Summary
This assignment required me to leverage csv, randomly generated numbers, and an online API to conduct exploratory analyses on weather patterns across hundreds of cities.
After importing my dependencies and entering my API key into a file I created called "api_keys", I initialized some list variables, and used numpy to generate random latitudes and longitudes to fill one of the lists. Then, I used citypy to find the nearest city to each latitude and logitude. Next, I used the openweathermap API to obtain data for several weather dimensions across a single timepoint. Finally, I created scatterplots of those weather dimensions against the cities generated earlier. For a fully deployed webpage that showcases the scatterplots created in this notebook, click here.
Solution
WeatherPy¶
Analysis¶
- As expected, the weather becomes significantly warmer as one approaches the equator (0 Deg. Latitude). More interestingly, however, is the fact that the southern hemisphere tends to be warmer this time of year than the northern hemisphere. This may be due to the tilt of the earth.
- There is no strong relationship between latitude and cloudiness. However, it is interesting to see that a strong band of cities sits at 0, 80, and 100% cloudiness.
- There is no strong relationship between latitude and wind speed. However, in northern hemispheres there is a flurry of cities with over 20 mph of wind.
In [3]:
# Dependencies and Setup
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import requests
import time
from pprint import pprint
# Import API key
from api_keys import api_key
# Incorporated citipy to determine city based on latitude and longitude
from citipy import citipy
# Output File (CSV)
output_data_file = "output_data/cities.csv"
# Range of latitudes and longitudes
lat_range = (-90, 90)
lng_range = (-180, 180)
In [4]:
print(api_key)
Generate Cities List¶
In [5]:
# List for holding lat_lngs and cities
lat_lngs = []
cities = []
# Create a set of random lat and lng combinations
lats = np.random.uniform(low=-90.000, high=90.000, size=1500)
lngs = np.random.uniform(low=-180.000, high=180.000, size=1500)
lat_lngs = zip(lats, lngs)
# Identify nearest city for each lat, lng combination
for lat_lng in lat_lngs:
city = citipy.nearest_city(lat_lng[0], lat_lng[1]).city_name
# If the city is unique, then add it to a our cities list
if city not in cities:
cities.append(city)
# Print the city count to confirm sufficient count
len(cities)
Out[5]:
In [6]:
print(cities)
Perform API Calls¶
- Perform a weather check on each city using a series of successive API calls.
- Include a print log of each city as it'sbeing processed (with the city number and city name).
In [9]:
url = "http://api.openweathermap.org/data/2.5/weather?"
units = "metric"
# Build partial query URL
query_url = f"{url}appid={api_key}&units={units}&q="
print(query_url)
In [10]:
city_list = []
lat_list = []
lng_list = []
max_temp_list = []
humidity_list = []
cloudiness_list = []
windspeed_list = []
counter = 0
record_counter = 0
setcount = 1
for city in cities:
counter += 1
if (counter % 50 == 0):
setcount += 1
record_counter = 0
print(setcount)
print(record_counter)
print(counter)
record_counter += 1
print("processing Record " + str(record_counter) + " of Set " + str(setcount) + " | " + city)
try:
weather_response = requests.get(query_url + city).json()
#pprint(weather_response)
city_list.append(weather_response['name'])
lat_list.append(weather_response['coord']['lat'])
lng_list.append(weather_response['coord']['lon'])
max_temp_list.append(weather_response['main']['temp_max'])
humidity_list.append(weather_response['main']['humidity'])
cloudiness_list.append(weather_response['clouds']['all'])
windspeed_list.append(weather_response['wind']['speed'])
except:
print("City not found: " + city)
print(len(weather_response))
In [11]:
print(len(city_list))
print(len(lat_list))
print(len(lng_list))
print(len(max_temp_list))
print(len(humidity_list))
print(len(cloudiness_list))
print(len(windspeed_list))
In [12]:
weather_dict = {"city":city_list,
"lat":lat_list,
"lng":lng_list,
"max_temp":max_temp_list,
"humidity":humidity_list,
"cloudiness":cloudiness_list,
"windspeed":windspeed_list
}
Convert Raw Data to DataFrame¶
- Export the city data into a .csv.
- Display the DataFrame
In [13]:
weather_df = pd.DataFrame(weather_dict)
weather_df.head()
Out[13]:
In [19]:
weather_df.to_html("table.html")
Plotting the Data¶
- Use proper labeling of the plots using plot titles (including date of analysis) and axes labels.
- Save the plotted figures as .pngs.
Latitude vs. Temperature Plot¶
In [14]:
plt.scatter(weather_df['lat'], weather_df['max_temp'], edgecolor="black", linewidths=1, marker="o",
alpha=0.8, label="Cities")
plt.xlabel("Latitude")
plt.ylabel("Max Temperature (F)")
plt.title("City Latitude vs. Max Temperature (04/03/2019)")
plt.savefig("Max_temp.png")
Latitude vs. Humidity Plot¶
In [15]:
plt.scatter(weather_df['lat'], weather_df['humidity'], edgecolor="black", linewidths=1, marker="o",
alpha=0.8, label="Cities")
plt.xlabel("Latitude")
plt.ylabel("Humidity (%)")
plt.title("City Latitude vs. Humidity (04/03/2019)")
plt.savefig("Humidity.png")
Latitude vs. Cloudiness Plot¶
In [16]:
plt.scatter(weather_df['lat'], weather_df['cloudiness'], edgecolor="black", linewidths=1, marker="o",
alpha=0.8, label="Cities")
plt.xlabel("Latitude")
plt.ylabel("Cloudiness (%)")
plt.title("City Latitude vs. Cloudiness (04/03/2019)")
plt.savefig("Cloudiness.png")
Latitude vs. Wind Speed Plot¶
In [17]:
plt.scatter(weather_df['lat'], weather_df['windspeed'], edgecolor="black", linewidths=1, marker="o",
alpha=0.8, label="Cities")
plt.xlabel("Latitude")
plt.ylabel("Wind Speed (mph)")
plt.title("City Latitude vs. Wind Speed (04/03/2019)")
plt.savefig("Wind_Speed.png")
In [ ]: