Cocoa Curations: Series Filtering with Chocolate Ratings¶

Let's Start¶

In [1]:
# Importing neccessary libraries

import pandas as pd
import numpy as np

import matplotlib.pyplot as plt
import seaborn as sns
In [2]:
# Loading the dataset
data = pd.read_csv('flavors_of_cocoa.csv')
In [3]:
data.head()
Out[3]:
Company \n(Maker-if known) Specific Bean Origin\nor Bar Name REF Review\nDate Cocoa\nPercent Company\nLocation Rating Bean\nType Broad Bean\nOrigin
0 A. Morin Agua Grande 1876 2016 63% France 3.75 Sao Tome
1 A. Morin Kpime 1676 2015 70% France 2.75 Togo
2 A. Morin Atsane 1676 2015 70% France 3.00 Togo
3 A. Morin Akata 1680 2015 70% France 3.50 Togo
4 A. Morin Quilla 1704 2015 70% France 3.50 Peru
In [4]:
data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1795 entries, 0 to 1794
Data columns (total 9 columns):
 #   Column                            Non-Null Count  Dtype  
---  ------                            --------------  -----  
 0   Company 
(Maker-if known)         1795 non-null   object 
 1   Specific Bean Origin
or Bar Name  1795 non-null   object 
 2   REF                               1795 non-null   int64  
 3   Review
Date                       1795 non-null   int64  
 4   Cocoa
Percent                     1795 non-null   object 
 5   Company
Location                  1795 non-null   object 
 6   Rating                            1795 non-null   float64
 7   Bean
Type                         1794 non-null   object 
 8   Broad Bean
Origin                 1794 non-null   object 
dtypes: float64(1), int64(2), object(6)
memory usage: 126.3+ KB
In [5]:
# Note the column have `\n` in their names so run the following line to know the column names

data.columns
Out[5]:
Index(['Company \n(Maker-if known)', 'Specific Bean Origin\nor Bar Name',
       'REF', 'Review\nDate', 'Cocoa\nPercent', 'Company\nLocation', 'Rating',
       'Bean\nType', 'Broad Bean\nOrigin'],
      dtype='object')
In [6]:
data.describe()
Out[6]:
REF Review\nDate Rating
count 1795.000000 1795.000000 1795.000000
mean 1035.904735 2012.325348 3.185933
std 552.886365 2.927210 0.478062
min 5.000000 2006.000000 1.000000
25% 576.000000 2010.000000 2.875000
50% 1069.000000 2013.000000 3.250000
75% 1502.000000 2015.000000 3.500000
max 1952.000000 2017.000000 5.000000
1. Which of the following methods is used to find the top 5 chocolates based on their Rating?¶
In [ ]:
# Identify the top 5 chocolates based on their rating
2. Identify Low-Rated Chocolate Bars¶
In [7]:
low_rated_count = data[data['Rating'] < 2].count()
low_rated_count
Out[7]:
Company \n(Maker-if known)           17
Specific Bean Origin\nor Bar Name    17
REF                                  17
Review\nDate                         17
Cocoa\nPercent                       17
Company\nLocation                    17
Rating                               17
Bean\nType                           17
Broad Bean\nOrigin                   17
dtype: int64
3. High Cocoa Percent Chocolates¶
In [9]:
data['Cocoa\nPercent'] = data['Cocoa\nPercent'].str.rstrip('%').astype(float)
high_cocoa_chocolates = data[data['Cocoa\nPercent'] > 70]
high_cocoa_chocolates
Out[9]:
Company \n(Maker-if known) Specific Bean Origin\nor Bar Name REF Review\nDate Cocoa\nPercent Company\nLocation Rating Bean\nType Broad Bean\nOrigin
26 Adi Vanua Levu, Toto-A 705 2011 80.0 Fiji 3.25 Trinitario Fiji
27 Adi Vanua Levu 705 2011 88.0 Fiji 3.50 Trinitario Fiji
28 Adi Vanua Levu, Ami-Ami-CA 705 2011 72.0 Fiji 3.50 Trinitario Fiji
32 Akesson's (Pralus) Bali (west), Sukrama Family, Melaya area 636 2011 75.0 Switzerland 3.75 Trinitario Indonesia
33 Akesson's (Pralus) Madagascar, Ambolikapiky P. 502 2010 75.0 Switzerland 2.75 Criollo Madagascar
... ... ... ... ... ... ... ... ... ...
1778 Zotter Raw 1205 2014 80.0 Austria 2.75
1779 Zotter Bocas del Toro, Cocabo Co-op 801 2012 72.0 Austria 3.50 Panama
1784 Zotter El Oro 879 2012 75.0 Austria 3.00 Forastero (Nacional) Ecuador
1785 Zotter Huiwani Coop 879 2012 75.0 Austria 3.00 Criollo, Trinitario Papua New Guinea
1786 Zotter El Ceibo Coop 879 2012 90.0 Austria 3.25 Bolivia

795 rows × 9 columns

4. Count Chocolates Above Average Rating¶
In [10]:
mean_rating = data['Rating'].mean()
above_avg_chocolates = data[data['Rating'] > mean_rating].count()
above_avg_chocolates
Out[10]:
Company \n(Maker-if known)           1005
Specific Bean Origin\nor Bar Name    1005
REF                                  1005
Review\nDate                         1005
Cocoa\nPercent                       1005
Company\nLocation                    1005
Rating                               1005
Bean\nType                           1004
Broad Bean\nOrigin                   1005
dtype: int64

Just for Exploration¶

Do you enjoy a bit of a bitter kick in your chocolate? I've heard the more cocoa solids, the more intense the bitterness. But, does a higher cocoa percentage translate into a better rating, or do our taste buds crave something different? Let’s dive in and see how the cocoa percentage really influences expert ratings. You might be surprised by what makes a chocolate bar truly elite!

