3
\n#### Group Members:Thaddeus Chong Zhuo Liang, Wai Hpone
","metadata":{}},{"cell_type":"markdown","source":"###1. Introduction
","metadata":{}},{"cell_type":"markdown","source":"**Purpose of the Assignment**\n\nThe purpose of this assignment is to leverage advanced data analytics techniques to address a critical challenge in the financial sector: predicting loan defaults. Participants are tasked with using a comprehensive dataset provided by Home Credit, which includes a myriad of tables from both internal and external sources. This dataset is structured to facilitate the application of both statistical and machine learning methods to develop predictive models. The assignment enhances proficiency in managing large-scale data, performing exploratory data analysis, conducting feature engineering, and constructing robust predictive models. It stresses the importance of model stability over time, reflecting the real-world necessity for financial models to retain predictive accuracy amid evolving economic conditions and borrower behaviors. This exercise aims to deepen participants' understanding of financial risk assessment, encouraging the development of reliable, sustainable, and effective predictive models.\n\n**Problem to be Addressed by the Case Study**\n\nThis case study addresses the significant problem of predicting client defaults on loans, a major issue for consumer finance providers, particularly when it comes to serving people with little or no credit history. The dataset provided is extensive, including various files that represent different dimensions of client information, such as basic personal data, previous application history, and credit bureau data. These data are crucial for developing predictive models that can assess the likelihood of default among potential borrowers. The challenge requires not only assessing the accuracy of these predictive models but also ensuring their stability over time. This dual focus is crucial because models need to adapt to changing borrower behaviors and broader economic shifts without frequent retraining. Stability in predictive performance is essential to prevent financial losses and maintain the reliability and credibility of the financial assessment models. Through this case study, participants are encouraged to explore non-traditional data sources and advanced analytical techniques to offer more inclusive financial services and improve the accuracy and reliability of loan default predictions.","metadata":{}},{"cell_type":"markdown","source":"###2. Data Collection
","metadata":{}},{"cell_type":"markdown","source":"### Importing Libraries","metadata":{}},{"cell_type":"code","source":"!pip install dash\n!pip install pyngrok\n!pip install --upgrade seaborn pandas","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:01:14.919621Z","iopub.execute_input":"2024-06-01T09:01:14.919967Z","iopub.status.idle":"2024-06-01T09:02:08.738800Z","shell.execute_reply.started":"2024-06-01T09:01:14.919936Z","shell.execute_reply":"2024-06-01T09:02:08.737655Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Importing necessary libraries for data manipulation, visualization, and machine learning\n\nimport gc # Import the garbage collection module to manually free up memory during runtime\nimport geopandas as gpd # Import GeoPandas for handling geospatial data\nimport joblib # Import joblib for saving and loading machine learning models efficiently\nimport lightgbm as lgb # Import LightGBM, a gradient boosting framework that uses tree-based learning algorithms\nimport matplotlib.pyplot as plt # Import matplotlib for creating static, animated, and interactive visualizations\nimport missingno as msno # Import missingno for visualizing missing data\nimport numpy as np # Import NumPy for powerful numerical computations and array handling\nimport os # Import os for interacting with the operating system, such as reading or writing files\nimport pandas as pd # Import pandas for data manipulation and analysis\nimport plotly.express as px # Import Plotly Express for creating interactive plots\nimport polars as pl # Import Polars for fast and efficient data manipulation using DataFrame structures optimized in Rust\nimport seaborn as sns # Import seaborn for statistical data visualization\nimport sys # Import sys for accessing system-specific parameters and functions\nimport threading\n\nfrom glob import glob # Import glob for finding all pathnames matching a specified pattern according to Unix shell rules\nfrom pathlib import Path # Import Path for working with filesystem paths in a platform-independent way\nfrom pyngrok import ngrok\nfrom scipy.stats import chi2_contingency # Import chi2_contingency for performing the Chi-Square test for independence\nfrom shapely.geometry import Point\n\n# Import Dash and its components for building interactive web applications\nfrom dash import Dash, dcc, html\nfrom dash.dependencies import Input, Output\n\n# Import scikit-learn components for creating custom estimators, model calibration, dimensionality reduction, and evaluation\nfrom sklearn.base import BaseEstimator, RegressorMixin # Base classes for creating custom scikit-learn estimators\nfrom sklearn.calibration import calibration_curve # Import calibration_curve to plot calibration curves\nfrom sklearn.decomposition import PCA # Import PCA for Principal Component Analysis\nfrom sklearn.metrics import (\n roc_curve, auc, precision_recall_curve, average_precision_score, \n roc_auc_score, confusion_matrix, ConfusionMatrixDisplay\n) # Import various metrics and plotting functions for model evaluation\nfrom sklearn.model_selection import (\n TimeSeriesSplit, GroupKFold, StratifiedGroupKFold, train_test_split\n) # Import utilities for splitting data into training and testing sets\nfrom sklearn.pipeline import Pipeline # Import Pipeline for creating a sequence of data transformations and a final estimator\nfrom sklearn.preprocessing import StandardScaler # Import StandardScaler for standardizing features\n\n# Define the data path using Path which ensures the path is correctly formatted