About This Project

# Let's import the project's cover image from PIL import Image image = Image.open("/work/front_internet_worlwide.png") image.show()

Set Up

# Import requried libraries import pandas as pd import numpy as np import plotly.express as px import time from datetime import datetime !pip install gspread !pip install gspread_dataframe from gspread_dataframe import set_with_dataframe import gspread gc = gspread.service_account(filename='/work/world-wide-web-356017-564c77672260.json')

Requirement already satisfied: gspread in /root/venv/lib/python3.9/site-packages (5.4.0)
Requirement already satisfied: google-auth>=1.12.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from gspread) (2.7.0)
Requirement already satisfied: google-auth-oauthlib>=0.4.1 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from gspread) (0.4.6)
Requirement already satisfied: six>=1.9.0 in /shared-libs/python3.9/py-core/lib/python3.9/site-packages (from google-auth>=1.12.0->gspread) (1.16.0)
Requirement already satisfied: pyasn1-modules>=0.2.1 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from google-auth>=1.12.0->gspread) (0.2.8)
Requirement already satisfied: rsa<5,>=3.1.4 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from google-auth>=1.12.0->gspread) (4.8)
Requirement already satisfied: cachetools<6.0,>=2.0.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from google-auth>=1.12.0->gspread) (5.2.0)
Requirement already satisfied: requests-oauthlib>=0.7.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from google-auth-oauthlib>=0.4.1->gspread) (1.3.1)
Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from pyasn1-modules>=0.2.1->google-auth>=1.12.0->gspread) (0.4.8)
Requirement already satisfied: oauthlib>=3.0.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread) (3.2.0)
Requirement already satisfied: requests>=2.0.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread) (2.27.1)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from requests>=2.0.0->requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread) (1.26.9)
Requirement already satisfied: certifi>=2017.4.17 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from requests>=2.0.0->requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread) (2022.5.18.1)
Requirement already satisfied: idna<4,>=2.5 in /shared-libs/python3.9/py-core/lib/python3.9/site-packages (from requests>=2.0.0->requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread) (3.3)
Requirement already satisfied: charset-normalizer~=2.0.0 in /shared-libs/python3.9/py-core/lib/python3.9/site-packages (from requests>=2.0.0->requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread) (2.0.12)
WARNING: You are using pip version 22.0.4; however, version 22.1.2 is available.
You should consider upgrading via the '/root/venv/bin/python -m pip install --upgrade pip' command.
Requirement already satisfied: gspread_dataframe in /root/venv/lib/python3.9/site-packages (3.3.0)
Requirement already satisfied: six>=1.12.0 in /shared-libs/python3.9/py-core/lib/python3.9/site-packages (from gspread_dataframe) (1.16.0)
Requirement already satisfied: gspread>=3.0.0 in /root/venv/lib/python3.9/site-packages (from gspread_dataframe) (5.4.0)
Requirement already satisfied: pandas>=0.24.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from gspread_dataframe) (1.2.5)
Requirement already satisfied: google-auth>=1.12.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from gspread>=3.0.0->gspread_dataframe) (2.7.0)
Requirement already satisfied: google-auth-oauthlib>=0.4.1 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from gspread>=3.0.0->gspread_dataframe) (0.4.6)
Requirement already satisfied: pytz>=2017.3 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from pandas>=0.24.0->gspread_dataframe) (2022.1)
Requirement already satisfied: python-dateutil>=2.7.3 in /shared-libs/python3.9/py-core/lib/python3.9/site-packages (from pandas>=0.24.0->gspread_dataframe) (2.8.2)
Requirement already satisfied: numpy>=1.16.5 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from pandas>=0.24.0->gspread_dataframe) (1.22.4)
Requirement already satisfied: cachetools<6.0,>=2.0.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from google-auth>=1.12.0->gspread>=3.0.0->gspread_dataframe) (5.2.0)
Requirement already satisfied: pyasn1-modules>=0.2.1 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from google-auth>=1.12.0->gspread>=3.0.0->gspread_dataframe) (0.2.8)
Requirement already satisfied: rsa<5,>=3.1.4 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from google-auth>=1.12.0->gspread>=3.0.0->gspread_dataframe) (4.8)
Requirement already satisfied: requests-oauthlib>=0.7.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from google-auth-oauthlib>=0.4.1->gspread>=3.0.0->gspread_dataframe) (1.3.1)
Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from pyasn1-modules>=0.2.1->google-auth>=1.12.0->gspread>=3.0.0->gspread_dataframe) (0.4.8)
Requirement already satisfied: requests>=2.0.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread>=3.0.0->gspread_dataframe) (2.27.1)
Requirement already satisfied: oauthlib>=3.0.0 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread>=3.0.0->gspread_dataframe) (3.2.0)
Requirement already satisfied: certifi>=2017.4.17 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from requests>=2.0.0->requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread>=3.0.0->gspread_dataframe) (2022.5.18.1)
Requirement already satisfied: charset-normalizer~=2.0.0 in /shared-libs/python3.9/py-core/lib/python3.9/site-packages (from requests>=2.0.0->requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread>=3.0.0->gspread_dataframe) (2.0.12)
Requirement already satisfied: idna<4,>=2.5 in /shared-libs/python3.9/py-core/lib/python3.9/site-packages (from requests>=2.0.0->requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread>=3.0.0->gspread_dataframe) (3.3)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in /shared-libs/python3.9/py/lib/python3.9/site-packages (from requests>=2.0.0->requests-oauthlib>=0.7.0->google-auth-oauthlib>=0.4.1->gspread>=3.0.0->gspread_dataframe) (1.26.9)
WARNING: You are using pip version 22.0.4; however, version 22.1.2 is available.
You should consider upgrading via the '/root/venv/bin/python -m pip install --upgrade pip' command.

