This project aims at teaching you the fundamentals of Machine Learning in python.
Simply open the Jupyter notebooks you are interested in:
- using Binder:
- this let's you experiment with the code examples
- or using jupyter.org's notebook viewer
- note: github.com's notebook viewer also works but it is slower and the math formulas are not displayed correctly
- or by cloning this repository and running Jupyter locally
- if you prefer this option, follow the installation instructions below.
No installation is required, just click the launch binder button above, and you're good to go! But if you insist, here's how to install these notebooks on your system.
Obviously, you will need git and python (python 3 is recommended, but python 2 should work as well).
First, clone this repository:
$ cd {your development directory}
$ git clone https://github.com/ageron/handson-ml.git
$ cd handson-ml
If you want an isolated environment, you can use virtualenv:
$ virtualenv env
$ source ./env/bin/activate
There are different packages for TensorFlow, depending on your platform. Please edit requirements.txt
using your favorite editor, and make sure only the right one for your platform is uncommented. Default is Python 3.5, Ubuntu/Linux 64-bits, CPU-only.
Then install the required python packages using pip:
$ pip install -r requirements.txt
If you want to install the Jupyter extensions, run the following command (this is optional but recommended as it allows you to view a table of contents in each notebook):
$ jupyter contrib nbextension install --user
Finally, launch Jupyter:
$ jupyter notebook
This should start the Jupyter server locally, and open your browser. Click on index.ipynb
to get started (note: you can visit /nbextensions
to turn extensions on or off).