for the current operating system\ndataPath = Path(\"/kaggle/input/home-credit-credit-risk-model-stability\")","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:35:39.916161Z","iopub.execute_input":"2024-06-01T09:35:39.917019Z","iopub.status.idle":"2024-06-01T09:35:39.930507Z","shell.execute_reply.started":"2024-06-01T09:35:39.916989Z","shell.execute_reply":"2024-06-01T09:35:39.929628Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Table Description","metadata":{}},{"cell_type":"code","source":"class Pipeline:\n # Method to set data types for various columns in the DataFrame according to their content and purpose\n @staticmethod\n def set_table_dtypes(df: pd.DataFrame) -> pd.DataFrame:\n for col in df.columns:\n # Handle specific columns by setting appropriate integer types\n if col in [\"case_id\", \"WEEK_NUM\", \"num_group1\", \"num_group2\"]:\n df = df.with_columns(pl.col(col).cast(pl.Int64))\n # Convert 'date_decision' column to a Date type for handling temporal information\n elif col in [\"date_decision\"]:\n df = df.with_columns(pl.col(col).cast(pl.Date))\n # Cast numeric columns ending with 'P' or 'A' to Float64 for numeric calculations\n elif col[-1] in (\"P\", \"A\"):\n df = df.with_columns(pl.col(col).cast(pl.Float64))\n # Cast columns ending with 'M' to String type for categorical handling\n elif col[-1] in (\"M\",):\n df = df.with_columns(pl.col(col).cast(pl.String))\n # Convert date columns (ending with 'D') to Date type for date operations\n elif col[-1] in (\"D\",):\n df = df.with_columns(pl.col(col).cast(pl.Date))\n return df\n\n # Method to handle date columns and calculate the difference in days from 'date_decision'\n @staticmethod\n def handle_dates(df: pd.DataFrame) -> pd.DataFrame:\n for col in df.columns:\n if col[-1] in (\"D\",):\n # Calculate the difference between the date columns and 'date_decision'\n df = df.with_columns(pl.col(col) - pl.col(\"date_decision\"))\n # Convert the time differences to total days\n df = df.with_columns(pl.col(col).dt.total_days())\n # Drop 'date_decision' and 'MONTH' columns as they are no longer necessary\n df = df.drop(\"date_decision\", \"MONTH\")\n return df\n\n # Method to filter out columns based on missing values and frequency of unique values\n @staticmethod\n def filter_cols(df: pd.DataFrame) -> pd.DataFrame:\n # Drop columns with more than 70% missing values\n for col in df.columns:\n if col not in [\"target\", \"case_id\", \"WEEK_NUM\"]:\n isnull = df[col].is_null().mean()\n if isnull > 0.7:\n df = df.drop(col)\n\n # Drop columns with only one unique value or more than 200 unique values\n for col in df.columns:\n if (col not in [\"target\", \"case_id\", \"WEEK_NUM\"]) & (df[col].dtype == pl.String):\n freq = df[col].n_unique()\n if (freq == 1) | (freq > 200):\n df = df.drop(col)\n return df\n \n # Method to convert string or object-type columns to categorical data type for efficient processing\n @staticmethod\n def convert_strings(df: pd.DataFrame) -> pd.DataFrame:\n for col in df.columns: \n if df[col].dtype.name in ['object', 'string']:\n # Convert to string and then to a categorical data type\n df[col] = df[col].astype(\"string\").astype('category')\n # Extend the existing categories with an 'Unknown' category\n current_categories = df[col].cat.categories\n new_categories = current_categories.to_list() + [\"Unknown\"]\n new_dtype = pd.CategoricalDtype(categories=new_categories, ordered=True)\n df[col] = df[col].astype(new_dtype)\n return df","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:02:14.128277Z","iopub.execute_input":"2024-06-01T09:02:14.129314Z","iopub.status.idle":"2024-06-01T09:02:14.152450Z","shell.execute_reply.started":"2024-06-01T09:02:14.129273Z","shell.execute_reply":"2024-06-01T09:02:14.150821Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class Aggregator:\n # Method to generate expressions for numeric columns\n @staticmethod\n def num_expr(df):\n # Identify columns ending with 'P' or 'A' as numeric columns\n cols = [col for col in df.columns if col[-1] in (\"P\", \"A\")]\n # Create max aggregation expressions for numeric columns\n expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n return expr_max\n\n # Method to generate expressions for date columns\n @staticmethod\n def date_expr(df):\n # Identify columns ending with 'D' as date columns\n cols = [col for col in df.columns if col[-1] in (\"D\",)]\n # Create max aggregation expressions for date columns\n expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n return expr_max\n\n # Method to generate expressions for string columns\n @staticmethod\n def str_expr(df):\n # Identify columns ending with 'M' as string columns\n cols = [col for col in df.columns if col[-1] in (\"M\",)]\n # Create max aggregation expressions for string columns\n expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n return expr_max\n\n # Method to generate expressions for other specific columns\n @staticmethod\n def other_expr(df):\n # Identify columns ending with 'T' or 'L' as other specific columns\n cols = [col for col in df.columns if col[-1] in (\"T\", \"L\")]\n # Create max aggregation expressions for these columns\n expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n return expr_max\n\n # Method to generate expressions for count-related columns\n @staticmethod\n def count_expr(df):\n # Identify columns containing 'num_group' as count-related columns\n cols = [col for col in df.columns if \"num_group\" in col]\n # Create max aggregation expressions for these columns\n expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n return expr_max\n\n # Method to aggregate all expression types\n @staticmethod\n def get_exprs(df):\n # Combine all types of aggregation expressions\n exprs = Aggregator.num_expr(df) + \\\n Aggregator.date_expr(df) + \\\n Aggregator.str_expr(df) + \\\n Aggregator.other_expr(df) + \\\n Aggregator.count_expr(df)\n return exprs\n\n# Feature engineering function to enhance the base DataFrame\ndef feature_eng(df_base, depth_0, depth_1, depth_2):\n df_base = (\n df_base\n .with_columns(\n # Extract month and weekday from 