Visualisations

Downlod data

Back to the Spell Book

1. Import

import pandas as pd
from pandas import DataFrame
import numpy as np
import matplotlib.pyplot as plt

smash = pd.read_csv("smash_3.csv")
smash.head()

	character	game	regular_fall	fast_fall	weight	walk_speed	run_speed	full_hop_height	short_hop_height	double_hop_height	before	after	expertise
0	mario	Super Mario	1.50	2.400	98	1.155	1.760	36.33	17.54	36.33	3	8	Amateur
1	donkey kong	Donkey Kong	1.63	2.608	127	1.365	1.873	34.00	17.30	35.50	8	10	Expert
2	link	Zelda	1.60	3.040	104	1.247	1.534	27.80	13.38	29.00	10	7	Expert
3	samus	Metroid	1.33	2.168	108	1.115	1.654	37.00	18.00	37.00	6	9	Expert
4	dark samus	Metroid	1.33	2.168	108	1.115	1.654	37.00	18.00	37.00	2	5	Amateur

2. Description

smash.shape

(72, 13)

smash.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 72 entries, 0 to 71
Data columns (total 13 columns):
 #   Column             Non-Null Count  Dtype  
---  ------             --------------  -----  
 0   character          72 non-null     object 
 1   game               72 non-null     object 
 2   regular_fall       72 non-null     float64
 3   fast_fall          72 non-null     float64
 4   weight             72 non-null     int64  
 5   walk_speed         72 non-null     float64
 6   run_speed          72 non-null     float64
 7   full_hop_height    70 non-null     float64
 8   short_hop_height   70 non-null     float64
 9   double_hop_height  70 non-null     float64
 10  before             72 non-null     int64  
 11  after              72 non-null     int64  
 12  expertise          72 non-null     object 
dtypes: float64(7), int64(3), object(3)
memory usage: 7.4+ KB

smash.describe()

	regular_fall	fast_fall	weight	walk_speed	run_speed	full_hop_height	short_hop_height	double_hop_height	before	after
count	72.000000	72.000000	72.00000	72.000000	72.000000	70.000000	70.000000	70.000000	72.000000	72.000000
mean	1.600069	2.546306	96.25000	1.104171	1.806428	33.211864	16.027429	34.440000	6.319444	7.722222
std	0.222869	0.353823	13.93218	0.213718	0.391566	5.112580	2.302452	6.379117	2.582860	1.576430
min	0.980000	1.568000	62.00000	0.620000	1.180000	19.790000	11.260000	19.790000	2.000000	5.000000
25%	1.472500	2.240000	88.00000	0.924000	1.591250	30.885375	14.500000	30.782500	4.000000	7.000000
50%	1.620000	2.592000	95.50000	1.133000	1.722500	33.000000	16.225000	33.500000	6.000000	8.000000
75%	1.770000	2.820000	104.50000	1.259000	1.979750	35.575000	17.482500	36.832500	8.250000	9.000000
max	2.100000	3.360000	135.00000	1.575000	3.850000	50.510000	22.110000	57.350000	10.000000	10.000000

smash["game"] = smash["game"].astype("category")
smash["expertise"] = smash["expertise"].astype("category")

smash.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 72 entries, 0 to 71
Data columns (total 13 columns):
 #   Column             Non-Null Count  Dtype   
---  ------             --------------  -----   
 0   character          72 non-null     object  
 1   game               72 non-null     category
 2   regular_fall       72 non-null     float64 
 3   fast_fall          72 non-null     float64 
 4   weight             72 non-null     int64   
 5   walk_speed         72 non-null     float64 
 6   run_speed          72 non-null     float64 
 7   full_hop_height    70 non-null     float64 
 8   short_hop_height   70 non-null     float64 
 9   double_hop_height  70 non-null     float64 
 10  before             72 non-null     int64   
 11  after              72 non-null     int64   
 12  expertise          72 non-null     category
dtypes: category(2), float64(7), int64(3), object(1)
memory usage: 7.9+ KB

smash.describe(include = "category")

	game	expertise
count	72	72
unique	30	2
top	Super Mario	Expert
freq	10	45

Rename columns.