Extract

# extract first df and create variable "prices". # list of default na values is redefined to exclude string 'NA' which happens to be Namibia's country code prices = pd.read_csv('/work/cerebro/worldwide internet prices in 2022 - IN 2022.csv',keep_default_na=False, na_values=['-1.#IND', '1.#QNAN', '1.#IND', '-1.#QNAN', '#N/A N/A', '#N/A', 'N/A', 'n/a', '#NA', 'NULL', 'null', 'NaN', '-NaN', 'nan', '-nan', ''])

# extract second df and create variable "speed" speed = pd.read_csv('/work/cerebro/worldwide internet speed in 2022 - avg speed.csv')

# extract third df and create variable "users" users = pd.read_csv('/work/cerebro/worldwide internet users - users.csv')

Transform

Prices

# This is our first DF prices

Country codeobject

TT0.8%

CW0.8%

238 others98.3%

Nameobject

Tuvalu0.8%

Israel0.4%

239 others98.8%

Israel

Kyrgyzstan

Fiji

Italy

Sudan

Russian Federation

Moldova

Bangladesh

Sri Lanka

Chile

# Checking prices columns, null values and data types prices.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 242 entries, 0 to 241
Data columns (total 9 columns):
 #   Column                                            Non-Null Count  Dtype  
---  ------                                            --------------  -----  
 0   Country code                                      242 non-null    object 
 1   Name                                              242 non-null    object 
 2   Continental region                                242 non-null    object 
 3   NO. OF Internet Plans                             230 non-null    float64
 4   Average price of 1GB (USD)                        242 non-null    object 
 5   Cheapest 1GB for 30 days (USD)                    230 non-null    object 
 6   Most expensive 1GB (USD)                          230 non-null    object 
 7   Average price of 1GB (USD  at the start of 2021)  230 non-null    object 
 8   Average price of 1GB (USD – at start of 2020)     230 non-null    object 
dtypes: float64(1), object(8)
memory usage: 17.1+ KB

prices.loc[prices['NO. OF Internet Plans '].isnull(),]

Country codeobject

CW8.3%

ER8.3%

10 others83.3%

Nameobject

Christmas Island8.3%

Eritrea8.3%

10 others83.3%

230

Christmas Island

231

Eritrea

232

North Korea

233

Marshall Islands

234

St. Pierre and Miquelon

235

South Sudan

236

Tuvalu

237

Vatican City (Holy See)

238

Venezuela

239

Wallis and Futuna

# drop rows with null values for selected columns only prices.dropna(subset=['NO. OF Internet Plans ','Average price of 1GB (USD)','Cheapest 1GB for 30 days (USD)', 'Most expensive 1GB (USD)','Average price of 1GB (USD at the start of 2021)', 'Average price of 1GB (USD – at start of 2020)'],inplace=True)

