A presentation at BreizhCamp in in Rennes, France by Jean-Luc Tromparent
mojo 🔥 A NEW Hope for ai ?
mojo 🔥 The next ai language ?
CONCLUSION Use Case Value Proposition AI in 2024
What is The current language of AI ?
c3-GPTo Advisor L’IA, ou intelligence artificielle, peut être développée et programmée dans différents langages de programmation. Certains des langages les plus couramment utilisés pour créer des systèmes d’IA incluent Python, Java, C++, et R, entre autres. Python est particulièrement populaire dans le domaine de l’IA en raison de sa simplicité, de sa flexibilité, de sa large gamme de bibliothèques et de frameworks dédiés à l’IA (comme TensorFlow, PyTorch, scikit-learn, etc.), et de sa communauté active de développeurs.
StackOverflow Advisor
StackOverflow Advisor
StackOverflow Advisor
StackOverflow Advisor
StackOverflow Advisor
How To Choose a Framework ? https://www.youtube.com/watch?v=k4Tfg6-7cyQ
AI Programming Landscape Model System Hardware CUDA, OpenCL, ROCm
NEW Kid in ToWN ! Mojo 🔥 02/05/2023
Mojo is born ! https://www.modular.com/blog/the-future-of-ai-depends-on-modularity
Mojo is born ! Modular Accelerated eXecution platform https://www.modular.com/blog/a-unified-extensible-platform-to-superpower-your-ai
Mojo is born ! • Member of the python family (superset of python) • Support modern chip architectures (thanks to MLIR) • Predictable low level performance https://www.modular.com/blog/a-unified-extensible-platform-to-superpower-your-ai
Mojo is born ! Chris Lattner 2000 beginning of the project LLVM 2003 release of LLVM 1.0 2007 release of CLang 1.0 2008 XCode 3.1 2011 Clang replace gcc on macos 2014 release of Swift 1.0 2018 beginning of the MLIR 2022 creation of Modular cie 2023 🔥 https://www.nondot.org/sabre/
Mojo is blazing fast ! https://www.modular.com/blog/how-mojo-gets-a-35-000x-speedup-over-python-part-1
Mojo is blazing fast ! Changelog 2022/01 incorporation 2022/07 seed round (30 M$) 2023/05 announce MAX & Mojo 2023/08 serie B (100 M$) 2023/09 release mojo 0.2.1 2023/10 release mojo 0.4.0 .. 2024/01 release mojo 0.7.0 2024/02 release MAX & mojo 24.1 https://www.modular.com/blog/how-mojo-gets-a-35-000x-speedup-over-python-part-1 2024/06 release MAX & mojo 24.4
Mojo is blazing fast ! https://www.modular.com/blog/how-mojo-gets-a-35-000x-speedup-over-python-part-1
Performance matters ! Performance matters : • for our users
Performance matters ! Your resume is being processed
Performance matters ! Performance matters : • for our users • for (artificial) intelligence
Performance matters !
Performance matters ! Performance matters : • for our users • for (artificial) intelligence • for the planet
Performance matters ! https://haslab.github.io/SAFER/scp21.pdf
Meetup python-rennes https://www.meetup.com/fr-FR/python-rennes/ https://www.youtube.com/watch?v=gE6HUsmh554
Performance matters ! Performance matters : • for our users • for (artificial) intelligence • for the planet
it’s demo time ! Laplacian filter (edge detection)
Edge Detection
Edge Detection kernel Convolve
Hop hop hop 2D Convolution Animation — M ichael Plotke, CC BY-SA 3.0 via W ikimedia Commons
Edge Detection 2D Convolution Animation — M ichael Plotke, CC BY-SA 3.0 via W ikimedia Commons
Edge Detection 2D Convolution Animation — M ichael Plotke, CC BY-SA 3.0 via W ikimedia Commons
naïve version : 500 ms numpy mul : 250 ms Recap numpy+numba : 50 ms opencv : 0.5 ms x2 x 10 x 1000 And now in mojo ?
And now in mojo ! https://www.modular.com/blog/implementing-numpy-style-matrix-slicing-in-mojo
Power ! Unlimited Power ! Credits to Georges L. and Corentin R.
“The greatest teacher, failure is”
it’s demo time ! Let’s optimize !