'date_decision' for additional features\n month_decision = pl.col(\"date_decision\").dt.month(),\n weekday_decision = pl.col(\"date_decision\").dt.weekday(),\n )\n )\n # Join additional dataframes on 'case_id' with left join\n for i, df in enumerate(depth_0 + depth_1 + depth_2):\n df_base = df_base.join(df, how=\"left\", on=\"case_id\", suffix=f\"_{i}\")\n # Handle dates by calculating differences in days\n df_base = df_base.pipe(Pipeline.handle_dates)\n return df_base\n\n# Function to convert Polars DataFrame to Pandas DataFrame and handle categorical columns\ndef to_pandas(df_data, cat_cols=None):\n df_data = df_data.to_pandas()\n if cat_cols is None:\n # Identify object type columns and convert them to category type\n cat_cols = list(df_data.select_dtypes(\"object\").columns)\n df_data[cat_cols] = df_data[cat_cols].astype(\"category\")\n return df_data, cat_cols\n\n# Function to reduce memory usage of the DataFrame\ndef reduce_mem_usage(df):\n # Calculate initial memory usage\n start_mem = df.memory_usage().sum() / 1024**2\n print('Memory usage of dataframe is {:.2f} MB'.format(start_mem))\n \n # Iterate through all columns and optimize data types\n for col in df.columns:\n col_type = df[col].dtype\n if str(col_type) == \"category\":\n continue\n \n if col_type != object:\n c_min = df[col].min()\n c_max = df[col].max()\n # Optimize integer columns\n if str(col_type)[:3] == 'int':\n if c_min > np.iinfo(np.int8).min and c_max < np.iinfo(np.int8).max:\n df[col] = df[col].astype(np.int8)\n elif c_min > np.iinfo(np.int16).min and c_max < np.iinfo(np.int16).max:\n df[col] = df[col].astype(np.int16)\n elif c_min > np.iinfo(np.int32).min and c_max < np.iinfo(np.int32).max:\n df[col] = df[col].astype(np.int32)\n elif c_min > np.iinfo(np.int64).min and c_max < np.iinfo(np.int64).max:\n df[col] = df[col].astype(np.int64) \n # Optimize float columns\n else:\n if c_min > np.finfo(np.float16).min and c_max < np.finfo(np.float16).max:\n df[col] = df[col].astype(np.float16)\n elif c_min > np.finfo(np.float32).min and c_max < np.finfo(np.float32).max:\n df[col] = df[col].astype(np.float32)\n else:\n df[col] = df[col].astype(np.float64)\n else:\n continue\n # Calculate and print the final memory usage\n end_mem = df.memory_usage().sum() / 1024**2\n print('Memory usage after optimization is: {:.2f} MB'.format(end_mem))\n print('Decreased by {:.1f}%'.format(100 * (start_mem - end_mem) / start_mem))\n \n return df","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:02:14.155268Z","iopub.execute_input":"2024-06-01T09:02:14.155657Z","iopub.status.idle":"2024-06-01T09:02:14.181912Z","shell.execute_reply.started":"2024-06-01T09:02:14.155627Z","shell.execute_reply":"2024-06-01T09:02:14.181042Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### File Reading","metadata":{}},{"cell_type":"code","source":"def read_file(path, depth=None):\n df = pl.read_parquet(path)\n df = df.pipe(Pipeline.set_table_dtypes)\n if depth in [1, 2]:\n df = df.group_by(\"case_id\").agg(Aggregator.get_exprs(df))\n return df\n\ndef read_files(regex_path, depth=None):\n # Initialize an empty list to store chunks of DataFrames\n chunks = []\n\n # Iterate over all paths that match the regex_path pattern\n for path in glob(str(regex_path)):\n # Read a Parquet file from each matching path\n df = pl.read_parquet(path)\n # Apply data type conversions and adjustments using a method from the Pipeline class\n df = df.pipe(Pipeline.set_table_dtypes)\n # If the depth is 1 or 2, group by 'case_id' and aggregate using custom expressions\n if depth in [1, 2]:\n df = df.group_by(\"case_id\").agg(Aggregator.get_exprs(df))\n # Append the processed DataFrame to the chunks list\n chunks.append(df)\n\n # Concatenate all DataFrame chunks vertically, allowing for relaxed schema matching\n df = pl.concat(chunks, how=\"vertical_relaxed\")\n # Remove duplicate entries based on 'case_id' to ensure unique cases in the final DataFrame\n df = df.unique(subset=[\"case_id\"])\n # Return the consolidated DataFrame\n return df\n","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:02:14.183106Z","iopub.execute_input":"2024-06-01T09:02:14.183467Z","iopub.status.idle":"2024-06-01T09:02:14.198361Z","shell.execute_reply.started":"2024-06-01T09:02:14.183424Z","shell.execute_reply":"2024-06-01T09:02:14.197369Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Paths For Train/Test Datasets","metadata":{}},{"cell_type":"code","source":"TRAIN_DIR = dataPath / \"parquet_files\" / \"train\"\nTEST_DIR = dataPath / \"parquet_files\" / \"test\"","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:02:14.199511Z","iopub.execute_input":"2024-06-01T09:02:14.199773Z","iopub.status.idle":"2024-06-01T09:02:14.212468Z","shell.execute_reply.started":"2024-06-01T09:02:14.199750Z","shell.execute_reply":"2024-06-01T09:02:14.211629Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Reading And Storing Train Data","metadata":{}},{"cell_type":"code","source":"%%time\n\n# Initialize a dictionary to store various DataFrames loaded from Parquet files\ndata_store_train = {\n # Load the base layer of train data from a Parquet file.\n \"df_base\": read_file(TRAIN_DIR / \"train_base.parquet\"), \n\n # Load depth 0 data, which might represent the most general data without historical depth\n \"depth_0\": [\n read_file(TRAIN_DIR / \"train_static_cb_0.parquet\"), \n read_files(TRAIN_DIR / \"train_static_0_*.parquet\"), \n ],\n\n # Load depth 1 data, potentially more detailed data that includes one level of historical context\n \"depth_1\": [\n read_files(TRAIN_DIR / \"train_applprev_1_*.parquet\", 1), \n read_file(TRAIN_DIR / \"train_tax_registry_a_1.parquet\", 1), \n read_file(TRAIN_DIR / \"train_tax_registry_b_1.parquet\", 1), \n read_file(TRAIN_DIR / \"train_tax_registry_c_1.parquet\", 1), \n read_files(TRAIN_DIR / \"train_credit_bureau_a_1_*.parquet\", 1), \n read_file(TRAIN_DIR / \"train_credit_bureau_b_1.parquet\", 1), \n read_file(TRAIN_DIR / \"train_other_1.parquet\", 1),\n read_file(TRAIN_DIR / \"train_person_1.parquet\", 1), \n read_file(TRAIN_DIR / \"train_deposit_1.parquet\", 1), \n read_file(TRAIN_DIR / \"train_debitcard_1.parquet\", 1), \n ],\n\n # Load depth 2 data, indicating data with two levels of historical or contextual depth\n \"depth_2\": [\n read_file(TRAIN_DIR / \"train_credit_bureau_b_2.parquet\", 2), \n read_files(TRAIN_DIR / \"train_credit_bureau_a_2_*.parquet\", 2), \n read_file(TRAIN_DIR / \"train_applprev_2.parquet\", 2),\n read_file(TRAIN_DIR / \"train_person_2.parquet\", 2) \n ]\n}","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:02:14.213517Z","iopub.execute_input":"2024-06-01T09:02:14.213920Z","iopub.status.idle":"2024-06-01T09:04:20.112986Z","shell.execute_reply.started":"2024-06-01T09:02:14.213895Z","shell.execute_reply":"2024-06-01T09:04:20.111881Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Reading And Storing Test Data","metadata":{}},{"cell_type":"code","source":"%%time\n\n# Initialize a dictionary to store various DataFrames loaded from Parquet files\ndata_store_test = {\n # Load the base layer of test data from a Parquet file.\n \"df_base\": read_file(TEST_DIR / \"test_base.parquet\"),\n \n # Load depth 0 data, typically static information that does not change over time.\n \"depth_0\": [\n read_file(TEST_DIR / \"test_static_cb_0.parquet\"), \n read_files(TEST_DIR / \"test_static_0_*.parquet\"),\n ],\n \n # Load depth 1 data, which includes data with one level of historical context.\n \"depth_1\": [\n read_files(TEST_DIR / \"test_applprev_1_*.parquet\", 1), \n read_file(TEST_DIR / \"test_tax_registry_a_1.parquet\", 1), \n read_file(TEST_DIR / \"test_tax_registry_b_1.parquet\", 1),\n read_file(TEST_DIR / \"test_tax_registry_c_1.parquet\", 1),\n read_files(TEST_DIR / \"test_credit_bureau_a_1_*.parquet\", 1), \n read_file(TEST_DIR / \"test_credit_bureau_b_1.parquet\", 1), \n read_file(TEST_DIR / \"test_other_1.parquet\", 1), \n read_file(TEST_DIR / \"test_person_1.parquet\", 1), \n read_file(TEST_DIR / \"test_deposit_1.parquet\", 1),\n read_file(TEST_DIR / \"test_debitcard_1.parquet\", 1), \n ],\n \n # Load depth 2 data, indicating more complex or detailed historical data.\n \"depth_2\": [\n read_file(TEST_DIR / \"test_credit_bureau_b_2.parquet\", 2), \n read_files(TEST_DIR / \"test_credit_bureau_a_2_*.parquet\", 2), \n read_file(TEST_DIR / \"test_applprev_2.parquet\", 2), \n read_file(TEST_DIR / \"test_person_2.parquet\", 2) \n ]\n}","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:04:20.114298Z","iopub.execute_input":"2024-06-01T09:04:20.114958Z","iopub.status.idle":"2024-06-01T09:04:20.455892Z","shell.execute_reply.started":"2024-06-01T09:04:20.114931Z","shell.execute_reply":"2024-06-01T09:04:20.454985Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"###3. Exploratory Data Analysis
","metadata":{}},{"cell_type":"markdown","source":"The exploratory data analysis (EDA) process undertaken is thorough and methodical, ensuring a deep understanding of the dataset before applying any machine learning algorithms. The EDA begins with preprocessing the train dataset, removing unnecessary columns, and renaming the remaining columns for better readability. This refined dataset serves as the foundation for all subsequent analyses.\n\n**Data Quality Assessment**\nThe first step in EDA is assessing data quality. This involves visualizing missing data patterns with heatmaps to determine if the data is missing at random. Outliers are identified using boxplots and scatter plots, which helps in evaluating data integrity and addressing any anomalies.\n\n**Univariate Analysis**\nUnivariate analysis focuses on understanding the distribution of individual variables. For numerical data, histograms and density plots are used to examine their distribution characteristics. For categorical data, bar charts are employed to analyze the frequency distributions of various categories, providing insights into the data's basic structure.\n\n**Bivariate Analysis**\nBivariate analysis explores relationships between two variables. Correlation analysis is conducted using heatmaps to investigate relationships between numerical variables. For categorical variables, cross-tabulation combined with Chi-square tests is used to examine associations, revealing potential dependencies and interactions between variables.\n\n**Multivariate Analysis**\nMultivariate analysis extends the exploration to multiple variables simultaneously. Pair plots are used to visualize relationships across several variables, highlighting interactions and correlations. Principal Component Analysis (PCA) is employed for dimensionality reduction, capturing the most significant variances within the dataset and aiding in data simplification while preserving essential information.\n\n**Advanced Visualizations**\nAdvanced visualizations enhance the understanding of complex patterns within the data. Heatmaps are used for temporal data to reveal time-based patterns. Interactive dashboards facilitate dynamic exploration of the dataset, allowing for a more intuitive and comprehensive analysis of the data.\n\n**Feature Engineering**\nFeature engineering involves creating new features to enrich the dataset. This includes generating aggregated features and feature interactions, enhancing the model’s input space. By incorporating domain knowledge and statistical methods, feature engineering improves the dataset's predictive power.