prices

Country codeobject

IL0.4%

KG0.4%

228 others99.1%

Nameobject

Israel0.4%

Kyrgyzstan0.4%

228 others99.1%

Israel

Kyrgyzstan

Fiji

Italy

Sudan

Russian Federation

Moldova

Bangladesh

Sri Lanka

Chile

# First, let's apply a function to delete the $ sign cols = ['Average price of 1GB (USD)', 'Cheapest 1GB for 30 days (USD)', 'Most expensive 1GB (USD)','Average price of 1GB (USD at the start of 2021)','Average price of 1GB (USD – at start of 2020)'] prices[cols] = prices[cols].apply(lambda x: x.str.lstrip('$')) # now let's convert these columns to numeric prices[cols] = prices[cols].apply(pd.to_numeric, errors= 'coerce') # Finally, column names are a bit messy, we could improve them prices.rename({'NO. OF Internet Plans ':'num_internet_plans','Average price of 1GB (USD)': 'avg_price_1gb','Cheapest 1GB for 30 days (USD)':'cheapest_1gb_30_days', 'Most expensive 1GB (USD)':'most_expensive_1gb','Average price of 1GB (USD at the start of 2021)':'avg_price_1gb_2021','Average price of 1GB (USD – at start of 2020)':'avg_price_1gb_2020'}, axis=1, inplace=True)

prices.head()

Country codeobject

Nameobject

Israel

Kyrgyzstan

Fiji

Italy

Sudan

# As a good practice, let's get rid of whitespaces at the begining and end in all three columns cols = ['Country code', 'Name', 'Continental region'] prices[cols] = prices[cols].apply(lambda x: x.str.strip()) # We can also improve the column names prices.rename({'Country code':'country_code','Name': 'name','Continental region':'continental_region'}, axis=1, inplace=True)

prices['continental_region'].value_counts()

# Reduce continental_region column to 5 regions conditions = [ (prices['continental_region'].isin(('SOUTH AMERICA', 'CENTRAL AMERICA','NORTHERN AMERICA','CARIBBEAN'))), (prices['continental_region'].isin(('WESTERN EUROPE', 'EASTERN EUROPE', 'BALTICS'))), (prices['continental_region'].isin(('SUB-SAHARAN AFRICA', 'NORTHERN AFRICA'))), (prices['continental_region'].isin(('ASIA (EX. NEAR EAST)', 'NEAR EAST','CIS (FORMER USSR)'))), (prices['continental_region'].str.contains(('OCEANIA'))) ] regions = ['Americas', 'Europe', 'Africa','Asia','Oceania'] prices["continental_region"] = np.select(conditions, regions, default=np.nan)

# There seems to be a problem with a few european countries that figure as Asian. Let's correct this. prices.loc[prices['name'] == 'Belarus','continental_region'] = 'Europe' prices.loc[prices['name'] == 'Ukraine','continental_region'] = 'Europe' prices.loc[prices['name'] == 'Cyprus','continental_region'] = 'Europe'

prices['continental_region'].value_counts()

prices.describe()

num_internet_plansfloat64

avg_price_1gbfloat64

count

230

mean

26.73043478

4.072956522

std

16.46821564

6.44300697

min

0.05

25%

1.0825

50%

2.205

75%

4.3575

max

49.67

fig = px.histogram(prices, x='avg_price_1gb', title='Distribution_avg_price_1gb', marginal='box') fig.show()

# Let's determine outliers for this distribution using a common formula to set minimum and maximum values: min_limit = q1 - 1.5*iqr and max_limit = q3 + 1.5*iqr q1 = prices['avg_price_1gb'].quantile(0.25) q3 = prices['avg_price_1gb'].quantile(0.75) print('Quantile 1: ' + str(q1) + ', '+ 'Quantile 3: ' + str(q3)) # Inter quantile range, IQR iqr = q3 - q1 print('Inter Quantile Range: ' + str(iqr)) # defining limits for outliers min_limit_col1 = q1 - 1.5*iqr max_limit_col1 = q3 + 1.5*iqr print('Min limit: ' + str(min_limit_col1) + ', ' + 'Max limit: ' + str(max_limit_col1))

Quantile 1: 1.0825, Quantile 3: 4.3575
Inter Quantile Range: 3.275
Min limit: -3.8299999999999996, Max limit: 9.27

fig = px.histogram(prices, x='cheapest_1gb_30_days', title='Distribution_cheapest_1gb_30_days', marginal='box') fig.show()

# Let's determine outliers for this distribution using a common formula to set minimum and maximum values: min_limit = q1 - 1.5*iqr and max_limit = q3 + 1.5*iqr q1 = prices['cheapest_1gb_30_days'].quantile(0.25) q3 = prices['cheapest_1gb_30_days'].quantile(0.75) print('Quantile 1: ' + str(q1) + ', '+ 'Quantile 3: ' + str(q3)) # Inter quantile range, IQR iqr = q3 - q1 print('Inter Quantile Range: ' + str(iqr)) # defining limits for outliers min_limit_col2 = q1 - 1.5*iqr max_limit_col2 = q3 + 1.5*iqr print('Min limit: ' + str(min_limit_col2) + ', ' + 'Max limit: ' + str(max_limit_col2))

