Pandas Auto Prep

A powerful pandas accessor that automates 80-90% of standard tabular data preprocessing tasks with a single command.

Installation

pip install pandas-auto-prep

Quick Start

import pandas as pd
import pandas_auto_prep  # This registers the .prep accessor

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

# Preprocess with one line!
X_train, X_test, y_train, y_test = df.prep.auto(target='target_column')

Features

The .prep.auto() method performs the following preprocessing steps automatically:

Parameters

df.prep.auto(
    target='target_column',      # Target column name (optional)
    test_size=0.2,               # Test split ratio
    random_state=42,             # Random seed
    cardinality_thresh=10,       # One-Hot vs Label Encoding threshold
    missing_thresh=0.5,          # Drop columns with >50% missing
    skewness_thresh=1.0,         # Skewness threshold for log transform
    verbose=True                 # Print progress
)

Usage Examples

With Target (Supervised Learning)

import pandas as pd
import pandas_auto_prep

df = pd.read_csv("customer_churn.csv")
X_train, X_test, y_train, y_test = df.prep.auto(target='churn')

# Ready for model training!
from sklearn.ensemble import RandomForestClassifier
model = RandomForestClassifier()
model.fit(X_train, y_train)

Without Target (Data Cleaning Only)

import pandas as pd
import pandas_auto_prep

df = pd.read_csv("messy_data.csv")
clean_df = df.prep.auto()  # Returns fully preprocessed DataFrame

Access Fitted Objects

# After running .prep.auto()
label_encoders = df.prep.get_label_encoders()  # For inverse transforms
scaler = df.prep.get_scaler()                   # Fitted StandardScaler

Requirements

• Python >= 3.8
• pandas >= 1.3.0
• numpy >= 1.20.0
• scikit-learn >= 1.0.0

License

MIT License - see LICENSE for details.

Author

Aarush Gupta