Neoadjuvant Therapy Classification Project

Project Overview

Table of Contents

Data Description

Setup

Requirements

pip install pandas numpy matplotlib seaborn scikit-learn imbalanced-learn xgboost tensorflow

File Structure

/path/to/project
│
├── Neoadj_base_luminaux - anonymise.xls # Raw dataset
├── perso_mec_imp.csv # tailored imputed dataset
├── data_preprocessing.py # Data processing and feature engineering script
├── model_building.py # Machine learning model training and evaluation script
├── neural_network.py # Neural network model implementation and training script
├── README.md # This README file
└── requirements.txt # List of Python dependencies

Data Preparation

# Load the dataset
df = pd.read_excel('/path/to/Neoadj_base_luminaux - anonymise.xls')

# Data Cleaning and Feature Selection
df1 = df.drop([...], axis=1)  # Drop irrelevant columns

# Encoding categorical variables
sata_n_map = {'A': 0, 'B': 1, 'C': 2, 'D': 3}
df1['sata_n'] = df1['sata_n'].replace(sata_n_map)

# Filtering data based on 'chim_typ'
filtered_df = df1.loc[(df1['chim_typ'].isin([1, 2, 5]))]
grouped_df = filtered_df.groupby('anap_kc').apply(lambda x: x.reset_index(drop=True))

# Load additional features
X_df = pd.read_csv('/path/to/perso_mec_imp.csv')

Feature Selection

Machine Learning Models and evaluation