Quantile 1: 0.2125, Quantile 3: 1.39
Inter Quantile Range: 1.1775
Min limit: -1.55375, Max limit: 3.15625

fig = px.histogram(prices, x='most_expensive_1gb', title='Distribution_most_expensive_1gb', marginal='box') fig.show()

# Let's determine outliers for this distribution using a common formula to set minimum and maximum values: min_limit = q1 - 1.5*iqr and max_limit = q3 + 1.5*iqr q1 = prices['most_expensive_1gb'].quantile(0.25) q3 = prices['most_expensive_1gb'].quantile(0.75) print('Quantile 1: ' + str(q1) + ', '+ 'Quantile 3: ' + str(q3)) # Inter quantile range, IQR iqr = q3 - q1 print('Inter Quantile Range: ' + str(iqr)) # defining limits for outliers min_limit_col3 = q1 - 1.5*iqr max_limit_col3 = q3 + 1.5*iqr print('Min limit: ' + str(min_limit_col3) + ', ' + 'Max limit: ' + str(max_limit_col3))

Quantile 1: 5.9025, Quantile 3: 34.295
Inter Quantile Range: 28.392500000000002
Min limit: -36.68625, Max limit: 76.88375

fig = px.histogram(prices, x='avg_price_1gb_2021', title='Distribution_avg_price_1gb_2021', marginal='box') fig.show()

# Let's determine outliers for this distribution using a common formula to set minimum and maximum values: min_limit = q1 - 1.5*iqr and max_limit = q3 + 1.5*iqr q1 = prices['avg_price_1gb_2021'].quantile(0.25) q3 = prices['avg_price_1gb_2021'].quantile(0.75) print('Quantile 1: ' + str(q1) + ', '+ 'Quantile 3: ' + str(q3)) # Inter quantile range, IQR iqr = q3 - q1 print('Inter Quantile Range: ' + str(iqr)) # defining limits for outliers min_limit_col4 = q1 - 1.5*iqr max_limit_col4 = q3 + 1.5*iqr print('Min limit: ' + str(min_limit_col4) + ', ' + 'Max limit: ' + str(max_limit_col4))

Quantile 1: 1.385, Quantile 3: 5.3325000000000005
Inter Quantile Range: 3.9475000000000007
Min limit: -4.536250000000001, Max limit: 11.25375

fig = px.histogram(prices, x='avg_price_1gb_2020', title='Distribution_avg_price_1gb_2020', marginal='box') fig.show()

# Let's determine outliers for this distribution using a common formula to set minimum and maximum values: min_limit = q1 - 1.5*iqr and max_limit = q3 + 1.5*iqr q1 = prices['avg_price_1gb_2020'].quantile(0.25) q3 = prices['avg_price_1gb_2020'].quantile(0.75) print('Quantile 1: ' + str(q1) + ', '+ 'Quantile 3: ' + str(q3)) # Inter quantile range, IQR iqr = q3 - q1 print('Inter Quantile Range: ' + str(iqr)) # defining limits for outliers min_limit_col5 = q1 - 1.5*iqr max_limit_col5 = q3 + 1.5*iqr print('Min limit: ' + str(min_limit_col5) + ', ' + 'Max limit: ' + str(max_limit_col5))

Quantile 1: 2.85, Quantile 3: 11.035
Inter Quantile Range: 8.185
Min limit: -9.4275, Max limit: 23.3125

# create column with % variation between 2020 and 2021. We can build a function for that def percentage_change(col1,col2): return ((col2 - col1) / col1) prices['%_change_2020_2021'] = round(percentage_change(prices['avg_price_1gb_2020'],prices['avg_price_1gb_2021']),4)

Speed

# Let's inspect our data frame speed

Countryobject

Lebanon1.4%

Canada0.7%

141 others97.9%

Avg (Mbit/s)Ooklafloat64

4.89 - 135.35

Canada

75.18

South Korea

117.95

Netherlands

108.33

Japan

44.05

Norway

134.73

Singapore

67.99

Australia

76.52

Switzerland

82.29

Belgium

58.65

Croatia

86.48

# Checking null values and data types speed.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 144 entries, 0 to 143
Data columns (total 2 columns):
 #   Column              Non-Null Count  Dtype  
---  ------              --------------  -----  
 0   Country             144 non-null    object 
 1   Avg 
(Mbit/s)Ookla  143 non-null    float64
dtypes: float64(1), object(1)
memory usage: 2.4+ KB

speed.columns

# Renaming the column speed.rename({'Avg \n(Mbit/s)Ookla':'avg_speed_mbit/s'}, axis=1, inplace=True)

speed.columns

# Let's see the only null value the second column has speed.loc[speed['avg_speed_mbit/s'].isnull(),]

Countryobject

avg_speed_mbit/sfloat64

Puerto Rico

nan

# Sorry Puerto Rico, but you need to go # drop rows with null values for selected columns only speed.dropna(subset=['avg_speed_mbit/s'],inplace=True)