SISD Architecture
SIMD Architecture
Algorithm vectorization fn naive(img: Matrix[DType.float32], kernel: Matrix[DType.float32]) -> Matrix[DType.float32]: var result = Matrix[DType.float32](img.height, img.width) # Loop through each pixel in the image # But skip the outer edges of the image for y in range(1, img.height-1): for x in range(1, img.width-1): # For each pixel, compute the product elements wise var acc: Float32 = 0 for k in range(3) : for l in range(3): acc += img[y-1+k, x-1+l] * kernel[k, l] # Normalize the result result[y, x] = min(255, max(0, acc)) return result
Algorithm vectorization alias nelts = simdwidthofDType.float32 x fn naive(img: Matrix[DType.float32], kernel: Matrix[DType.float32]) -> Matrix[DType.float32]: var result = Matrix[DType.float32](img.height, img.width) # Loop through each pixel in the image # But skip the outer edges of the image for y in range(1, img.height-1): for x in range(1, img.width-1): # For each pixel, compute the product elements wise var acc: Float32 = 0 for k in range(3) : for l in range(3): acc += img[y-1+k, x-1+l] * kernel[k, l] # Normalize the result result[y, x] = min(255, max(0, acc)) return result
Algorithm vectorization alias nelts = simdwidthofDType.float32 fn naive(img: Matrix[DType.float32], kernel: Matrix[DType.float32]) -> Matrix[DType.float32]: var result = Matrix[DType.float32](img.height, img.width) # Loop through each pixel in the image # But skip the outer edges of the image for y in range(1, img.height-1): for x in range(1, img.width-1, nelts): # For each pixel, compute the product elements wise var acc: Float32 = 0 for k in range(3) : for l in range(3): acc += img[y-1+k, x-1+l] * kernel[k, l] # Normalize the result result[y, x] = min(255, max(0, acc)) return result
Algorithm vectorization alias nelts = simdwidthofDType.float32 fn naive(img: Matrix[DType.float32], kernel: Matrix[DType.float32]) -> Matrix[DType.float32]: var result = Matrix[DType.float32](img.height, img.width) # Loop through each pixel in the image # But skip the outer edges of the image for y in range(1, img.height-1): for x in range(1, img.width-1, nelts): # For each pixel, compute the product elements wise # var acc: Float32 = 0 var acc: SIMD[DType.float32,nelts] = 0 for k in range(3) : for l in range(3): acc += img[y-1+k, x-1+l] * kernel[k, l] # Normalize the result result[y, x] = min(255, max(0, acc)) return result
Algorithm vectorization fn naive(img: Matrix[DType.float32], kernel: Matrix[DType.float32]) -> Matrix[DType.float32]: var result = Matrix[DType.float32](img.height, img.width) # Loop through each pixel in the image # But skip the outer edges of the image for y in range(1, img.height-1): for x in range(1, img.width-1, nelts): # For each pixel, compute the product elements wise # var acc: Float32 = 0 var acc: SIMD[DType.float32,nelts] = 0 for k in range(3) : for l in range(3): # acc += img[y-1+k, x-1+l] * kernel[k, l] acc += img.simd_load[nelts](y-1+k, x-1+l) * kernel[k, l] # Normalize the result # result[y, x] = min(255, max(0, acc)) result.simd_store[nelts](y, x, min(255, max(0, acc))) return result
Algorithm vectorization fn naive(img: Matrix[DType.float32], kernel: Matrix[DType.float32]) -> Matrix[DType.float32]: var result = Matrix[DType.float32](img.height, img.width) # Loop through each pixel in the image # But skip the outer edges of the image for y in range(1, img.height-1): for x in range(1, img.width-1, nelts): # For each pixel, compute the product elements wise # var acc: Float32 = 0 var acc: SIMD[DType.float32,nelts] = 0 for k in range(3) : for l in range(3): # acc += img[y-1+k, x-1+l] * kernel[k, l] acc += img.simd_load[nelts](y-1+k, x-1+l) * kernel[k, l] # Normalize the result # result[y, x] = min(255, max(0, acc)) result.simd_store[nelts](y, x, min(255, max(0, acc))) # Handle remaining elements with scalars. for n in range(nelts * (img.width-1 // nelts), img.width-1) : var acc: Float32 = 0 for k in range(3) : for l in range(3): acc += img[y-1+k, n-1+l] * kernel[k, l] result[y, n] = min(255, max(0, acc)) return result
