View the following notebooks on nbviewer:
All notebooks are stored at the notebooks directory.
This repository contains a series of data science notebooks that performs data science, visualization, and predictive analytics against the kaggle Dataset: 1.6M accidents & traffic flow over 16 years - Visualise and analyse rich data on traffic demographics
All notebooks are stored under the notebooks directory.
Table of Content
Download the zip file 1-6m-accidents-traffic-flow-over-16-years.zip from the kaggle website,
and extract it to somewhere suitable on your local machine. I assume it goes to your ~/Downloads
At the root of the repository, create a directory input and download (and extract) all the flat files there.
Step 1: Create input directory:
$ mkdir input
Step 2: Move the 1-6m-accidents-traffic-flow-over-16-years.zip to thie input directory
$ mv ~/Downloads/1-6m-accidents-traffic-flow-over-16-years.zip ./input
Step 3: extract all files within the same input directory:
$ cd input
$ tar -xvzf 1-6m-accidents-traffic-flow-over-16-years.zip`
Once you have the zip file extracted you should see some .csv files.
Local_Authority_Districts_Dec_2016.geojson
accidents_2005_to_2007.csv
accidents_2009_to_2011.csv
accidents_2012_to_2014.csv
accident_coords_update.ipynb
ukTrafficAADF.csv
$ conda env create -f environment.yml
$ source activate
$ source activate ds2
At the repository root directory:
$ jupyter notebook
Navigate to the notebooks directory, and pick the notebook you'd like.
All notebooks are stored under the notebooks directory. The Notebook README contains links to view the notebooks via the Jupyter nbviewer tool.