python201_workshop_script.py

"""
Python Data Visualization

Data download:
https://figshare.com/ndownloader/articles/1314459/versions/10

Adapted from Data Carpentry:
https://datacarpentry.org/python-ecology-lesson/07-visualization-ggplot-python.html

Accompanying Slides:
https://docs.google.com/presentation/d/1Iy72RxX-J7O5kg-odqqzOw1DijCPijP9yoHmWCkvaeQ/edit?usp=drive_link
"""

# -----------------------------------
# Plotnine (ggplot-style plotting)
# -----------------------------------

import plotnine as p9
import pandas as pd

# Load and clean data
surveys_complete = pd.read_csv('data/surveys.csv')
surveys_complete = surveys_complete.dropna()

# -----------------------------------
# Basic Scatter Plot
# -----------------------------------

surveys_plot = p9.ggplot(
    data=surveys_complete,
    mapping=p9.aes(x='weight', y='hindfoot_length')
)

surveys_plot + p9.geom_point()


# -----------------------------------
# Adding elements step-by-step
# -----------------------------------

surveys_plot = p9.ggplot(
    data=surveys_complete,
    mapping=p9.aes(x='weight', y='hindfoot_length', color='species_id')
)

(
    surveys_plot
    + p9.geom_point()
    + p9.xlab("Weight (g)")
    + p9.scale_x_log10()
    + p9.theme_bw()
    + p9.theme(text=p9.element_text(size=16))
)


# -----------------------------------
# Boxplot
# -----------------------------------

surveys_plot = p9.ggplot(
    data=surveys_complete,
    mapping=p9.aes(x='species_id', y='weight')
)

surveys_plot + p9.geom_boxplot()


# -----------------------------------
# Time Series Line Chart
# -----------------------------------

yearly_counts = surveys_complete.groupby(
    ['year', 'species_id']
)['species_id'].count()

yearly_counts = yearly_counts.reset_index(name='counts')

surveys_plot = p9.ggplot(
    data=yearly_counts,
    mapping=p9.aes(x='year', y='counts', color='species_id')
)

surveys_plot + p9.geom_line()


# -----------------------------------
# Split Plots (Facet Wrap)
# -----------------------------------

surveys_plot = p9.ggplot(
    data=surveys_complete,
    mapping=p9.aes(x='weight', y='hindfoot_length', color='species_id')
)

surveys_plot + p9.geom_point(alpha=0.1) + p9.facet_wrap("plot_id")


# -----------------------------------
# Facet Grid (rows ~ columns)
# -----------------------------------

survey_2000_2001 = surveys_complete[
    surveys_complete["year"].isin([2000, 2001])
]

surveys_plot = p9.ggplot(
    data=survey_2000_2001,
    mapping=p9.aes(x='weight', y='hindfoot_length', color='species_id')
)

surveys_plot + p9.geom_point(alpha=0.1) + p9.facet_grid("year ~ sex")


# -----------------------------------
# Visualization Customizations
# -----------------------------------

surveys_plot = p9.ggplot(
    data=surveys_complete,
    mapping=p9.aes(x='factor(year)')
)

surveys_plot + p9.geom_bar()
surveys_plot + p9.geom_bar() + p9.theme_bw() + p9.theme(
    axis_text_x=p9.element_text(angle=90)
)


# Custom theme
my_custom_theme = p9.theme(
    axis_text_x=p9.element_text(color="grey", size=10, angle=90, hjust=.5),
    axis_text_y=p9.element_text(color="grey", size=10)
)

surveys_plot = p9.ggplot(
    data=surveys_complete,
    mapping=p9.aes(x='factor(year)')
)

surveys_plot + p9.geom_bar() + my_custom_theme


# -----------------------------------
# Saving Plots
# -----------------------------------

my_plot = (
    p9.ggplot(
        data=surveys_complete,
        mapping=p9.aes(
            x='weight',
            y='hindfoot_length',
            color='species_id'
        )
    )
    + p9.geom_point()
)

my_plot.save("my_bar_graph.png", width=10, height=10, dpi=300)


# -----------------------------------
# matplotlib Examples
# -----------------------------------

import matplotlib.pyplot as plt
import numpy as np

surveys_complete = pd.read_csv('data/surveys.csv')
surveys_complete = surveys_complete.dropna()

x = surveys_complete.weight
y = surveys_complete.hindfoot_length

plt.scatter(x, y, s=10, c='black')
plt.show()


# -----------------------------------
# Category-wise coloring
# -----------------------------------

labels, index = np.unique(
    surveys_complete.species_id,
    return_inverse=True
)

scatter = plt.scatter(x, y, s=10, c=index)

plt.legend(
    scatter.legend_elements(num=None)[0],
    labels,
    ncol=6,
    loc='upper left',
    bbox_to_anchor=(-0.05, 1.15)
)

plt.xlabel("Weight (g)")
plt.xscale("log")
plt.show()


# -----------------------------------
# Boxplots with matplotlib
# -----------------------------------

data = []
labels = []

for element in np.unique(surveys_complete.species_id):
    data.append(
        surveys_complete.loc[
            surveys_complete['species_id'] == element,
            'weight'
        ].to_numpy()
    )
    labels.append(element)

plt.boxplot(data, labels=labels)
plt.xlabel("Species ID")
plt.ylabel("Weight distribution")
plt.show()