Algorithm vectorization alias nelts = simdwidthofDType.float32 fn naive(img: Matrix[DType.float32], kernel: Matrix[DType.float32]) -> Matrix[DType.float32]: var result = Matrix[DType.float32](img.height, img.width) # Loop through each pixel in the image # But skip the outer edges of the image for y in range(1, img.height-1): @parameter fn dot[nelts: Int](x: Int): # For each pixel, compute the product elements wise var acc: SIMD[DType.float32,nelts] = 0 for k in range(3) : for l in range(3): acc += img.simd_load[nelts](y-1+k, x-1+l) * kernel[k, l] # Normalize the result result.simd_store[nelts](y, x, min(255, max(0, acc))) vectorizedot, nelts return result
Algorithm vectorization alias nelts = simdwidthofDType.float32 fn naive(img: Matrix[DType.float32], kernel: Matrix[DType.float32]) -> Matrix[DType.float32]: var result = Matrix[DType.float32](img.height, img.width) # Loop through each pixel in the image # But skip the outer edges of the image for y in range(1, img.height-1): @parameter fn dot[nelts: Int](x: Int): # For each pixel, compute the product elements wise var acc: SIMD[DType.float32,nelts] = 0 for k in range(3) : for l in range(3): acc += img.simd_load[nelts](y-1+k, x-1+l) * kernel[k, l] # Normalize the result result.simd_store[nelts](y, x, min(255, max(0, acc))) vectorizedot, nelts return result
Algorithm vectorization alias nelts = simdwidthofDType.float32 fn naive(img: Matrix[DType.float32], kernel: Matrix[DType.float32]) -> Matrix[DType.float32]: var result = Matrix[DType.float32](img.height, img.width) # Loop through each pixel in the image # But skip the outer edges of the image for y in range(1, img.height-1): @parameter fn dot[nelts: Int](x: Int): # For each pixel, compute the product elements wise var acc: SIMD[DType.float32,nelts] = 0 for k in range(3) : for l in range(3): acc += img.simd_load[nelts](y-1+k, x+l) * kernel[k, l] # Normalize the result result.simd_store[nelts](y, x+1, min(255, max(0, acc))) vectorizedot, nelts return result
• Far from stable • Compilation AOT or JIT Recap • Python friendly but not Python • Dynamic Python vs Static Mojo • Python interoperability • Predictable behavior with semantic ownership • Low level optimization • Blazingly fast
Mojo 🔥 The next AI Language?
Conclusion • Python is not yet dead ! But he moves slowly • This is a great team ! Will they be able to deploy their platform strategy ? • Will they be able to unite a community? To be open-source or not to be
Jean-Luc Tromparent Principal Engineer @ Merci ! https://linkedin.com/in/jltromparent https://github.com/jiel/laplacian_filters_benchmark https://noti.st/jlt/qheM0m 👉 I need your feedback 👈
Dans notre galaxie où l’innovation en matière d’intelligence artificielle est constamment en lutte avec la fragmentation, alimentée par une diversité de langages et de plateformes d’accélération matérielle, une dualité s’installe : d’un côté, Python règne en maître pour la modélisation et le prototypage, de l’autre, C, C++, Rust sont régulièrement convoqués pour la performance et la fiabilité. Cette dichotomie bride l’innovation, les experts fullstack se retrouvant souvent sous l’emprise de l’empire GAFAM.
Pourtant, un nouvel allié, le langage Mojo 🔥 émerge de l’ombre. Lancé il y a un an par la startup Modular créee par Chris Lattner (LLVM, CLang, Swift…) et désormais libéré en open source, Mojo se profile comme un puissant outil pour façonner l’avenir de l’IA.
J’ai voulu approfondir les concepts du langage et évaluer la courbe d’apprentissage en réalisant un benchmark de différentes implémentations d’un use-case de traitement d’image. Je vous raconterai dans ce talk comment après m’être laissé tenté par le coté obscur et connu un échec cuisant, je suis finalement revenu dans la voie du Jedi et obtenu des performances blazing fast.
for free. You
can too.