Every dataset is a matrix. Every model transformation is a matrix operation. When you implement PCA from scratch, you stop fearing linear algebra forever. Read from here: tensortonic.com/ml-math Probability and Statistics > Bayes' theorem > Distributions: Gaussian, Bernoulli, Multinomial > Expectation, variance, covariance > Maximum Likelihood Estimation > MAP estimation > Conditional probability Every ML model is making a probabilistic bet. Naive Bayes, GMMs, VAEs: all just probability applied differently. Read here: tensortonic.com/ml-math Calculus > Derivatives and partial derivatives > Chain rule (this is literally backpropagation) > Gradients and Jacobians > Hessians (second-order intuition, not computation) > Multivariate function optimization You need to understand what a gradient means geometrically: which direction makes the loss go down. Read here: tensortonic.com/ml-math Optimization > Gradient descent and its variants (SGD, Adam, RMSProp) > Learning rates and convergence > Convexity (and why non-convex still works) > Loss functions: MSE, cross-entropy, hinge loss > Regularization: L1, L2, and why they work geometrically This is where math becomes code. Read here: tensortonic.com/ml-math Information Theory > Entropy > Cross-entropy (yes, your loss function) > KL divergence > Mutual information Most people skip this. Don't. Read here: tensortonic.com/ml-math That's it. These five areas, practiced through real implementations, cover 95% of the math behind every ML model from linear regression to transformers. That's what we built TensorTonic for. 200+ problems where you implement the math from scratch, with interactive visualizations that show you what every operation does geometrically. Link - tensortonic.com
TensorTonic tweet media
English
5
102
799
26.4K
SQuirreL retweetledi
Jay 🇳🇱
Jay 🇳🇱@Griezmenace·
A civilization of scholars without soldiers is just a library waiting to be burned.
English
56
1.2K
5.3K
134.7K
SQuirreL
SQuirreL@IndianInExile·
WHY TF AADHAR OTP HAS TO COME ONLY ON PHONE NO AND NOT EMAIL ??? IM NOT RECHARGING FOR A WHOLE MONTH JUST FOR AN OTP FFS
English
0
0
0
2
SQuirreL
SQuirreL@IndianInExile·
@stellaoki Damn so murder 3 was copied from here
English
1
0
10
20.4K
egbidi
egbidi@stellaoki·
Now I'm curious
English
179
1.1K
16.2K
445.8K
SQuirreL
SQuirreL@IndianInExile·
@sidhant Why does this UN thing even happens? Do they actually serve any purpose? Same speech same lines just from a different seat lol
English
0
0
1
583
Sidhant Sibal
Sidhant Sibal@sidhant·
Pakistan lives in la la land, Indian Diplomat Anupama Singh's sharp rebuke to Pakistan at UN Human Rights Council; Points to J&K's development budget being bigger than Pakistan's IMF bailout
English
66
1.3K
8.7K
166.8K
SQuirreL
SQuirreL@IndianInExile·
@InvestorOfJAMMU There are so many accounts impersonating you, even copying and posting your tweet word to word lol
SQuirreL tweet media
English
0
0
0
57
Margin of Safety🇮🇳
Margin of Safety🇮🇳@InvestorOfJAMMU·
You will be surprised to know that Nifty Capital Market index grown its Q3FY26 PAT by 65% YoY against Nifty muted growth of just 1%. Supersonic growth lead by MCX & BSE. Do you think this growth is sustainable??
English
15
4
143
9K
SQuirreL
SQuirreL@IndianInExile·
@p_chauhan1406 @ankurpandeyvns @MeghUpdates bro why would you do any such stuffs on anyone's property that too a religious thing? he knew exactly what he was doing assault could have been avoided but laaton k bhoot baaton se ni maante maybe you try do something similar on a waqf property & then we can talk about reactions
English
0
0
0
14
Prakhar Chauhan
Prakhar Chauhan@p_chauhan1406·
@ankurpandeyvns @MeghUpdates First thing, it wasn’t a temple. It is someone else’s property with no fence or gate. I won’t pray in any mosque but in a temple Even if someone mistakenly does it which is very rare, it would be better to deal with it peacefully.
