durham_trees_analysis.py

# -*- coding: utf-8 -*-
"""durham_trees_analysis

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/drive/1OjlRC7F_UICGJM59jzSoy1o2crZxqXCl
"""

# Save this code as a file named 'durham_trees_analysis.py'
# Import display function and Image class
from IPython.display import display, Image
# Install required packages
import subprocess
import seaborn as sns
import matplotlib.pyplot as plt
# Install datasets package
subprocess.run(["pip", "install", "datasets", "geopandas", "seaborn", "matplotlib", "mplcursors", "pandas"])

# Import libraries
from datasets import load_dataset
import seaborn as sns
import matplotlib.pyplot as plt
import mplcursors
import pandas as pd
import geopandas as gpd

# Load dataset
dataset = load_dataset("Ziyuan111/DurhamTrees")

# Convert dataset to pandas DataFrame
df = pd.DataFrame(dataset['train'])

# Interactive scatter plot with seaborn and mplcursors
def plot_interactive_scatter():
    scatter = sns.scatterplot(data=df, x='X', y='Y', hue='species')
    plt.xlabel('X')
    plt.ylabel('Y')
    plt.legend([],[], frameon=False)
    cursor = mplcursors.cursor(hover=True)

    @cursor.connect("add")
    def on_add(sel):
        sel.annotation.set_text(df.iloc[sel.target.index]['species'])
    plt.savefig('interactive_scatter.png')
    plt.show()
    display(Image('interactive_scatter.png'))

# Plot tree planting sites with geopandas
def plot_tree_sites():
    gdf = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.X, df.Y))
    durham_center = {'x': -78.898619, 'y': 35.994033}  # Durham, NC coordinates
    fig, ax = plt.subplots(figsize=(10, 10))
    gdf.plot(ax=ax, color='green')
    buffer = 0.05
    ax.set_xlim([durham_center['x'] - buffer, durham_center['x'] + buffer])
    ax.set_ylim([durham_center['y'] - buffer, durham_center['y'] + buffer])
    ax.set_title('Tree Planting Sites in Durham')
    ax.set_xlabel('Longitude')
    ax.set_ylabel('Latitude')
    ax.set_axis_off()
    plt.savefig('tree_sites.png')
    plt.show()
    display(Image('tree_sites.png'))

# Print correlation matrix
def print_correlation_matrix():
    correlation_matrix = df[['vocs', 'coremoved_ozperyr', 'coremoved_dolperyr']].corr()
    print(correlation_matrix)

# Call the functions
plot_interactive_scatter()
plot_tree_sites()
print_correlation_matrix()


# Show the plot
plt.show()