In [ ]:
# Create scatter plot for cocoa percentage vs rating
plt.figure(figsize=(16, 12))
sns.scatterplot(data=data, x='Cocoa\nPercent', y='Rating', hue='Cocoa\nPercent', palette='coolwarm')
plt.title('Cocoa Percentage vs. Chocolate Rating')
plt.xlabel('Cocoa Percent')
plt.ylabel('Rating')
plt.show()
5. Identify Beans with High Cocoa and High Rating!¶
In [12]:
filtered_chocolates_series = data[(data['Cocoa\nPercent'] > 60) & (data['Rating'] >= 4.0)]['Specific Bean Origin\nor Bar Name']
6. Count Extreme Chocolates¶
In [13]:
extreme_chocolates = data[(data['Rating'] < 2) | (data['Cocoa\nPercent'] > 90)].count()
extreme_chocolates
Out[13]:
Company \n(Maker-if known)           34
Specific Bean Origin\nor Bar Name    34
REF                                  34
Review\nDate                         34
Cocoa\nPercent                       34
Company\nLocation                    34
Rating                               34
Bean\nType                           34
Broad Bean\nOrigin                   34
dtype: int64
7. What is the correct syntax to filter chocolates with a rating greater than 4.5 and a cocoa percentage less than 70%?¶
In [ ]:
# Provide the correct syntax to filter chocolates with a rating greater than `4.5` and a cocoa percentage less than `70%`.
8. Count High-Rated Venezuelan Chocolates¶
In [16]:
venezuela_chocolates = data[(data['Broad Bean\nOrigin'] == 'Venezuela') & (data['Rating'] > 3.5)].count()
venezuela_chocolates
Out[16]:
Company \n(Maker-if known)           54
Specific Bean Origin\nor Bar Name    54
REF                                  54
Review\nDate                         54
Cocoa\nPercent                       54
Company\nLocation                    54
Rating                               54
Bean\nType                           54
Broad Bean\nOrigin                   54
dtype: int64

Just for Exploration¶

Ever wondered which country produces the finest chocolate? Is it the lush rainforests of South America, or perhaps the exotic plantations of Africa? This visualization uncovers the countries that consistently produce the highest-rated chocolate bars. You might find some surprising contenders that produce top-quality bars recognized by chocolate connoisseurs worldwide!

In [17]:
# Group by 'Company Location' and calculate the mean rating
top_countries = data.groupby('Company\nLocation')['Rating'].mean().sort_values(ascending=False).head(10)

# Plot the top 10 countries with highest-rated chocolates
plt.figure(figsize=(10, 6))
sns.barplot(x=top_countries.values, y=top_countries.index, palette='viridis')
plt.title('Top 10 Countries Producing Highest-Rated Chocolate Bars')
plt.xlabel('Average Rating')
plt.ylabel('Company Location')
plt.show()

/tmp/ipykernel_18/2659750790.py:6: FutureWarning: 

Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `y` variable to `hue` and set `legend=False` for the same effect.

  sns.barplot(x=top_countries.values, y=top_countries.index, palette='viridis')
No description has been provided for this image
9. Recent High-Rated Bars¶
In [18]:
recent_high_rated_count = data[(data['Review\nDate'] > 2015) & (data['Rating'] >= 4)].count()
recent_high_rated_count
Out[18]:
Company \n(Maker-if known)           9
Specific Bean Origin\nor Bar Name    9
REF                                  9
Review\nDate                         9
Cocoa\nPercent                       9
Company\nLocation                    9
Rating                               9
Bean\nType                           9
Broad Bean\nOrigin                   9
dtype: int64
10. Most Common Bean Origin for Highly Rated Chocolates¶
In [ ]:
top_rated_common_origin = ...
11. Average Rating by Company Location¶
In [ ]:
# Get unique company locations
unique_locations = ...
In [21]:
# Get unique company locations
unique_locations = data['Company\nLocation'].unique()

#Initialize a dictionary to store results
location_stats = {}

#Calculate stats for each location
for location in unique_locations:
    location_data = data[data['Company\nLocation'] == location]
    count = len(location_data)
    if count >= 10:  # Only consider locations with at least 10 reviews
        mean_rating = location_data['Rating'].mean()
        location_stats[location] = mean_rating

#Convert results to a Series and sort
avg_rating_by_location = pd.Series(location_stats).sort_values(ascending=False)
avg_rating_by_location
Out[21]:
Vietnam        3.409091
Brazil         3.397059
Australia      3.357143
Guatemala      3.350000
Switzerland    3.342105
Italy          3.325397
Scotland       3.325000
Canada         3.324000
Denmark        3.283333
Spain          3.270000
France         3.251603
Austria        3.240385
Hungary        3.204545
New Zealand    3.191176
Germany        3.178571
Venezuela      3.175000
Colombia       3.173913
U.S.A.         3.154123
Madagascar     3.147059
Belgium        3.093750
Japan          3.088235
U.K.           3.054688
Ecuador        3.009259
Peru           2.897059
dtype: float64
In [ ]:
# Convert results to a Series and sort
avg_rating_by_location = ...