English
3
0
0
100
Megh Updates 🚨™
Megh Updates 🚨™@MeghUpdates·
🚨 Rudrapur, Uttarakhand: A Muslim offered namaz on temple grounds. Ataria temple manager Arvind Sharma allegedly assaulted a fasting man during prayers and forced him to chant “Jai Shri Ram,” later apologizing before letting him go.
English
376
1.2K
7.4K
235.5K
SQuirreL
SQuirreL@IndianInExile·
Its not the AI bubble, its the twitter bubble
English
0
0
0
8
SQuirreL
SQuirreL@IndianInExile·
@talk2anuradha you are saying "a section of under-construction Metro bridge" and "No construction was going on" in the same post?
English
0
0
0
23
Anuradha Tiwari
Anuradha Tiwari@talk2anuradha·
This country is really unlivable now. Today, a section of under-construction Metro bridge collapsed onto vehicles in Mulund, Mumbai. 1 person killed & 3 injured. No construction was going on. It was a finished section. Imagine the corruption in broad daylight!
English
288
2.2K
7.6K
174.2K
SQuirreL
SQuirreL@IndianInExile·
@ajeetbharti they can save so much money from the paper which they can save by reducing the total marks to 360 for the reserved class coz no one is scoring above 300 anyways😂
English
0
0
0
62
Ajeet Bharti
Ajeet Bharti@ajeetbharti·
Putting without comments. Revised cutoff notified on Jan 9.
Ajeet Bharti tweet media
English
290
2.9K
7.6K
156.2K
Vandhitha
Vandhitha@Vandhitha31·
What do you think valuation actually means — hype, math, or future potential? 🤔
English
1
0
1
78
SQuirreL
SQuirreL@IndianInExile·
@Mayawati Which timeline am I living in lol where BJP brings UGC and Mayawati is defending GCs😂
English
0
0
0
6
Mayawati
Mayawati@Mayawati·
यह बड़े दुख व चिन्ता की बात है कि पिछले कुछ समय से अकेले यू.पी. में ही नहीं बल्कि अब तो फिल्मों में भी ’पंडत’ को घूसखोर आदि बताकर पूरे देश में जो इनका अपमान व अनादर किया जा रहा है तथा जिससे समूचे ब्राह्मण समाज में इस समय ज़बरदस्त रोष व्याप्त है, इसकी हमारी पार्टी भी कड़े शब्दों में निन्दा करती है। ऐसी इस जातिसूचक फिल्म (वेब सीरीज) ’घूसखोर पंडत’ पर केन्द्र सरकार को तुरन्त प्रतिबन्ध लगाना चाहिये, बी.एस.पी. की यह माँग। साथ ही, इसको लेकर लखनऊ पुलिस द्वारा प्राथमिकी दर्ज करना उचित कदम।
हिन्दी
2.1K
11.1K
37.1K
1.1M
SQuirreL
SQuirreL@IndianInExile·
@gdb Whats the difference between CODEX and CHATGPT AGENT ? Are they same? If not, how different are they in terms of their function?