\n\n**Temporal Analysis**\nTemporal analysis provides insights into how data evolves over time. Time series plots are used to visualize changes in variables across different time periods. This analysis helps in understanding trends, seasonality, and temporal dependencies, which are crucial for time-sensitive predictions.","metadata":{}},{"cell_type":"markdown","source":"### Feature Selection","metadata":{}},{"cell_type":"code","source":"df_train = feature_eng(**data_store_train)\ndf_train = df_train.pipe(Pipeline.filter_cols)\nprint(\"Train data shape:\\t\", df_train.shape)\ndf_train, cat_cols = to_pandas(df_train)\ndf_train = reduce_mem_usage(df_train)\ndf_train","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:04:20.457180Z","iopub.execute_input":"2024-06-01T09:04:20.457535Z","iopub.status.idle":"2024-06-01T09:05:00.610265Z","shell.execute_reply.started":"2024-06-01T09:04:20.457504Z","shell.execute_reply":"2024-06-01T09:05:00.609271Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_test = feature_eng(**data_store_test)\ndf_test = df_test.select([col for col in df_train.columns if col != \"target\"])\nprint(\"test data shape:\\t\", df_test.shape)\ndf_test, cat_cols = to_pandas(df_test)\ndf_test = reduce_mem_usage(df_test)\ndf_test","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:00.615167Z","iopub.execute_input":"2024-06-01T09:05:00.615524Z","iopub.status.idle":"2024-06-01T09:05:00.914534Z","shell.execute_reply.started":"2024-06-01T09:05:00.615495Z","shell.execute_reply":"2024-06-01T09:05:00.913428Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"del data_store_train, data_store_test\ngc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:00.916105Z","iopub.execute_input":"2024-06-01T09:05:00.917010Z","iopub.status.idle":"2024-06-01T09:05:01.456930Z","shell.execute_reply.started":"2024-06-01T09:05:00.916968Z","shell.execute_reply":"2024-06-01T09:05:01.455790Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Removing Unnecessary Columns From Train Data","metadata":{}},{"cell_type":"code","source":"df_train_refined = df_train","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:01.459884Z","iopub.execute_input":"2024-06-01T09:05:01.460177Z","iopub.status.idle":"2024-06-01T09:05:01.468296Z","shell.execute_reply.started":"2024-06-01T09:05:01.460153Z","shell.execute_reply":"2024-06-01T09:05:01.467283Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Function to summarize data\ndef summarize_data(df):\n missing_data = df.isnull().sum()\n missing_percent = (missing_data / len(df)) * 100\n non_null_counts = df.notnull().sum()\n unique_counts = df.nunique()\n \n summary = pd.DataFrame({\n 'Missing Count': missing_data,\n 'Missing Percent': missing_percent,\n 'Non-Null Count': non_null_counts,\n 'Unique Count': unique_counts\n })\n return summary\n\n# Adjust display settings to show all rows and columns\npd.set_option('display.max_rows', None)\npd.set_option('display.max_columns', None)\npd.set_option('display.max_colwidth', None)\npd.set_option('display.width', 1000)\n\n# Generate summary for df_train_refined\ndata_summary = summarize_data(df_train_refined)\n\n# Print the full summary\nprint(data_summary)\n\n# Reset display options to defaults after printing\npd.reset_option('display.max_rows')\npd.reset_option('display.max_columns')\npd.reset_option('display.max_colwidth')\npd.reset_option('display.width')","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:01.469662Z","iopub.execute_input":"2024-06-01T09:05:01.470211Z","iopub.status.idle":"2024-06-01T09:05:11.998549Z","shell.execute_reply.started":"2024-06-01T09:05:01.470176Z","shell.execute_reply":"2024-06-01T09:05:11.997501Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# List of columns to retain based on EDA objectives\ncolumns_to_keep = [\n 'case_id', 'WEEK_NUM', 'target', 'month_decision', 'weekday_decision', \n 'dateofbirth_337D', 'datefirstoffer_1144D', 'datelastunpaid_3546854D',\n 'dtlastpmtallstes_4499206D', 'firstdatedue_489D', 'lastapplicationdate_877D',\n 'lastapprdate_640D', 'lastrejectdate_50D', 'days120_123L', 'days180_256L',\n 'days30_165L', 'days360_512L', 'days90_310L', 'pmtssum_45A', 'annuity_780A',\n 'annuitynextmonth_57A', 'amtinstpaidbefduel24m_4187115A', 'max_annuity_853A',\n 'credamount_770A', 'currdebt_22A', 'totaldebt_9A', 'totalsettled_863A',\n 'pmtnum_254L', 'sumoutstandtotal_3546847A', 'sumoutstandtotalest_4493215A',\n 'education_1103M', 'maritalst_385M', 'education_88M', 'maritalst_893M',\n 'requesttype_4525192L', 'lastrejectreason_759M', 'lastrejectreasonclient_4145040M'\n]\n\n# Filter the DataFrame to keep only the necessary columns\ndf_train_refined = df_train_refined[columns_to_keep]\n\n# Display the filtered DataFrame summary\nprint(df_train_refined.info())","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:11.999842Z","iopub.execute_input":"2024-06-01T09:05:12.000147Z","iopub.status.idle":"2024-06-01T09:05:12.411008Z","shell.execute_reply.started":"2024-06-01T09:05:12.000121Z","shell.execute_reply":"2024-06-01T09:05:12.410085Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Renaming The Remaining Columns","metadata":{}},{"cell_type":"code","source":"def rename_necessary_columns(df, columns_to_rename):\n \"\"\"\n Rename the columns in the dataframe.\n\n Args:\n df (DataFrame): The DataFrame to be used.\n columns_to_rename (dict): Dictionary of columns to be renamed.\n\n Returns:\n DataFrame: The modified DataFrame with the columns renamed.\n \"\"\"\n df_renamed = df.rename(columns=columns_to_rename)\n return df_renamed","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:12.412349Z","iopub.execute_input":"2024-06-01T09:05:12.413099Z","iopub.status.idle":"2024-06-01T09:05:12.418437Z","shell.execute_reply.started":"2024-06-01T09:05:12.413061Z","shell.execute_reply":"2024-06-01T09:05:12.417425Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"columns_mapping = {\n 'case_id': 'Case ID',\n 'WEEK_NUM': 'Week Number',\n 'target': 'Target',\n 'month_decision': 'Month Decision',\n 'weekday_decision': 'Weekday Decision',\n 'dateofbirth_337D': 'Date of Birth',\n 'datefirstoffer_1144D': 'Date First Offer',\n 'datelastunpaid_3546854D': 'Date Last Unpaid',\n 'dtlastpmtallstes_4499206D': 'Date Last Payment All States',\n 'firstdatedue_489D': 'First Due Date',\n 'lastapplicationdate_877D': 'Last Application Date',\n 'lastapprdate_640D': 'Last Approval Date',\n 'lastrejectdate_50D': 'Last Rejection Date',\n 'days120_123L': '120 Days',\n 'days180_256L': '180 Days',\n 'days30_165L': '30 Days',\n 'days360_512L': '360 Days',\n 'days90_310L': '90 Days',\n 'pmtssum_45A': 'Payments Sum',\n 'annuity_780A': 'Annuity',\n 'annuitynextmonth_57A': 'Next Month Annuity',\n 'amtinstpaidbefduel24m_4187115A': 'Amount Installments Paid Before Due Last 24 Months',\n 'max_annuity_853A': 'Maximum Annuity',\n 'credamount_770A': 'Credit Amount',\n 'currdebt_22A': 'Current Debt',\n 'totaldebt_9A': 