smash = smash.rename(columns = {"full_hop_height": "FULL_HOP_HEIGHT_RENAME"})
smash.head()

	character	game	regular_fall	fast_fall	weight	walk_speed	run_speed	FULL_HOP_HEIGHT_RENAME	short_hop_height	double_hop_height	before	after	expertise
0	mario	Super Mario	1.50	2.400	98	1.155	1.760	36.33	17.54	36.33	3	8	Amateur
1	donkey kong	Donkey Kong	1.63	2.608	127	1.365	1.873	34.00	17.30	35.50	8	10	Expert
2	link	Zelda	1.60	3.040	104	1.247	1.534	27.80	13.38	29.00	10	7	Expert
3	samus	Metroid	1.33	2.168	108	1.115	1.654	37.00	18.00	37.00	6	9	Expert
4	dark samus	Metroid	1.33	2.168	108	1.115	1.654	37.00	18.00	37.00	2	5	Amateur

3. Visualisations

3.1 Histogram

import seaborn as sns

sns.histplot(x = smash["run_speed"])

<AxesSubplot:xlabel='run_speed', ylabel='Count'>

Use ";" at the end to remove the AxesSubplot. Change the number of bins.

sns.histplot(x = smash["run_speed"], bins = 20);

histogram = sns.histplot(x = smash["run_speed"], bins = 20, stat = "probability", color = "purple");

histogram.set_xlabel("Run Speed", fontsize = 15);
histogram.set_ylabel("Probability", fontsize = 15);

3.2 Column

sns.barplot(x = "expertise", y = "short_hop_height", data = smash, color = "green");

count = sns.countplot(x = "expertise", data = smash, color = "red");

count.set_xlabel("Expertise", fontsize = 15);
count.set_ylabel("Number", fontsize = 15);

bar = sns.barplot(x = "expertise", y = "short_hop_height", data = smash, color = "blue", estimator = sum);

bar.set_xlabel("Expertise", fontsize = 15);
bar.set_ylabel("Total Short Hop Height", fontsize = 15);

bar = sns.barplot(x = "expertise", y = "short_hop_height", data = smash, color = "lightblue", estimator = max);

bar.set_xlabel("Expertise", fontsize = 15);
bar.set_ylabel("Maximum Short Hop Height", fontsize = 15);

Alternatively from numpy import me

import numpy as np
bar = sns.barplot(x = "expertise", y = "short_hop_height", data = smash, color = "lightblue", estimator = np.median);

bar.set_xlabel("Expertise", fontsize = 15);
bar.set_ylabel("Median Short Hop Height", fontsize = 15);

Convert to a horizontal orientation.

bar = sns.barplot(x = "short_hop_height", y = "expertise", data = smash, color = "purple", estimator = np.median);

bar.set_ylabel("Expertise", fontsize = 15);
bar.set_xlabel("Median Short Hop Height", fontsize = 15);

3.3 Boxplot

sns.boxplot(x = smash["run_speed"]);

box = sns.boxplot(x = smash["run_speed"], showmeans = True, color = "yellow");

box.set_xlabel("Run Speed", fontsize = 15);

3.4 Line

Just for illustration. Does not make sense here.

line = sns.lineplot(x = "expertise", y = "double_hop_height", hue = "walk_speed", data = smash);

line.set_xlabel("Expertise", fontsize = 15);
line.set_ylabel("Double Hop Height", fontsize = 15);

line.legend(title = "Walk Speed", title_fontsize = "10");

3.5 Scatter plot

scatter = sns.scatterplot(x = "run_speed",
                          y = "walk_speed",
                          hue = "double_hop_height",
                          data = smash);

scatter.set_xlabel("Run Speed", fontsize = 20);
scatter.set_ylabel("Walk Speed", fontsize = 20);

scatter.legend(title = "Houdle Hop height", title_fontsize = "10");

Show distributions with the scatter plot.

sns.jointplot(x = "run_speed", y = "walk_speed", data = smash);

sns.jointplot(x = "run_speed", y = "walk_speed", data = smash, kind = "reg");

sns.jointplot(x = "run_speed", y = "walk_speed", data = smash, kind = "hex");

sns.jointplot(x = "run_speed", y = "walk_speed", data = smash, kind = "kde", color = "purple");

3.6 Violin plot

sns.catplot(x = "expertise", y = "run_speed", kind = "violin", data = smash);