So, If you have read my previous blog in this quickbites series, you probably understood what are the types of supervised learning, and also more about Regression and Classification.
And also Python is really famous language for Data science and ML. Hence it has lots of libraries for ML that are really good.
The libraries in python
The good and neat libraries for ML in python:
This is a famous library and the easiest way to get started with machine learning in python. Its really neat, and easy to use library, and follows the principle of DRY, where each components use same convention and same methods so as to make it easier to remember, and control. Also it allows you to make your own methods and mixins for custom access and control.
This is a old library for deep learning that was worked from 2015 to 2018. The code isn't still maintained anymore, but it can produce neat well performing neural networks and various other model and stuff for manipulations and training. It provides a rich API consisting of various interesting stuff, allowing a lot of new access back in the year of 2018.
As their description says "The core open source library to help you develop and train ML models." and Yeah, that's really true. It's open source since 2018, and providing a lot of unique methods and ways for training methods, and counting. It has recently added audio based model predictions, and tensorflow reinforcement learning too. Amazing right?
PyTorch is an amazing library constituting to the deep learning ecosystem. It is an open source machine learning framework that accelerates the path from research prototyping to production deployment. Providing numerous ways to deploy your model everywhere, training your model, and also exposing the C++ API!
You can integrate the C++ API for optimized code and integration into Python.
OpenCV is a great library based on computer vision and camera based model detection. It is used for face detection and various more implementations requiring vision. It also has amazing prebuilt models ready for use.
Machine learning algorithms for classification
- Logistic regression.
- Decision trees.
- Random forest.
- Gradient boosting.
- SVC
- K neighbors.
- Naive Bayes.
Machine learning regression algorithms
- Simple Linear Regression model.
- Lasso Regression.
- Logistic regression.
- Support Vector Machines.
- Multivariate Regression algorithm.
- Multiple Regression Algorithm.
We just had a simple overview about which ones are used. We'll be talking about 4 of the most used algorithms from each of the categories :)
See you next time!