m0unpredictable's profile picture. Computer Engineer By Profession .
Car Mechanic By Hobby .
I RT 🔁 interesting 💡 posts 📎

Vishal Patel

@m0unpredictable

Computer Engineer By Profession . Car Mechanic By Hobby . I RT 🔁 interesting 💡 posts 📎

Vishal Patel reposted

I built Multi-Head Attention in Excel because I wanted to understand how it works. It helped me — now it can help you too. 🔽 Download: byhand.ai/mha


Vishal Patel reposted

Self-Attention by hand ✍️ Excel ~ I designed this exercise for students to practice the QKV math. I also created a medium and a large version to show how the attention matrix grows quadratically as the sequence gets longer. 👇Join the 'AI Math' community. Download xlsx.


Vishal Patel reposted

Missed by many experts yesterday was the astonishing power of Google Gemini in research. Gemini allows scientists to scan through 200,000 papers, find ~250 relevant ones and extract data from those papers. Very powerful element to Gemini.


Vishal Patel reposted

Time Series Decomposition

hamptonism's tweet image. Time Series Decomposition

I traveled 1,593 kilometers with my Fitbit, and just earned the New Zealand badge for it! fitbit.com/user/BWV5XQ #Fitstats_AU


Vishal Patel reposted

xLSTM: Extended Long Short-Term Memory “performs favorably when compared to state-of-the-art Transformers and State Space Models, both in performance and scaling.” LSTM is not dead! Looking forward to see the comeback of RNNs🔥 github.com/NX-AI/xlstm arxiv.org/abs/2405.04517

hardmaru's tweet image. xLSTM: Extended Long Short-Term Memory

“performs favorably when compared to  state-of-the-art Transformers and State Space Models, both in  performance and scaling.”

LSTM is not dead! Looking forward to see the comeback of RNNs🔥

github.com/NX-AI/xlstm
arxiv.org/abs/2405.04517

Vishal Patel reposted

Activation of Neural Networks.


Vishal Patel reposted

Kolmogorov-Arnold Network is just an ordinary MLP. Here is the Colab, which explains: colab.research.google.com/drive/1v3AHz5J… The main point is, that if we consider KAN interaction as a piece-wise linear function, it can be rewritten like this: 1/n

bozavlado's tweet image. Kolmogorov-Arnold Network is just an ordinary MLP.
Here is the Colab, which explains:
colab.research.google.com/drive/1v3AHz5J…

The main point is, that if we consider KAN interaction as a piece-wise linear function, it can be rewritten like this:

1/n

Do you remember when you joined X? I do! #MyXAnniversary

m0unpredictable's tweet image. Do you remember when you joined X? I do! #MyXAnniversary

I traveled 563 kilometers with my Fitbit, and just earned the Hawaii badge for it! fitbit.com/user/BWV5XQ #Fitstats_AU


I traveled 402 kilometers with my Fitbit, and just earned the London Underground badge for it! fitbit.com/user/BWV5XQ #Fitstats_AU


Vishal Patel reposted

• Study hard. • What others think of you is none of your business. • It's OK not to have all the answers. • Experiment, Fail, Learn and Repeat. • Knowledge comes from experience. • Imagination is important. • Do what interests you the most. • Stay curious

ProfFeynman's tweet image. • Study hard.
• What others think of you is none of your business.
• It's OK not to have all the answers.
• Experiment, Fail, Learn and Repeat.
• Knowledge comes from experience.
• Imagination is important.
• Do what interests you the most.
• Stay curious

Vishal Patel reposted

Now you can render Iron Man in your favorite poses but much faster 🏎 We now support T2I adapters in 🧨 diffusers 🔥 T2I adapters are lightweight auxiliary networks & run ONLY once for the entire diffusion process giving ~ControlNet-like quality. Docs 📝huggingface.co/docs/diffusers…

RisingSayak's tweet image. Now you can render Iron Man in your favorite poses but much faster 🏎

We now support T2I adapters in 🧨 diffusers 🔥

T2I adapters are lightweight auxiliary networks & run ONLY once for the entire diffusion process giving ~ControlNet-like quality.

Docs 📝huggingface.co/docs/diffusers…
RisingSayak's tweet image. Now you can render Iron Man in your favorite poses but much faster 🏎

We now support T2I adapters in 🧨 diffusers 🔥

T2I adapters are lightweight auxiliary networks & run ONLY once for the entire diffusion process giving ~ControlNet-like quality.

Docs 📝huggingface.co/docs/diffusers…

Vishal Patel reposted

Derivatives (VIII): Second Derivatives bit.ly/2NUeiEK (The Mechanical Universe) #math #science #iteachmath #mtbos #visualization #elearning #calculus


Vishal Patel reposted

Integration (III): Area under a Function bit.ly/2JDMG18 (The Mechanical Universe) #math #science #iteachmath #mtbos #visualization #elearning #calculus


Vishal Patel reposted

The @arduino Opta explained in under 45 seconds. If you want to know more about Arduino's #IoT #PLC for industrial #automation, including watching an interview, follow this link: electromaker.io/blog/article/e…


Vishal Patel reposted

#Linux meme based upon @LoadingArtist comic ...

nixcraft's tweet image. #Linux meme based upon @LoadingArtist comic ...

Vishal Patel reposted

You can download a FREE Deep Learning book from this page: Understanding Deep Learning: udlbook.github.io/udbook It's a draft, and the book is almost finished, so it may not stay free for much longer.

abacusai's tweet image. You can download a FREE Deep Learning book from this page:

Understanding Deep Learning: udlbook.github.io/udbook

It's a draft, and the book is almost finished, so it may not stay free for much longer.

Loading...

Something went wrong.


Something went wrong.