Python Project 5 - K-Means Clustering

# import libraries and K-Means function import pandas as pd from sklearn.cluster import KMeans import numpy as np import seaborn as sns import matplotlib.pyplot as plt

# Create dataframe for data df = pd.read_csv('cereal.csv') df

nameobject

100% Bran1.3%

100% Natural Bran1.3%

75 others97.4%

mfrobject

K29.9%

G28.6%

5 others41.6%

Cream of Wheat (Quick)

Crispix

Crispy Wheat & Raisins

Double Chex

Froot Loops

Frosted Flakes

Frosted Mini-Wheats

Fruit & Fibre Dates; Walnuts; and Oats

Fruitful Bran

Fruity Pebbles

# drop features we aren't using or create a new dataframe with only the features we want df = df[["name", "mfr", "sugars", "carbo", "rating"]] df.head(10)

nameobject

100% Bran10%

100% Natural Bran10%

8 others80%

mfrobject

K30%

R20%

4 others50%

100% Bran

100% Natural Bran

All-Bran

All-Bran with Extra Fiber

Almond Delight

Apple Cinnamon Cheerios

Apple Jacks

Basic 4

Bran Chex

Bran Flakes

# Check to see if any instances have NaN for entries using .info df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 77 entries, 0 to 76
Data columns (total 5 columns):
 #   Column  Non-Null Count  Dtype  
---  ------  --------------  -----  
 0   name    77 non-null     object 
 1   mfr     77 non-null     object 
 2   sugars  77 non-null     int64  
 3   carbo   77 non-null     float64
 4   rating  77 non-null     float64
dtypes: float64(2), int64(1), object(2)
memory usage: 3.1+ KB

# Check to see if any cereals have negative values for sugars or carbohydrates; if 2 or less cereals, drop those # data instances; otherwise replace negative values with 0 neg_sugars = df[df.sugars < 0] print(neg_sugars) neg_carbo = df[df.carbo < 0] print(neg_carbo)

              name mfr  sugars  carbo     rating
57  Quaker Oatmeal   Q      -1   -1.0  50.828392
              name mfr  sugars  carbo     rating
57  Quaker Oatmeal   Q      -1   -1.0  50.828392

# Address negative values x_sugars = df[df.sugars >= 0] print(len(x_sugars)) x_carbo = df[df.carbo >= 0] print(len(x_carbo))

76
76

# set background grid for plots sns.set_style('whitegrid')

# Plot number of products per manufacturer sns.histplot(data = df, x = "mfr")

# There is only one cereal from American Home Foods Company so we drop that data sample # Type df[df.mfr == 'A'] and it will give you all data from this manufacturer. # Make sure you set df = drop command; use command to make sure it was deleted df = df[df.mfr != "A"] # need to have no space after ! or else it won't work sns.histplot (data = df, x = "mfr")

# For plots we would like the name of manufacturer instead of just "N" or "Q" # Use .map or .apply df["mfr_name"] = df["mfr"].map ({"G": "General Mills", "K": "Kelloggs", "N": "Nabisco", "P": "Post", "Q": "Quaker Oats", "R": "Ralston Purina"}) # Then drop 'mfr' column df.drop(columns = ["mfr"])

nameobject

100% Bran1.3%

100% Natural Bran1.3%

74 others97.4%

sugarsint64

-1 - 15

100% Bran

100% Natural Bran

All-Bran

All-Bran with Extra Fiber

Almond Delight

Apple Cinnamon Cheerios

Apple Jacks

Basic 4

Bran Chex

Bran Flakes

# Plot to see which manufacturer has highest rated cereals sns.relplot (x = "mfr_name", y = "rating", data = df, hue = "mfr_name").set(title = "Manufacturers and their ratings") plt.xticks(rotation = 45)

# Which cereal is rated highest? df[df.rating > 93] # don't know why the mfr column is still present even thought I droppped it in a previous question

nameobject

mfrobject

All-Bran with Extra Fiber

# Look at sugars per brand by plotting sns.catplot (x = "mfr_name", y = "sugars", data = df, hue = "mfr_name").set(title = "Sugars by brand") plt.xticks(rotation = 45) plt.legend() # Don't know how to move the legend away from being in front of 2 points, sorry

# Get data for clustering x = df.sugars.values n = len(x) x = np.reshape(x, (n, 1))

# Form model, fit data and print out cluster centers km = KMeans (n_clusters = 3, init = "random", random_state = 0) y_km = km.fit(x) print (f"Cluster centers for sugars clusters:") print(km.cluster_centers_)

Cluster centers for sugars clusters:
[[ 1.84      ]
 [ 6.84      ]
 [12.03846154]]