'Total Debt',\n 'totalsettled_863A': 'Total Settled',\n 'pmtnum_254L': 'Number of Payments',\n 'sumoutstandtotal_3546847A': 'Total Outstanding Amount',\n 'sumoutstandtotalest_4493215A': 'Total Outstanding Amount Estimated',\n 'education_1103M': 'Education Level',\n 'maritalst_385M': 'Marital Status',\n 'education_88M': 'Education Level 2',\n 'maritalst_893M': 'Marital Status 2',\n 'requesttype_4525192L': 'Request Type',\n 'lastrejectreason_759M': 'Last Rejection Reason',\n 'lastrejectreasonclient_4145040M': 'Last Client Rejection Reason'\n}\n\ndf_train_refined = rename_necessary_columns(df_train_refined, columns_mapping)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:12.419848Z","iopub.execute_input":"2024-06-01T09:05:12.420184Z","iopub.status.idle":"2024-06-01T09:05:12.579596Z","shell.execute_reply.started":"2024-06-01T09:05:12.420152Z","shell.execute_reply":"2024-06-01T09:05:12.578449Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Data Quality Assessment","metadata":{}},{"cell_type":"markdown","source":"#### Missing Data Analysis","metadata":{}},{"cell_type":"code","source":"def plot_missing_data_heatmap(df):\n \"\"\"\n Visualizes missing data in a DataFrame using a heatmap. This enhanced version also displays the percentage\n of missing data for each column as annotations on the heatmap.\n\n Args:\n df (DataFrame): The DataFrame to analyze for missing data.\n \"\"\"\n # Calculating the percentage of missing data in each column\n missing_percent = df.isnull().mean() * 100\n\n # Creating a DataFrame for visualization\n missing_data = missing_percent.to_frame(name='percent').T\n\n # Setting up the matplotlib figure\n plt.figure(figsize=(12, 1)) # Adjust the figure size to be more narrow, as we have only one row\n\n # Creating a heatmap to visualize the missing data presence\n sns.heatmap(missing_data, annot=True, fmt=\".1f\", cmap='viridis', cbar=False,\n yticklabels=False, annot_kws={\"color\": \"white\", \"fontsize\": 10})\n\n plt.title('Missing Data Heatmap')\n plt.xlabel('Columns')\n plt.ylabel('Rows')\n plt.show()\n \n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n \n # Collect garbage to free up memory\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:12.581028Z","iopub.execute_input":"2024-06-01T09:05:12.581801Z","iopub.status.idle":"2024-06-01T09:05:12.589001Z","shell.execute_reply.started":"2024-06-01T09:05:12.581766Z","shell.execute_reply":"2024-06-01T09:05:12.588036Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_missing_data_heatmap(df_train_refined)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:12.590316Z","iopub.execute_input":"2024-06-01T09:05:12.590599Z","iopub.status.idle":"2024-06-01T09:05:13.430169Z","shell.execute_reply.started":"2024-06-01T09:05:12.590577Z","shell.execute_reply":"2024-06-01T09:05:13.429267Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Outlier Detection","metadata":{}},{"cell_type":"code","source":"def plot_boxplots(df):\n \"\"\"\n Enhanced boxplot function that plots boxplots for all numerical columns in a DataFrame to identify outliers.\n Plots are displayed in a grid for better organization and readability, optimized to reduce memory usage by plotting in batches.\n\n Args:\n df (DataFrame): The DataFrame with numerical columns.\n \"\"\"\n # Selecting numerical columns only\n num_cols = df.select_dtypes(include=['number']).columns\n num_plots = len(num_cols)\n max_plots_per_figure = 9 # Limiting the number of plots per figure to manage memory usage\n\n # Calculating the number of figures needed\n num_figures = (num_plots // max_plots_per_figure) + (1 if num_plots % max_plots_per_figure else 0)\n\n # Iterating over each figure set\n for fig_index in range(num_figures):\n start_index = fig_index * max_plots_per_figure\n end_index = start_index + max_plots_per_figure\n current_cols = num_cols[start_index:end_index]\n\n # Creating subplots for the current set of columns\n plt.figure(figsize=(18, 12))\n for i, col in enumerate(current_cols, 1):\n plt.subplot(3, 3, i)\n try:\n # Drop missing values for plotting\n data = df[col].dropna()\n \n # Ensure the column is not empty and contains valid data for plotting\n if not data.empty and pd.api.types.is_numeric_dtype(data):\n sns.boxplot(x=data)\n plt.title(f'Boxplot of {col}')\n plt.grid(True)\n else:\n print(f\"Skipping column {col} as it is empty or not numeric.\")\n except Exception as e:\n print(f\"Could not plot boxplot for column {col}: {e}\")\n \n plt.tight_layout()\n plt.show()\n \n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n\n # Collect garbage to free up memory\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:13.431668Z","iopub.execute_input":"2024-06-01T09:05:13.432137Z","iopub.status.idle":"2024-06-01T09:05:13.442536Z","shell.execute_reply.started":"2024-06-01T09:05:13.432105Z","shell.execute_reply":"2024-06-01T09:05:13.441684Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_boxplots(df_train_refined)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:05:13.443863Z","iopub.execute_input":"2024-06-01T09:05:13.444266Z","iopub.status.idle":"2024-06-01T09:06:51.173253Z","shell.execute_reply.started":"2024-06-01T09:05:13.444234Z","shell.execute_reply":"2024-06-01T09:06:51.172275Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def plot_scatter(df, x, y):\n \"\"\"\n Enhanced scatter plot function that generates a scatter plot to detect outliers between two variables.\n Includes a regression line to indicate trends and potential outlier influence on the relationship.\n\n Args:\n df (DataFrame): DataFrame containing the data.\n x (str): Column name for the x-axis.\n y (str): Column name for the y-axis.