Awesome Open Source
Awesome Open Source

animated

What's new in Hyperlearn 2022!

animated

We're building a full Earth simulation to predict the future of everything and make JARVIS reality.
Ask complex questions you never thought possible like

How long will I live?
When will the market next crash?
How will climate change affect me?
Hyperlearn
Would you wait a second for a Google search? How about 5 hours for a prediction?
Hyperlearn makes Moonshot run fast and makes ML algos faster and use less memory.

! Hyperlearn is under construction! A stable package will be reuploaded mid 2022! Stay tuned!

Moonshot Website (under SEVERE construction)

Documentation

50 Page Modern Big Data Algorithms PDF

In 2018-2020, I was at NVIDIA helping make GPU ML algos faster! I incorporated Hyperlearn's methods to make TSNE 2000x faster, and others faster. Since then, I have 50+ fast algos, but didn't have time to update Hyperlearn since Moonshot was priority one! I'll be updating Hyperlearn mid 2022!


Hyperlearn's algorithms, methods and repo has been featured or mentioned in 5 research papers!

+ Microsoft, UW, UC Berkeley, Greece, NVIDIA

Hyperlearn's methods and algorithms have been incorporated into 5 organizations and repositories!

+ Facebook's Pytorch, Scipy, Cupy, NVIDIA, UNSW

During Hyperlearn's development, bugs and issues were notified to GCC!


Packages Used

HyperLearn is written completely in PyTorch, NoGil Numba, Numpy, Pandas, Scipy & LAPACK, C++, C, Python, Cython and Assembly, and mirrors (mostly) Scikit Learn. HyperLearn also has statistical inference measures embedded, and can be called just like Scikit Learn's syntax.

Some key current achievements of HyperLearn:

  • 70% less time to fit Least Squares / Linear Regression than sklearn + 50% less memory usage
  • 50% less time to fit Non Negative Matrix Factorization than sklearn due to new parallelized algo
  • 40% faster full Euclidean / Cosine distance algorithms
  • 50% less time LSMR iterative least squares
  • New Reconstruction SVD - use SVD to impute missing data! Has .fit AND .transform. Approx 30% better than mean imputation
  • 50% faster Sparse Matrix operations - parallelized
  • RandomizedSVD is now 20 - 30% faster

Around mid 2022, Hyperlearn will evolve to GreenAI and aims to incorporate:

  • New Paratrooper optimizer - fastest SGD variant combining Lookahead, Learning Rate Range Finder, and more!
  • 30% faster Matrix Multiplication on CPUs
  • Software Support for brain floating point (bfloat16) on nearly all hardware
  • Easy compilation on old and new CPU hardware (x86, ARM)
  • 100x faster regular expressions
  • 50% faster and 50% less memory usage for assembly kernel accelerated methods
  • Fast and parallelized New York Times scraper
  • Fast and parallelized NYSE Announcements scraper
  • Fast and parallelized FRED scraper
  • Fast and parallelized Yahoo Finance scraper

I also published a mini 50 page book called "Modern Big Data Algorithm"!

Modern Big Data Algorithms PDF

Modern Big Data Algorithms

Comparison of Speed / Memory

Algorithm n p Time(s) RAM(mb) Notes
Sklearn Hyperlearn Sklearn Hyperlearn
QDA (Quad Dis A) 1000000 100 54.2 22.25 2,700 1,200 Now parallelized
LinearRegression 1000000 100 5.81 0.381 700 10 Guaranteed stable & fast

Time(s) is Fit + Predict. RAM(mb) = max( RAM(Fit), RAM(Predict) )

I've also added some preliminary results for N = 5000, P = 6000

drawing

Help is really needed! Message me!


Key Methodologies and Aims

1. Embarrassingly Parallel For Loops

2. 50%+ Faster, 50%+ Leaner

3. Why is Statsmodels sometimes unbearably slow?

4. Deep Learning Drop In Modules with PyTorch

5. 20%+ Less Code, Cleaner Clearer Code

6. Accessing Old and Exciting New Algorithms


1. Embarrassingly Parallel For Loops

  • Including Memory Sharing, Memory Management
  • CUDA Parallelism through PyTorch & Numba

2. 50%+ Faster, 50%+ Leaner

3. Why is Statsmodels sometimes unbearably slow?

  • Confidence, Prediction Intervals, Hypothesis Tests & Goodness of Fit tests for linear models are optimized.
  • Using Einstein Notation & Hadamard Products where possible.
  • Computing only what is necessary to compute (Diagonal of matrix and not entire matrix).
  • Fixing the flaws of Statsmodels on notation, speed, memory issues and storage of variables.

4. Deep Learning Drop In Modules with PyTorch

  • Using PyTorch to create Scikit-Learn like drop in replacements.

5. 20%+ Less Code, Cleaner Clearer Code

  • Using Decorators & Functions where possible.
  • Intuitive Middle Level Function names like (isTensor, isIterable).
  • Handles Parallelism easily through hyperlearn.multiprocessing

6. Accessing Old and Exciting New Algorithms

  • Matrix Completion algorithms - Non Negative Least Squares, NNMF
  • Batch Similarity Latent Dirichelt Allocation (BS-LDA)
  • Correlation Regression
  • Feasible Generalized Least Squares FGLS
  • Outlier Tolerant Regression
  • Multidimensional Spline Regression
  • Generalized MICE (any model drop in replacement)
  • Using Uber's Pyro for Bayesian Deep Learning

Goals & Development Schedule

Hyperlearn will be revamped in the following months to become Moonshot GreenAI with over an extra 150 optimized algorithms! Stay tuned!! Also you made it this far! If you want to join Moonshot, complete the secretive quiz!

Join Moonshot!


Extra License Terms

  1. Hyperlearn is intended for academic, research and personal purposes only. Any explicit commercialisation of the algorithms, methods and anything inside Hyperlearn is strictly prohibited unless explicit notice is given to Daniel Han-Chen. The usage must also be approved by Daniel Han-Chen.
  2. Hyperlearn uses the BSD 3 License now (previously GNU v3). However, as stated, commercialisation on top of Hyperlearn must be first approved by Daniel Han-Chen.

ask the impossible


Get A Weekly Email With Trending Projects For These Topics
No Spam. Unsubscribe easily at any time.
Python (1,118,602
Jupyter Notebook (234,411
Machine Learning (30,860
Deep Learning (22,971
Pytorch (11,190
Data Science (8,861
Neural Network (8,440
Statistics (4,210
Data Analysis (3,655
Optimization (2,527
Scikit Learn (1,943
Gpu (1,647
Research (1,489
Regression Models (385
Tensor (287
Econometrics (210
Related Projects