[파이썬/proj] Weather Forecasting

WeatherForecasting.html

617.6 kB

WeatherForecasting.ipynb

24.7 kB

 

 

Weather Forecasting using Python¶

 

Website: https://thecleverprogrammer.com/2022/10/17/weather-forecasting-using-python/

 

In Data Science, weather forecasting is an application of Time Series Forecasting where we use time-series data and algorithms to make forecasts for a given time. If you want to learn how to forecast the weather using your Data Science skills, this article is for you. In this article, I will take you through the task of weather forecasting using Python.

Weather Forecasting¶

Weather forecasting is the task of forecasting weather conditions for a given location and time. With the use of weather data and algorithms, it is possible to predict weather conditions for the next n number of days.

For forecasting weather using Python, we need a dataset containing historical weather data based on a particular location. I found a dataset on Kaggle based on the Daily weather data of New Delhi. We can use this dataset for the task of weather forecasting. You can download the dataset from here.

dataset download: https://www.kaggle.com/datasets/sumanthvrao/daily-climate-time-series-data

In the section below, you will learn how we can analyze and forecast the weather using Python.

Analyzing Weather Data using Python¶

Now let’s start this task by importing the necessary Python libraries and the dataset we need:

In [ ]:

import cmdstanpy
cmdstanpy.install_cmdstan()
cmdstanpy.install_cmdstan(compiler=True) # only valid on Windows

In [ ]:

import kaggle

# check api for dataset downloading from the website: 
!kaggle datasets download -d sumanthvrao/daily-climate-time-series-data

# unzip dataset 
from zipfile import ZipFile
dataset = "./daily-climate-time-series-data.zip"

with ZipFile(dataset, "r") as zip:
    zip.extractall()
    print("The dataset is extracted")

 

daily-climate-time-series-data.zip: Skipping, found more recently modified local copy (use --force to force download)
The dataset is extracted

In [ ]:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.express as px

data = pd.read_csv("./DailyDelhiClimateTrain.csv")
print(data.head())

 

         date   meantemp   humidity  wind_speed  meanpressure
0  2013-01-01  10.000000  84.500000    0.000000   1015.666667
1  2013-01-02   7.400000  92.000000    2.980000   1017.800000
2  2013-01-03   7.166667  87.000000    4.633333   1018.666667
3  2013-01-04   8.666667  71.333333    1.233333   1017.166667
4  2013-01-05   6.000000  86.833333    3.700000   1016.500000

 

Let’s have a look at the descriptive statistics of this data before moving forward:

In [ ]:

data.describe()

Out[ ]:

	meantemp	humidity	wind_speed	meanpressure
count	1462.000000	1462.000000	1462.000000	1462.000000
mean	25.495521	60.771702	6.802209	1011.104548
std	7.348103	16.769652	4.561602	180.231668
min	6.000000	13.428571	0.000000	-3.041667
25%	18.857143	50.375000	3.475000	1001.580357
50%	27.714286	62.625000	6.221667	1008.563492
75%	31.305804	72.218750	9.238235	1014.944901
max	38.714286	100.000000	42.220000	7679.333333

 

Now let’s have a look at the information about all the columns in the dataset:

In [ ]:

# Checking information of data type
data.info()

 

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1462 entries, 0 to 1461
Data columns (total 5 columns):
 #   Column        Non-Null Count  Dtype  
---  ------        --------------  -----  
 0   date          1462 non-null   object 
 1   meantemp      1462 non-null   float64
 2   humidity      1462 non-null   float64
 3   wind_speed    1462 non-null   float64
 4   meanpressure  1462 non-null   float64
dtypes: float64(4), object(1)
memory usage: 57.2+ KB

 

The date column in this dataset is not having a datetime data type. We will change it when required. Let’s have a look at the mean temperature in Delhi over the years:

In [ ]:

figure = px.line(data, x="date", y="meantemp",
                title = "Mean Temperature in Delhi Over the Years")
figure.show()

 

 

Now let’s have a look at the humidity in Delhi over the years:

In [ ]:

figure = px.line(data, x ="date", y = "humidity", 
                title = "Humidity in Delhi Over the Years")
figure.show()

 

 

Now let’s have a look at the wind speed in Delhi over the years:

In [ ]:

figure = px.line(data, x="date", y="wind_speed",
                title = "Wind Speed in Delhi Over the Years")
figure.show()

 

 

Till 2015, the wind speed was higher during monsoons (August & September) and retreating monsoons (December & January). After 2015, there were no anomalies in wind speed during monsoons. Now let’s have a look at the relationship between temperature and humidity:

In [ ]:

figure = px.scatter(data_frame=data, x="humidity", 
                    y="meantemp", size="meantemp", 
                    trendline="ols", 
                    title="Relationship Between Temperature and Humidity")
figure.show()

 

 

There’s a negative correlation between temperature and humidity in Delhi. It means higher temperature results in low humidity and lower temperature results in high humidity.

Analyzing Temperature Change¶

Now let’s analyze the temperature change in Delhi over the years. For this task, I will first convert the data type of the date column into datetime. Then I will add two new columns in the dataset for year and month values.

Here’s how we can change the data type and extract year and month data from the date column:

In [ ]:

data["date"] = pd.to_datetime(data["date"], format="%Y-%m-%d")
data["year"] = data["date"].dt.year
data["month"] = data["date"].dt.month
data.head()

Out[ ]:

	date	meantemp	humidity	wind_speed	meanpressure	year	month
0	2013-01-01	10.000000	84.500000	0.000000	1015.666667	2013	1
1	2013-01-02	7.400000	92.000000	2.980000	1017.800000	2013	1
2	2013-01-03	7.166667	87.000000	4.633333	1018.666667	2013	1
3	2013-01-04	8.666667	71.333333	1.233333	1017.166667	2013	1
4	2013-01-05	6.000000	86.833333	3.700000	1016.500000	2013	1

 

Now let’s have a look at the temperature change in Delhi over the years:

In [ ]:

plt.style.use('fivethirtyeight')
plt.figure(figsize=(15, 10))
plt.title("Temperature Change in Delhi Over the Years")
sns.lineplot(data = data, x = "month", y = "meantemp", hue = "year");

 

 

Although 2017 was not the hottest year in the summer, we can see a rise in the average temperature of Delhi every year.

Forecasting Weather using Python¶

Now let’s move to the task of weather forecasting. I will be using the Facebook prophet model for this task. The Facebook prophet model is one of the best techniques for time series forecasting. If you have never used this model before, you can install it on your system by using the command mentioned below in your command prompt or terminal:

 

The prophet model accepts time data named as “ds”, and labels as “y”. So let’s convert the data into this format:

In [ ]:

forcast_data = data.rename(columns = {"date": "ds","meantemp": "y" })
forcast_data

Out[ ]:

	ds	y	humidity	wind_speed	meanpressure	year	month
0	2013-01-01	10.000000	84.500000	0.000000	1015.666667	2013	1
1	2013-01-02	7.400000	92.000000	2.980000	1017.800000	2013	1
2	2013-01-03	7.166667	87.000000	4.633333	1018.666667	2013	1
3	2013-01-04	8.666667	71.333333	1.233333	1017.166667	2013	1
4	2013-01-05	6.000000	86.833333	3.700000	1016.500000	2013	1
...	...	...	...	...	...	...	...
1457	2016-12-28	17.217391	68.043478	3.547826	1015.565217	2016	12
1458	2016-12-29	15.238095	87.857143	6.000000	1016.904762	2016	12
1459	2016-12-30	14.095238	89.666667	6.266667	1017.904762	2016	12
1460	2016-12-31	15.052632	87.000000	7.325000	1016.100000	2016	12
1461	2017-01-01	10.000000	100.000000	0.000000	1016.000000	2017	1

1462 rows × 7 columns

 

Now below is how we can use the Facebook prophet model for weather forecasting using Python:

In [ ]:

from prophet import Prophet
from prophet.plot import plot_plotly, plot_components_plotly
model = Prophet()

model.fit(forcast_data)
forcasts = model.make_future_dataframe(periods=365)
predictions = model.predict(forcasts)
plot_plotly(model, predictions)

 

 

So this is how you can analyze and forecast the weather using Python.

Summary¶

Weather forecasting is the task of forecasting weather conditions for a given location and time. With the use of weather data and algorithms, it is possible to predict weather conditions for the next n number of days. I hope you liked this article on Weather Analysis and Forecasting using Python. Feel free to ask valuable questions in the comments section below.