kasunUdayanga/Tea_yield_6_features

Tea Yield Prediction Dataset (6 Features)

📋 Quick Info

• Samples: 53,264
• Features: 6
• Task: Regression (predict tea yield)
• Type: Synthetic (realistic simulation)

🎯 Purpose

Simple dataset for machine learning beginners to practice:

• Data preprocessing (missing values, outliers)
• Feature engineering
• Regression modeling
• Model evaluation

📊 Features

#	Feature	Description	Range
1	rainfall_mm	Annual rainfall in mm	10-350
2	temperature_avg	Average temperature in °C	18-40
3	soil_ph	Soil acidity/alkalinity	4.5-6.0
4	fertilizer_kg_ha	Fertilizer used in kg/ha	200-500
5	plant_age_years	Age of tea plants in years	2-25
6	altitude_m	Altitude in meters	500-2000
Target	yield_kg_ha	Tea yield in kg/ha	1000-5000

🚀 Quick Start

import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestRegressor
from sklearn.metrics import mean_absolute_error, r2_score

# Load data
df = pd.read_csv('tea_yield_6_features.csv')

# Handle missing values
df = df.fillna(df.median())

# Split data
X = df.drop('yield_kg_ha', axis=1)
y = df['yield_kg_ha']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# Train model
model = RandomForestRegressor()
model.fit(X_train, y_train)

# Evaluate
predictions = model.predict(X_test)
print(f"R²: {r2_score(y_test, predictions):.3f}")
print(f"MAE: {mean_absolute_error(y_test, predictions):.2f}")