Home

This is the public challenge for the Machine Learning Specialist and Data Scientist positions at 3778 Healthcare.

The dataset used for this challenge consists of time-series data on economic and health indicators of multiple countries. The objective is to predict the values for 2013, 2014 and 2015 using previous years as well as 2016 and 2017. For more information regarding the dataset please see Dataset.

Instructions

1. Generate a private personal repository using this template.
2. Download hnp_stats_csv.zip, extract it into data/. The final directory should look like this:

data
├── HNP_StatsCountry.csv
├── HNP_StatsCountry-Series.csv
├── HNP_StatsData.csv
├── HNP_StatsFootNote.csv
├── HNP_StatsSeries.csv
└── HNP_StatsSeries-Time.csv

3. Install dependencies and run python -m challenge._internal.make_data to create files data/{data,test,answers}.csv.
4. Train your model and evaluate it with the supplied function;
5. Use README.md or a Jupyter notebook to summarize your results (see Request an evaluation for more details).
6. Request an evaluation.

Make sure you followed the rules, otherwise you will be ignored.

Remember, this is an iterative process. Refer to the diagram for more information.

Feel free to create an issue regarding questions and suggestions.

Table of Contents

1. Dataset
2. Rules
3. How to evaluate your model
4. How to request an evaluation by 3778