\n \"\"\"\n # Ensure the specified columns are in the DataFrame to avoid KeyError\n if x not in df.columns or y not in df.columns:\n raise ValueError(f\"Columns {x} or {y} not found in DataFrame\")\n\n # Plotting the scatter plot and regression line\n plt.figure(figsize=(10, 6))\n sns.scatterplot(x=df[x], y=df[y]) # Plot directly using columns to avoid copying data\n sns.regplot(x=df[x], y=df[y], scatter=False, color='red') # Regression line without scatter points\n\n # Adding titles and labels\n plt.title(f'Scatter Plot between {x} and {y}')\n plt.xlabel(x)\n plt.ylabel(y)\n plt.grid(True)\n\n # Show the plot\n plt.show()\n\n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n\n # Collect garbage to free up memory\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:06:51.174623Z","iopub.execute_input":"2024-06-01T09:06:51.174935Z","iopub.status.idle":"2024-06-01T09:06:51.185671Z","shell.execute_reply.started":"2024-06-01T09:06:51.174909Z","shell.execute_reply":"2024-06-01T09:06:51.184592Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_scatter(df_train_refined, 'Week Number', 'Target')","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:06:51.186981Z","iopub.execute_input":"2024-06-01T09:06:51.187340Z","iopub.status.idle":"2024-06-01T09:10:32.558201Z","shell.execute_reply.started":"2024-06-01T09:06:51.187306Z","shell.execute_reply":"2024-06-01T09:10:32.557325Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Univariate Analysis","metadata":{}},{"cell_type":"markdown","source":"#### Histogram and Density Plots for Numerical Data","metadata":{}},{"cell_type":"code","source":"def plot_histograms(df):\n \"\"\"\n Plots histograms for all numerical columns in a DataFrame.\n\n Args:\n df (DataFrame): The DataFrame with numerical data.\n \"\"\"\n # Convert infinite values to NaN\n df.replace([np.inf, -np.inf], np.nan, inplace=True)\n \n # Converting float16 columns to float32 for compatibility\n for col in df.select_dtypes(include=['float16']).columns:\n df[col] = df[col].astype('float32')\n \n # Selecting numerical columns only\n num_cols = df.select_dtypes(include=['number']).columns\n num_plots = len(num_cols)\n rows = (num_plots // 3) + 1 if num_plots % 3 else num_plots // 3\n\n # Creating the plot\n fig, axes = plt.subplots(rows, 3, figsize=(18, 6 * rows))\n axes = axes.flatten()\n\n for i, col in enumerate(num_cols):\n sns.histplot(df[col].dropna(), kde=False, color='skyblue', edgecolor='black', ax=axes[i])\n axes[i].set_title(f'Histogram of {col}')\n axes[i].grid(True)\n \n # Hide any unused subplots\n for j in range(i + 1, len(axes)):\n fig.delaxes(axes[j])\n\n plt.tight_layout()\n plt.show()\n\n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n\n # Collect garbage to free up memory\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:10:32.559374Z","iopub.execute_input":"2024-06-01T09:10:32.559637Z","iopub.status.idle":"2024-06-01T09:10:32.569344Z","shell.execute_reply.started":"2024-06-01T09:10:32.559615Z","shell.execute_reply":"2024-06-01T09:10:32.568426Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_histograms(df_train_refined)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:10:32.570465Z","iopub.execute_input":"2024-06-01T09:10:32.570752Z","iopub.status.idle":"2024-06-01T09:12:54.893279Z","shell.execute_reply.started":"2024-06-01T09:10:32.570728Z","shell.execute_reply":"2024-06-01T09:12:54.892283Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def plot_density_plots(df):\n \"\"\"\n Plots density plots for all numerical columns in a DataFrame.\n \n Args:\n df (DataFrame): The DataFrame with numerical data.\n \"\"\"\n # Selecting numerical columns only\n num_cols = df.select_dtypes(include=['number']).columns\n num_plots = len(num_cols)\n rows = (num_plots // 3) + 1 if num_plots % 3 else num_plots // 3\n\n # Creating the plot\n fig, axes = plt.subplots(rows, 3, figsize=(18, 6 * rows))\n axes = axes.flatten()\n\n for i, col in enumerate(num_cols):\n sns.kdeplot(df[col], fill=True, color='salmon', ax=axes[i], warn_singular=False)\n axes[i].set_title(f'Density Plot of {col}')\n axes[i].grid(True)\n \n # Hide any unused subplots\n for j in range(i + 1, len(axes)):\n fig.delaxes(axes[j])\n\n plt.tight_layout()\n plt.show()\n\n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n\n # Collect garbage to free up memory\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:12:54.894802Z","iopub.execute_input":"2024-06-01T09:12:54.895627Z","iopub.status.idle":"2024-06-01T09:12:54.905957Z","shell.execute_reply.started":"2024-06-01T09:12:54.895594Z","shell.execute_reply":"2024-06-01T09:12:54.904951Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_density_plots(df_train_refined)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:12:54.907155Z","iopub.execute_input":"2024-06-01T09:12:54.907548Z","iopub.status.idle":"2024-06-01T09:15:57.486877Z","shell.execute_reply.started":"2024-06-01T09:12:54.907520Z","shell.execute_reply":"2024-06-01T09:15:57.485738Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Bar Chart for Categorical Data","metadata":{}},{"cell_type":"code","source":"def plot_bar_charts(df, top_n=None):\n \"\"\"\n Plots bar charts for all categorical columns in a DataFrame to examine the frequency of categories,\n optionally showing only the top N categories for clarity if specified.\n \n Args:\n df (DataFrame): The DataFrame with categorical data.\n top_n (int, optional): The number of top categories to display. If None, displays all categories.\n \"\"\"\n # Identify categorical columns\n cat_cols = df.select_dtypes(include=['object', 'category']).columns\n if not cat_cols.size:\n print(\"No categorical columns to plot.\")\n return\n \n for col in cat_cols:\n # Use value_counts with dropna=False to ensure no warning about observed\n data = df[col].value_counts(dropna=False).sort_values(ascending=False)\n \n # If top_n is specified, limit the data to the top N categories\n if top_n:\n data = data.head(top_n)\n \n # Create the bar plot\n plt.figure(figsize=(10, 5))\n sns.barplot(x=data.index, y=data.values, palette='viridis', dodge=False, hue=data.index, legend=False)\n plt.title(f'Frequency of Categories in {col}')\n plt.ylabel('Frequency')\n plt.xlabel(col)\n plt.xticks(rotation=45, ha='right') # Improve visibility of labels\n plt.grid(True, axis='y') # Adding grid lines for easier readability\n plt.show()\n\n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n\n # Collect garbage to free up memory\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:35:58.680735Z","iopub.execute_input":"2024-06-01T09:35:58.681086Z","iopub.status.idle":"2024-06-01T09:35:58.690435Z","shell.execute_reply.started":"2024-06-01T09:35:58.681059Z","shell.execute_reply":"2024-06-01T09:35:58.689557Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_bar_charts(df_train_refined,)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:36:02.748167Z","iopub.execute_input":"2024-06-01T09:36:02.748558Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Bivariate Analysis","metadata":{}},{"cell_type":"markdown","source":"#### Correlation Analysis","metadata":{}},{"cell_type":"code","source":"def plot_correlation_matrix(df, annot=True):\n \"\"\"\n Generates a heatmap of the correlation matrix for numerical features in the DataFrame, with options to toggle annotations.