ml-light-talk-02-2022

Light talk about introduction to machine learning

Setup

libraries

All python dependencies are inside requirements.txt file. Short sum up of what's included:

• jupyter is a notebook we will use to nicely and interactivelly play with our code
• pandas is a base library for data manipulation, it includes numpy which is a math library
• matplotlib is basic chart library. seaborn is a wrapper with high level API
• scikit-learn in short sklearn, is a simple data analysis tool with machine learning capabilities
• keras is a tensorflow 2 based library, used for deep learning machine learning algorithms

Dockerfile

We use docker to run the project so we do not need to worry about python environment etc.. It is based on python base image, and includes tini to make jupyter notebook work through docker.

Makefile

Simple makefile setup for building and running our docker setup locally.

Data

Taken from https://www.kaggle.com/aleksandrglotov/car-prices-poland?select=Car_Prices_Poland_Kaggle.csv

Links

• Numpy: https://numpy.org/doc/stable/reference/
• Pandas: https://pandas.pydata.org/docs/user_guide/index.html
• Scikit Learn: https://scikit-learn.org/stable/getting_started.html
• Keras: https://keras.io/getting_started/
• Seaborn: https://seaborn.pydata.org/

Resources

• kaggle: https://www.kaggle.com/
• medium/towardsdatascience: https://towardsdatascience.com/
• datacamp: https://datacamp.com
• statquest yt: https://www.youtube.com/c/joshstarmer

Images

• Linear regression: https://i.stack.imgur.com/zoYKG.png
• Logistic regression: https://miro.medium.com/max/725/0*tLu8lvAEomHZm2YK.png
• Random forest: https://static.javatpoint.com/tutorial/machine-learning/images/random-forest-algorithm2.png
• Classification scoring metrics: https://www.researchgate.net/publication/336849041/figure/fig2/AS:818957984468994@1572266207473/Confusion-matrix-and-classification-metrics-terminology.ppm
• Overfitting / underfitting: https://datascience.foundation/img/pdf_images/underfitting_and_overfitting_in_machine_learning_image.png
• Neural network: https://1.cms.s81c.com/sites/default/files/2021-04-15/ICLH_Diagram_Batch_01_03-DeepNeuralNetwork-WHITEBG.png
• Neural network example: https://pbs.twimg.com/media/ELEHKR8UEAAsIHt.jpg
• Activation functions: https://qph.fs.quoracdn.net/main-qimg-65a7c3bf75549bad04875d0e789bb5bf