GitHub - horodnicm7/PocketML: Machine Learning models and helper methods written in C++. Ideal for high speed apps.

horodnicm7 / PocketML Public

Notifications You must be signed in to change notification settings
Fork 0
Star 0

Machine Learning models and helper methods written in C++. Ideal for high speed apps.

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 17 Commits
models		models
python test scripts		python test scripts
reference		reference
tests		tests
utils		utils
Makefile		Makefile
README		README
test.cpp		test.cpp
test.txt		test.txt

Repository files navigation

Description:
        C/C++ implementation of common ML algorithms, helper classes and
methods, using efficient algorithms and highly optimized for cache usage.
The main goal is to get a good performance on SBCs (especially Raspberry Pi Zero
and Raspberry Pi 3 B+).

TODO:
        [0. Makefile]
        [1. Finish the linear regression model]
        [2. Write some tests]
        3. Write linear matrix operations using efficient algorithms
        [4. Finish the logistic regression model]
        5. Stress test and optimize the existing models