# No null values in our data set speed.isna().sum()

# The speed value doesn't need to be with decimals speed['avg_speed_mbit/s'] = round(speed['avg_speed_mbit/s'],0)

# Finally, as a good practice, let's get rid of whitespaces at the begining and end speed['Country'] = speed['Country'].str.strip()

# Setting up speed categories: conditions = [ (speed['avg_speed_mbit/s'] <= 12), (speed['avg_speed_mbit/s'] > 12) & (speed['avg_speed_mbit/s'] <= 25), (speed['avg_speed_mbit/s'] > 25) ] values = ['low_speed (12 or less)','medium_speed (12 to 25)','high_speed (over 25)'] speed['speed_category'] = np.select(conditions, values)

speed['speed_category'].value_counts()

speed.shape

speed.nunique()

# After a quick search, the country Lebanon is duplicated speed.loc[speed['Country']=='Lebanon',]

Countryobject

avg_speed_mbit/sfloat64

Lebanon

130

Lebanon

speed.drop_duplicates(subset='Country',inplace=True)

speed.loc[speed['Country']=='Lebanon',]

Countryobject

avg_speed_mbit/sfloat64

Lebanon

Users

users

Country or areaobject

China0.5%

India0.5%

212 others99.1%

Subregionobject

Caribbean9.3%

Western Asia8.4%

20 others82.2%

China

Eastern Asia

India

Southern Asia

United States

Northern America

Indonesia

South-eastern Asia

Brazil

South America

Nigeria

Western Africa

Bangladesh

Southern Asia

Russia

Eastern Europe

Pakistan

Southern Asia

Japan

Eastern Asia

users.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 214 entries, 0 to 213
Data columns (total 5 columns):
 #   Column           Non-Null Count  Dtype 
---  ------           --------------  ----- 
 0   Country or area  214 non-null    object
 1   Subregion        214 non-null    object
 2   Region           214 non-null    object
 3   Internet users   214 non-null    object
 4   Population       212 non-null    object
dtypes: object(5)
memory usage: 8.5+ KB

# Let's now convert numeric columns to numeric type, after doing a little trick we need to be able to do the convertion: delete the commas users['Population'] = pd.to_numeric(users['Population'].str.replace(',',''),errors='coerce') users['Internet users'] = pd.to_numeric(users['Internet users'].str.replace(',',''),errors='coerce')

# Checking null values for the only column that has any. users.loc[users['Population'].isnull(),]

Country or areaobject

Subregionobject

Palestine

Western Asia

197

Jersey

Northern Europe

# Filling the missing values in our DF users.loc[[98,197],'Population' ] = 4803000, 99000

users.loc[(users['Country or area'] == 'Palestine') | (users['Country or area'] == 'Jersey'),]

Country or areaobject

Subregionobject

Palestine

Western Asia

197

Jersey

Northern Europe

# As a good practice, let's get rid of whitespaces at the begining and end in all three columns cols = ['Country or area', 'Subregion', 'Region'] users[cols] = users[cols].apply(lambda x: x.str.strip()) # We can also improve the column names users.rename({'Country or area':'country','Region': 'continent','Internet users':'internet_users','Population':'population'}, axis=1, inplace=True) # We could drop the 'Subregion' column as we won't be using it, just the continent information users.drop(['Subregion'],axis=1, inplace=True)

users['%_internet_users_vs_pop'] = users['internet_users'] / users['population']

users.head()

countryobject

continentobject

China

Asia

India

Asia

United States

Americas

Indonesia

Asia

Brazil

Americas

Load

#load data to google sheets #Create key key = "1ta2iOwjKzIRu7CL60DqzXpVlYhvaTOz7618GmgF8Kts" #Selects the sheet to export to sheet = gc.open_by_key(key).worksheet('prices') sheet1 = gc.open_by_key(key).worksheet('speed') sheet2 = gc.open_by_key(key).worksheet('users') #Clears the sheet and exports data sheet.batch_clear(["A1:L40000"]) set_with_dataframe(sheet, prices) sheet1.batch_clear(["A1:L4000"]) set_with_dataframe(sheet1, speed) sheet2.batch_clear(["A1:L4000"]) set_with_dataframe(sheet2, users)