theilsen: code optimization(refer issue #52)

[Tanvi-Jain01]

@nipunbatra , @patel-zeel
This PR proposes solution for issue #52

BEFORE:

CODE:

vayu/vayu/TheilSen.py

Lines 1 to 113 in ef99aef

	def TheilSen(df, pollutant):
	"""Connected scatter plot.
	Plots a connected scatter plot of the average value of
	the pollutant every month of every year. Then plots a
	line of best fit through the plot showing the user
	the overall trend of the pollutant through the years.
	Parameters
	----------
	df: data frame
	minimally containing date and at least one other
	pollutant
	pollutant: type string
	A pollutant name correspoinding to
	a variable in a data frame, ex: 'pm25'
	"""
	import datetime as dt
	import matplotlib.pyplot as plt
	import matplotlib as mpl
	import numpy as np
	import pandas as pd
	from numpy import array
	import seaborn as sns
	import scipy
	from scipy import stats
	import math
	# =============================================================================
	# df = pd.read_csv("mydata.csv")
	# =============================================================================
	df.index = pd.to_datetime(df.date)
	unique_years = np.unique(df.index.year)
	# df = df[pd.notnull(df[pollutant])]
	i = 0
	year = []
	while i < len(unique_years):
	year.append(str(unique_years[i]))
	i = i + 1
	num_unique_years = len(year)
	# df = df.drop("date", axis=1)
	# print(df)
	i = 0
	x = 0
	j = 0
	var2 = []
	while i < num_unique_years:
	df_new = df[year[j]].resample("1D").mean()
	df_new = df_new.fillna(method="ffill")
	df_new["month"] = df_new.index.month
	# df_new['day']=df_new.index.dayofweek
	# df_new['hour']=df_new.index.hour
	i = i + 1
	j = j + 1
	x = 0
	while x < 12:
	a = df_new[df_new.month == x]
	mean_var2 = a[pollutant].mean()
	var2.append(mean_var2)
	x = x + 1
	i = 0
	while i < len(var2):
	if pd.notnull(var2[i]) == False:
	var2[i] = (var2[i - 1] + var2[i + 1]) / 2
	i = i + 1
	scatterX = []
	t = 0
	while t < num_unique_years:
	r = 0
	while r < 12:
	scatterX.append(t + (r / 12))
	r = r + 1
	t = t + 1
	y = var2
	x = scatterX
	def best_fit(X, Y):
	xbar = sum(X) / len(X)
	ybar = sum(Y) / len(Y)
	n = len(X) # or len(Y)
	numer = sum([xi * yi for xi, yi in zip(X, Y)]) - n * xbar * ybar
	denum = sum([xi ** 2 for xi in X]) - n * xbar ** 2
	b = numer / denum
	a = ybar - b * xbar
	# print('best fit line:\ny = {:.2f} + {:.2f}x'.format(a, b))
	return a, b
	a, b = best_fit(x, y)
	fig = plt.figure()
	ax = fig.add_subplot(111)
	# print(len(x))
	ax.plot(x, y, "-o")
	ax.set_xlabel("Year")
	ax.set_ylabel(pollutant)
	ax.set_title("TheilSen plot")
	plt.plot(np.unique(x), np.poly1d(np.polyfit(x, y, 1))(np.unique(x)), color="red")
	plt.show()
	# =============================================================================
	# df = pd.read_csv("mydata.csv")
	# TheilSen(df, 'o3')

OUTPUT:

AFTER:

CODE:

import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from matplotlib.ticker import ScalarFormatter

def theilsen(df:pd.DataFrame, pollutant:str):
    """Connected scatter plot.

    Plots a connected scatter plot of the average value of
    the pollutant every month of every year. Then plots a
    line of best fit through the plot showing the user
    the overall trend of the pollutant through the years.
    
    Parameters
    ----------
    df: data frame
        minimally containing date and at least one other
        pollutant 
    pollutant: type string
        A pollutant name correspoinding to 
        a variable in a data frame, ex: 'pm25'

    """
    
    df.index = pd.to_datetime(df.date)
    unique_years = np.unique(df.index.year)

    var2 = []
    scatterX = []

    for year in unique_years:
        df_year = df[df.index.year == year]
        df_monthly = df_year.resample("M").mean()
        monthly_mean = df_monthly[pollutant].values
        var2.extend(monthly_mean)
        scatterX.extend(np.arange(len(monthly_mean)) / 12 + year)

    y = np.array(var2)
    x = np.array(scatterX)

    def best_fit(X, Y):
        xbar = np.mean(X)
        ybar = np.mean(Y)
        n = len(X)

        numer = np.sum(X * Y) - n * xbar * ybar
        denum = np.sum(X ** 2) - n * xbar ** 2

        b = numer / denum
        a = ybar - b * xbar

        return a, b

    a, b = best_fit(x, y)

    fig, ax = plt.subplots()
    ax.plot(x, y, "-o")
    ax.set_xlabel("Year")
    ax.set_ylabel(pollutant)
    ax.set_title("TheilSen plot")
    plt.plot(np.unique(x), np.poly1d(np.polyfit(x, y, 1))(np.unique(x)), color="red")

    # Format y-axis tick labels
    ax.xaxis.set_major_formatter(ScalarFormatter(useOffset=False))
    plt.savefig("TheilSenplot.png", bbox_inches="tight")
    print("Your plot has also been saved")

    plt.show()


============================================================================

USAGE:

df = pd.read_csv("mydata.csv")
theilsen(df, 'pm25')

OUTPUT: