World_weather_analysis
Overview
Using multiple API sources, create an app that allows users to filter through random locations based on their temperature range choices.
Process
CitiPy and OpenWeather API
With a random list of 2000 latitudes and longitudes, we used citipy to locate the nearest city and then looped through the OpenWeather API to gather weather data about each city and formatted into a DataFrame.
Google API
After cleaning the DataFrame of any null values, we ask the user to input their desired temperature min and max values and then using the .loc method, we created a new DataFrame to include cities that fell in the users temperature range. After inserting a new column, "Hotel Name" into our DataFrame, we used Googles 'nearbysearch' endpoint API with a set of parameters to fill in a hotel for each city in our DF. A marker layer was made with each cities info and a figure was displayed using gmaps.figure.
Create a travel itinerary map
With our filtered DataFrame, we narrowed it down to four cities in the same country. Creating DataFrames for each country and then combining them, we used gmaps.directions to create a route between each city and displayed the map.