machine-learning-101

This repository contains Jupyter notebooks for learning the basics of Machine Learning. Each notebook has a run script that will start a docker container with the appropriate environment for the Jupyter server.

Basic machine learning concepts

training dataset - the data you use for training your model
test dataset - the data you use to test your model
features - the attributes of your data that are used by the model as input
label(s) - the attribute(s) of your data that the model is trying to predict

Workflow of solving a machine learning problem

1. Analyse and improve the quality of your data
2. Perform transformations on your data (example: change the type of some features or represent the differently)
3. Understand your data
4. Choose a model for training
5. Train using part of your training dataset
6. Test using the other part of your training dataset
7. Measure accuracy of your model
8. Go back to any of the steps above in order to improve the accuracy
9. Test your model using the test dataset
10. Celebrate! :)

Notebooks

Notebook: machine-learning/inspecting-data/inspecting-data.ipynb
Description: contains basic commands for inspecting the structure and quality of your data sets
Run: schipyrun.sh

Notebook: machine-learning/linear-regression/linear-regression.ipynb
Description: contains a simple example of training a model using Linear Regression
Run: schipyrun.sh

Notebook: tensor-flow/tf-simple-operations
Description: contains basic TensorFlow evaluations
Run: tfrun.sh

Author: Oana Cioara
Copyright: The content of this repository is public. You are allowed to access the "source-code" and use it as you wish, for study or change it for personal use.