The API provides the list of sports which can be practiced given the current weather conditions or the weather forecast. The inputs of the API are the geolocation (lat and lng) of interest and the number of days (zero to four) over which we want to consider the weather. If days=0 is passed as an argument, the current weather conditions are used, while the weather forecast is used when days > 0. The output is the list of sports matching the weather conditions.
For any additional information, please contact samuel.mercier@decathlon.com
When the API is called, the weather at the given location is obtained by calling pyowm, a python wrapper around the OpenWeatherMap web API. Two features are extracted: if the current weather is used (days=0), the two features are the current temperature and if it currently rains or not; if we want to match sports to the forecast (days > 0), the two features are the average temperature at noon and 3 pm (local time) over the forecasting period and whether rain is forecasted at more than half of these time steps. The sports then are recommended following these general guidelines
| Weather | Sports recommended |
|---|---|
| Rainy | indoor sports |
| Cold temperature (< 0 °C), not rainy | outdoor winter sports |
| Chilly temperature (0-10 °C), not rainy | outdoor year-long sports (hiking, jogging, etc.) |
| Comfortable temperature (10-20 °C), not rainy | outdoor summer sports, excluding water sports |
| Warm temperature (20-30 °C), not rainy | outdoor summer sports |
| Very warm temperature (> 30 °C), not rainy | outdoor water sports |
The association between the sports and the weather can be found and edited in the ./data/weather-sport-master-list.csv file. Note that a given sport can be found accross a number of weather conditions (exemple: ice hockey, which is both an indoor sport and an outdoor winter sport).
- git clone the project to the desired location
git clone https://github.com/decathloncanada/weather-sport-API.git
-
Make sure you have python 3 installed
-
Install the dependencies:
sudo pip install -r requirements.txt
- Add a config.py file, in the root of the project, containing your OpenWeatherMap API key. The config.py file should be as follows:
owm_key={OWM_KEY}
where you replace {OWM_KEY} with your Open Weather Map key.
To get the sport recommendations at a given location given the weather forecast, run the project as follows:
python app.py
The API will then be available at http://localhost:5000/weather-sport-API. The API takes two arguments:
- origin : longitude and latitude of the location of interest
- days: the period (number of days) over which we consider the weather forecast. Should be an integer between 1 and 4.
The API returns a json containing the ID of the recommended sports and a short description of the weather forecast in a json format.
As an example, here's a call to recommend the sports given the weather forecasted over the next two days at a given location in Montreal:
http://localhost:5000/weather-sport-API?origin=-73.582,45.511&days=2
The returned json should look like:
{"forecast":"Warm temperature (between 20 and 30 Celsius), not rainy","sports_recommended":[12,29,37,45,49,51,74,76,78,81,87,89,98,134,141,143,147,150,163,166,171,178,180,182,186,219,224,280,281,282,283,284,285,286,287,288,289,292,294,296,298,305,306,308,309,310,311,312,318,319,320,321,322,323,324,328,329,330,331,389,392,395,404,409,449,461,466,471,473,487,490]}
The information about the sports given their IDs can be obtained from the Sports API.
The next steps of the project will be to sync the API with the updated version of the Sports API, and include informations about the sun, wind and snow conditions to refine the recommendations.