import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

# Data Source: https://www.kaggle.com/datasets/samuelotiattakorah/agriculture-crop-yield?resource=download

# Load data
df = pd.read_csv("crop_yield.csv")

# Filter for Soybean only
df = df[df["Crop"] == "Soybean"]

# --- Single Boxplot ---
plt.figure(figsize=(6, 6))
sns.boxplot(y="Yield_tons_per_hectare", data=df, color="#D44803", width=0.3)

plt.ylabel("Yield", fontsize=14)
plt.xlabel("")
plt.title("Boxplots with PikBioStat: Soybean Yield", fontsize=16, weight="bold")

# Clean theme (no gridlines, bigger y labels)
sns.despine()  
plt.grid(False)
plt.yticks(fontsize=14)
plt.xticks([])

plt.show()

# --- Group Comparisons (by Fertilizer_Used) ---
plt.figure(figsize=(8, 6))
sns.boxplot(x="Fertilizer_Used", y="Yield_tons_per_hectare", data=df,
            color="#D44803", width=0.5)

plt.ylabel("Yield (tons/ha)", fontsize=14, weight="bold")
plt.xlabel("Fertilizer Used", fontsize=14, weight="bold")
plt.title("Boxplots with PikBioStat: Soybean Yield by Fertilizer Used", fontsize=16, weight="bold")

# Clean theme
sns.despine()
plt.grid(False)
plt.xticks(rotation=45, ha="right", fontsize=12)
plt.yticks(fontsize=14)

plt.show()