static_plots.py

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import pickle

# read data from pickle file
data = pd.read_pickle('dataframe.pkl')

pd.set_option('display.max_columns', None)

# Create figure with 4 subplots
fig, axs = plt.subplots(2, 2, figsize = (16, 12))
fig.subplots_adjust(hspace = 0.6, wspace = 0.1)


##### Plot 1: Platform Heights #####
grouped =data.groupby(['Stop name', 'Platform number'])['Platform edge height'].first().reset_index()
height=grouped['Platform edge height']

# plot the plot 1
plot1 = plt.subplot(2,2,1)

#heights =data['Platform edge height']
plot1.hist(height, color='lightgreen', ec='darkgreen', bins=10)
plot1.set_xlabel('platform edge height [cm]')
plot1.set_ylabel('frequency')
plot1.set_title('Distribution of Platform Edge Heights', size=18)


##### Plot 2: Material Overview #####

# Group by 'Stop name' and 'Platform number'
data = data[data != 'Not Applicable']
grouped_data = data.groupby(['Stop name', 'Platform number'])['Edge type'].unique().reset_index()

# Flatten the list of unique edge types
flattened_edge_types = [edge_type for sublist in grouped_data['Edge type'] for edge_type in sublist]

# Calculate the occurrences of each edge type
edge_type_counts = pd.Series(flattened_edge_types).value_counts()
print(f"Here are the absolute numbers of each group: {edge_type_counts}")

# Calculate percentages
total=edge_type_counts.sum()
percentages = [(count / total) * 100 for count in edge_type_counts]

# Create labels with percentages
labels = [f'{edge_type}: {percentage:.1f}%' for edge_type, percentage in zip(edge_type_counts.index, percentages)]

# Rearrange the order of edge types and corresponding labels
edge_type_counts = edge_type_counts.iloc[[0, 2, 1, 3]]  # Change the order as desired
labels = [labels[i] for i in [0, 2, 1, 3]]  # Change the order of labels accordingly

# plot the plot 2
plot2 = plt.subplot(2,2,2)

plot2.pie(edge_type_counts,
        labels= labels,#edge_type_counts.index,
        colors=['lightseagreen', 'skyblue', 'mediumaquamarine', 'cyan', 'seagreen'],
        startangle=90,
        textprops={'fontsize': 8}
        )
plot2.set_title("Edge Types", size=18)


##### PLOT 3 #####
#making categories for bars
min=data['Length platform total'].min()
med=data['Length platform total'].median()
max=data['Length platform total'].max()

min_station=(data[data['Length platform total'] == 19.67])

print("The Station with the minimal platform length is:")
print(min_station)

len_data=[min, med, max]
print("Here you see the minimal, median and the maximal platform length:")
print(len_data) #[19.67, 342.69, 1206.78]


lmax=data[data['Length platform total']==1206.78] 

# Get unique values for 'Stop name', 'Length platform total', and 'Platform number'
stop_name = lmax['Stop name'].unique()[0]
length_platform_total = lmax['Length platform total'].unique()[0]
platform_number = lmax['Platform number'].unique()[0]

print("The maximal platform length is located in:")
label_max=print(f"{stop_name} with {length_platform_total}m on edge {platform_number}")

# category names
x=('minimum', 'median', 'maximum')

# plot the plot 3
plot3 = plt.subplot(2,2,3)

#creating horizontal plot
plot3.barh(x, len_data) #add , label= ???

#setting title
plot3.set_title("Platform Lenghts", size=18)

#naming axes
plot3.set_xlabel("platform length [m]")
plot3.set_ylabel("")

##### Plot 4: Platform Types #####
# Filter out 'Not Applicable' values
data = data[data['Platform type'] != 'Not Applicable']

grouped_data = data.groupby(['Stop name', 'Platform number'])['Platform type'].unique().reset_index()

# Flatten the list of unique platform types
flattened_platform_types = [platform_type for sublist in grouped_data['Platform type'] for platform_type in sublist]

# Calculate the occurrences of each platform type
platform_type_counts = pd.Series(flattened_platform_types).value_counts()
print(f"Here are the absolute numbers of each group: {platform_type_counts}")

# Calculate percentages
total = platform_type_counts.sum()
percentages = [(count / total) * 100 for count in platform_type_counts]

# Rearrange the order of edge types and corresponding labels
platform_type_counts = platform_type_counts.iloc[[0, 2, 1, 3]]  # Change the order as desired
labels = [labels[i] for i in [0, 2, 1, 3]]  # Change the order of labels accordingly


# Create labels with percentages
labels = [f'{platform_type}: {percentage:.1f}%' for platform_type, percentage in zip(platform_type_counts.index, percentages)]

# plot the plot 4
plot4 = plt.subplot(2,2,4)

plot4.pie(platform_type_counts,
        labels=labels,
        colors=['lightseagreen', 'skyblue', 'mediumaquamarine', 'cyan', 'seagreen'],
        startangle=90,
        textprops={'fontsize': 8}
        )
plot4.set_title("Platform Types", size=18)

##### Print the figure #####
plt.show()