English
0
0
0
56
Greg Brockman
Greg Brockman@gdb·
Software development is undergoing a renaissance in front of our eyes. If you haven't used the tools recently, you likely are underestimating what you're missing. Since December, there's been a step function improvement in what tools like Codex can do. Some great engineers at OpenAI yesterday told me that their job has fundamentally changed since December. Prior to then, they could use Codex for unit tests; now it writes essentially all the code and does a great deal of their operations and debugging. Not everyone has yet made that leap, but it's usually because of factors besides the capability of the model. Every company faces the same opportunity now, and navigating it well — just like with cloud computing or the Internet — requires careful thought. This post shares how OpenAI is currently approaching retooling our teams towards agentic software development. We're still learning and iterating, but here's how we're thinking about it right now: As a first step, by March 31st, we're aiming that: (1) For any technical task, the tool of first resort for humans is interacting with an agent rather than using an editor or terminal. (2) The default way humans utilize agents is explicitly evaluated as safe, but also productive enough that most workflows do not need additional permissions. In order to get there, here's what we recommended to the team a few weeks ago: 1. Take the time to try out the tools. The tools do sell themselves — many people have had amazing experiences with 5.2 in Codex, after having churned from codex web a few months ago. But many people are also so busy they haven't had a chance to try Codex yet or got stuck thinking "is there any way it could do X" rather than just trying. - Designate an "agents captain" for your team — the primary person responsible for thinking about how agents can be brought into the teams' workflow. - Share experiences or questions in a few designated internal channels - Take a day for a company-wide Codex hackathon 2. Create skills and AGENTS[.md]. - Create and maintain an AGENTS[.md] for any project you work on; update the AGENTS[.md] whenever the agent does something wrong or struggles with a task. - Write skills for anything that you get Codex to do, and commit it to the skills directory in a shared repository 3. Inventory and make accessible any internal tools. - Maintain a list of tools that your team relies on, and make sure someone takes point on making it agent-accessible (such as via a CLI or MCP server). 4. Structure codebases to be agent-first. With the models changing so fast, this is still somewhat untrodden ground, and will require some exploration. - Write tests which are quick to run, and create high-quality interfaces between components. 5. Say no to slop. Managing AI generated code at scale is an emerging problem, and will require new processes and conventions to keep code quality high - Ensure that some human is accountable for any code that gets merged. As a code reviewer, maintain at least the same bar as you would for human-written code, and make sure the author understands what they're submitting. 6. Work on basic infra. There's a lot of room for everyone to build basic infrastructure, which can be guided by internal user feedback. The core tools are getting a lot better and more usable, but there's a lot of infrastructure that currently go around the tools, such as observability, tracking not just the committed code but the agent trajectories that led to them, and central management of the tools that agents are able to use. Overall, adopting tools like Codex is not just a technical but also a deep cultural change, with a lot of downstream implications to figure out. We encourage every manager to drive this with their team, and to think through other action items — for example, per item 5 above, what else can prevent a lot of "functionally-correct but poorly-maintainable code" from creeping into codebases.
English
413
1.6K
12.3K
2.1M
Grok
Grok@grok·
@IndianInExile @the_fauxy The female anchor in the video looks like Palki Sharma Upadhyay, a well-known Indian journalist from Firstpost and formerly WION. However, since it's from a satire account like The Fauxy, it's likely an AI-generated or edited parody.
English
2
0
0
89
The Fauxy
The Fauxy@the_fauxy·
🚨#𝗕𝗥𝗘𝗔𝗞𝗜𝗡𝗚: “𝗦𝗮𝗯𝗸𝗮 𝗦𝗮𝗮𝘁𝗵, 𝗦𝗮𝗯𝗸𝗮 𝗩𝗶𝗸𝗮𝘀” 𝗚𝗼𝗲𝘀 𝗚𝗹𝗼𝗯𝗮𝗹 𝗮𝘀 𝗠𝗼𝗱𝗶 𝗔𝗴𝗿𝗲𝗲𝘀 𝘁𝗼 𝗗𝗲𝘃𝗲𝗹𝗼𝗽 𝗔𝗺𝗲𝗿𝗶𝗰𝗮 𝗙𝘂𝗿𝘁𝗵𝗲𝗿
CY
25
377
1.5K
36.9K
SQuirreL
SQuirreL@IndianInExile·
@ajeetbharti ॐ शांति ॐ सद्गति 🙏🏻
हिन्दी
0
0
0
1
Ajeet Bharti
Ajeet Bharti@ajeetbharti·
नमस्कार मित्रो! आज प्रातःकाल 4:30 पर माताजी ने देह त्याग दिया। आप सभी की प्रार्थनाओं ने उसे बहुत संबल दिया, जिजीविषा दी परंतु भगवान की इच्छा कुछ और थी। आप सब मेरे परिजन की भाँति इस पूरे समय साथ खड़े रहे, उसका आभारी हूँ। मेरी माँ, हर माँ की तरह, मेरे जीवन का आधार थी। उसी की कृपा ने मुझे शिक्षा की ओर मोड़ा, उसी ने पिताजी से जिद कर-कर के मुझे दिल्ली तक पढ़ाने को भेजा। मैं चाहता था कि वो लम्बे समय तक मेरे साथ रहें, परंतु लिवर सिरोसिस जैसे रोग लगभग असाध्य ही होते हैं। अंतिम दिनों में कुछ कष्ट में थीं, और वह मेरे लिए असह्य अनुभव था। आप सब ने इस कठिन समय में मेरा सहयोग किया, मेरी माँ के लिए प्रार्थनाएँ कीं, डॉक्टर और हॉस्पिटल बताए, मैं इसे भूल नहीं सकता। अब उनकी सद्गति हेतु प्रार्थना कीजिए! ईश्वर उन्हें मुक्ति दें! ॐ शांति!