# Add column to dataframe for this clusters, say sugars_clusters df["sugars_clusters"] = km.labels_ df.head()

nameobject

mfrobject

100% Bran

100% Natural Bran

All-Bran

All-Bran with Extra Fiber

Almond Delight

# Plot clusters sns.catplot(x = "sugars_clusters", y = "sugars", data = df)

# Determine which cluster number corresponds to lowest, middle and highest level and create a new # column in dataframe using .map df["sugars_clusters_label"] = df["sugars_clusters"].map ({0: "low sugar", 1:"medium sugar", 2: "high sugar"}) df.head(10)

nameobject

100% Bran10%

100% Natural Bran10%

8 others80%

mfrobject

K30%

R20%

4 others50%

100% Bran

100% Natural Bran

All-Bran

All-Bran with Extra Fiber

Almond Delight

Apple Cinnamon Cheerios

Apple Jacks

Basic 4

Bran Chex

Bran Flakes

# How are cereals distributed among the 3 levels? print (f"The cereals are distributed evenly among the 3 levels")

The cereals are distributed evenly among the 3 levels

# Which cereals have the highest sugar levels df[df.sugars_clusters == 2]

nameobject

Apple Cinnamon Cheerios3.8%

Apple Jacks3.8%

24 others92.3%

mfrobject

G34.6%

K30.8%

3 others34.6%

Apple Cinnamon Cheerios

Apple Jacks

Cap'n'Crunch

Cocoa Puffs

Corn Pops

Count Chocula

Crispy Wheat & Raisins

Froot Loops

Frosted Flakes

Fruit & Fibre Dates; Walnuts; and Oats

# Which cereals have the lowest sugar levels df[df.sugars_clusters == 0]

nameobject

All-Bran with Extra Fiber4%

Cheerios4%

23 others92%

mfrobject

K28%

G24%

4 others48%

All-Bran with Extra Fiber

Cheerios

Corn Chex

Corn Flakes

Cream of Wheat (Quick)

Crispix

Grape-Nuts

Great Grains Pecan

Kix

Nutri-grain Wheat

# If you eat a particular cereal like Apple Jacks, Froot Loops, etc. what cluster is it in? my_cereal = 'Fruity Pebbles' print(f"The data instance and sugar cluster for {my_cereal} is {df[df.name == my_cereal].sugars_clusters}")

The data instance and sugar cluster for Fruity Pebbles is 29    2
Name: sugars_clusters, dtype: int32

sns.catplot (x = "mfr_name", y = "carbo", data = df, hue = "mfr_name") plt.xticks(rotation = 45)

X = df.carbo.values n = len(X) X = np.reshape (X, (n, 1)) km = KMeans (n_clusters = 3, init = "random", random_state = 0) y_km = km.fit(X) print (f"Clusters centers for carbo clusters:") print(km.cluster_centers_)

Clusters centers for carbo clusters:
[[19.52173913]
 [ 8.8       ]
 [13.86842105]]

df["carbo_clusters"] = km.labels_ df.head(10)

nameobject

100% Bran10%

100% Natural Bran10%

8 others80%

mfrobject

K30%

R20%

4 others50%

100% Bran

100% Natural Bran

All-Bran

All-Bran with Extra Fiber

Almond Delight

Apple Cinnamon Cheerios

Apple Jacks

Basic 4

Bran Chex

Bran Flakes

sns.catplot(x = "carbo_clusters", y = "carbo", data = df)

df["carbo_clusters_label"] = df["carbo_clusters"].map ({0: "low carbs", 1:"medium carbs", 2: "high carbs"}) df.head(10)

nameobject

100% Bran10%

100% Natural Bran10%

8 others80%

mfrobject

K30%

R20%

4 others50%

100% Bran

100% Natural Bran

All-Bran

All-Bran with Extra Fiber

Almond Delight

Apple Cinnamon Cheerios

Apple Jacks

Basic 4

Bran Chex

Bran Flakes

# High carbs and low sugar df[df.sugars_clusters == 0] df[df.carbo_clusters == 2] # When I do just the low sugar version it shows the low version, # But when I add the high carb version it prioritizes that and disregards the first part

nameobject

Almond Delight2.6%

Bran Chex2.6%

36 others94.7%

mfrobject

G34.2%

K21.1%

4 others44.7%

Almond Delight

Bran Chex

Bran Flakes

Cap'n'Crunch

Cinnamon Toast Crunch

Clusters

Cocoa Puffs

Corn Pops

Count Chocula

Frosted Flakes

# Low carbs and low sugar df[df.sugars_clusters == 0] df[df.carbo_clusters == 0] # I don't know why it shows the medium sugar clusters

nameobject

Basic 44.3%

Cheerios4.3%

21 others91.3%

mfrobject

K39.1%

G26.1%

3 others34.8%

Basic 4

Cheerios

Corn Chex

Corn Flakes

Cream of Wheat (Quick)

Crispix

Double Chex

Grape-Nuts

Just Right Crunchy Nuggets

Just Right Fruit & Nut