हिन्दी
17.7K
5K
39K
1.2M
The Data of Everything
The Data of Everything@TheDataHubX·
Top 50 Greatest Anime Series of All Time 🎌🔥 1. 🇯🇵 Attack on Titan 2. 🇯🇵 Death Note 3. 🇯🇵 Fullmetal Alchemist: Brotherhood 4. 🇯🇵 Code Geass 5. 🇯🇵 One Piece 6. 🇯🇵 Naruto: Shippuden 7. 🇯🇵 One Punch Man 8. 🇯🇵 Dragon Ball 9. 🇯🇵 No Game No Life 10. 🇯🇵 Food Wars: Shokugeki no Soma 11. 🇯🇵 Your Lie in April 12. 🇯🇵 Demon Slayer: Kimetsu no Yaiba 13. 🇯🇵 My Hero Academia 14. 🇯🇵 Naruto 15. 🇯🇵 Hunter x Hunter 16. 🇯🇵 Steins;Gate 17. 🇯🇵 Cowboy Bebop 18. 🇯🇵 Neon Genesis Evangelion 19. 🇯🇵 Gintama 20. 🇯🇵 Haikyuu!! 21. 🇯🇵 Monster 22. 🇯🇵 Clannad 23. 🇯🇵 Mob Psycho 100 24. 🇯🇵 Tokyo Ghoul 25. 🇯🇵 Blue Exorcist 26. 🇯🇵 Black Clover 27. 🇯🇵 Fairy Tail 28. 🇯🇵 Death Parade 29. 🇯🇵 Psycho-Pass 30. 🇯🇵 Bleach 31. 🇯🇵 Sword Art Online 32. 🇯🇵 JoJo’s Bizarre Adventure 33. 🇯🇵 Re:Zero 34. 🇯🇵 Kill la Kill 35. 🇯🇵 Soul Eater 36. 🇯🇵 Dororo 37. 🇯🇵 Spirited Away 38. 🇯🇵 Your Name 39. 🇯🇵 The Promised Neverland 40. 🇯🇵 Assassination Classroom 41. 🇯🇵 That Time I Got Reincarnated as a Slime 42. 🇯🇵 Tokyo Revengers 43. 🇯🇵 Vinland Saga 44. 🇯🇵 Dr. Stone 45. 🇯🇵 Spy x Family 46. 🇯🇵 Samurai Champloo 47. 🇯🇵 Akame ga Kill! 48. 🇯🇵 Erased 49. 🇯🇵 Parasyte 50. 🇯🇵 Chainsaw Man Source: IMDb, MyAnimeList, Anime fans polls
The Data of Everything tweet media
English
755
1.4K
10.6K
1.1M
SQuirreL
SQuirreL@IndianInExile·
@heavensbvnny Whats the need to write it here? You can just write it to your partner privately 🤷🏻its perfomative action which has many synonyms like cheating and gaslighting
English
0
0
2
363
🤠
🤠@heavensbvnny·
i'm dating to marry, not to play around for a couple months and then break up, that's why i'm willing to fight for a relationship. you're crazy if you think I'm leaving after i’ve seen your insecurities, your flaws. you trusted me with a side of you that no one else saw. i would never change anything about you. your problems have become my problems. your happiness is a priority to me. i wanna know your struggles so i can help and love you the right way
English
45
469
2.5K
113.5K
SQuirreL
SQuirreL@IndianInExile·
@Tanu8839 @weirdest1910 Are you aarna's batchmate? I've a friend who's from IIT Mandi. I wanted to confirm this news, can you tell me which branch she was from and her full name? You can dm me if you can't reveal here
English
0
0
0
33