\n \n Args:\n df (DataFrame): The DataFrame containing numerical data.\n annot (bool): Whether to display annotations over the heatmap. Default is True.\n \"\"\"\n # Selecting numerical columns\n num_cols = df.select_dtypes(include=['number']).columns\n if len(num_cols) < 2:\n print(\"Not enough numerical data to compute a correlation matrix.\")\n return\n \n # Compute the correlation matrix\n corr_matrix = df[num_cols].corr()\n \n # Setting the size of the heatmap depending on the number of variables\n fig_size = max(10, len(num_cols) * 0.5)\n plt.figure(figsize=(fig_size, fig_size))\n \n # Creating the heatmap\n sns.heatmap(corr_matrix, annot=annot, cmap='coolwarm', fmt=\".2f\", square=True, linewidths=.5, cbar_kws={'shrink': .5})\n plt.title('Correlation Matrix')\n plt.xticks(rotation=45, ha='right') # Rotate column names for better readability\n plt.yticks(rotation=0) # Keep the row names horizontal for readability\n plt.show()\n\n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n\n # Collect garbage to free up memory\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:16:01.038603Z","iopub.execute_input":"2024-06-01T09:16:01.038959Z","iopub.status.idle":"2024-06-01T09:16:01.047330Z","shell.execute_reply.started":"2024-06-01T09:16:01.038930Z","shell.execute_reply":"2024-06-01T09:16:01.046336Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_correlation_matrix(df_train_refined, annot=True)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:16:01.048544Z","iopub.execute_input":"2024-06-01T09:16:01.048842Z","iopub.status.idle":"2024-06-01T09:16:07.152524Z","shell.execute_reply.started":"2024-06-01T09:16:01.048810Z","shell.execute_reply":"2024-06-01T09:16:07.151613Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Cross-Tabulations and Chi-Square Tests for Categorical Variables","metadata":{}},{"cell_type":"code","source":"def plot_crosstabs_and_chi_square(df, col1, col2):\n \"\"\"\n Performs cross-tabulation and Chi-Square test between two categorical variables to find associations, \n and visualizes the results in a heatmap.\n \n Args:\n df (DataFrame): The DataFrame containing the data.\n col1 (str): First categorical column name.\n col2 (str): Second categorical column name (often the target variable).\n \"\"\"\n # Validate that the columns contain appropriate data\n if df[col1].isnull().any() or df[col2].isnull().any():\n raise ValueError(\"Columns should not contain null values. Consider filling or dropping them before this analysis.\")\n \n # Convert columns to 'category' dtype if not already \n if not (isinstance(df[col1].dtype, pd.CategoricalDtype) and isinstance(df[col2].dtype, pd.CategoricalDtype)):\n df[col1] = df[col1].astype('category')\n df[col2] = df[col2].astype('category')\n\n # Create a cross-tabulation\n crosstab = pd.crosstab(df[col1], df[col2])\n\n # Chi-Square Test\n chi2, p, dof, expected = chi2_contingency(crosstab)\n print(f\"Chi-Square Statistic: {chi2:.2f}, p-value: {p:.3f}\")\n\n # Plotting the crosstab\n fig, ax = plt.subplots(figsize=(10, 25)) # Setting size\n sns.heatmap(crosstab, annot=True, fmt=\"d\", cmap=\"Blues\", ax=ax)\n plt.title('Cross-tabulation Heatmap')\n plt.xlabel(col2)\n plt.ylabel(col1)\n plt.xticks(rotation=45, ha='right') # Rotate x labels for better readability\n plt.yticks(rotation=0) # Ensure y labels are horizontal\n plt.show()\n\n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n\n # Clear variables from memory\n del crosstab, fig, ax\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:16:07.154034Z","iopub.execute_input":"2024-06-01T09:16:07.154548Z","iopub.status.idle":"2024-06-01T09:16:07.167200Z","shell.execute_reply.started":"2024-06-01T09:16:07.154510Z","shell.execute_reply":"2024-06-01T09:16:07.166272Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_crosstabs_and_chi_square(df_train_refined, 'Week Number', 'Target')","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:16:07.168569Z","iopub.execute_input":"2024-06-01T09:16:07.168882Z","iopub.status.idle":"2024-06-01T09:16:08.944694Z","shell.execute_reply.started":"2024-06-01T09:16:07.168855Z","shell.execute_reply":"2024-06-01T09:16:08.943754Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Multivariate Analysis","metadata":{}},{"cell_type":"markdown","source":"#### Pair Plots","metadata":{}},{"cell_type":"code","source":"def plot_pair_plots(df, columns=None, hue=None, diag_kind='auto', markers=None, plot_kws=None, sample_size=5000):\n \"\"\"\n Generates pair plots for selected numerical variables in the DataFrame to visualize relationships and distributions.\n \n Args:\n df (DataFrame): The DataFrame containing the data.\n columns (list, optional): Specific columns to include in the pair plot. Defaults to all numerical columns.\n hue (str, optional): Column name to use for color-coding points by categories.\n diag_kind (str, optional): Kind of plot for the diagonal elements ('hist', 'kde', or 'auto'). Defaults to 'auto'.\n markers (list, optional): Marker styles for different levels of the hue variable, if hue is specified.\n plot_kws (dict, optional): Keyword arguments to pass to the pairplot function for non-diagonal elements.\n sample_size (int, optional): Number of samples to plot. Defaults to 5000.\n \"\"\"\n # Validate and select appropriate columns\n if columns:\n numerical_data = df[columns].copy() # Ensure we are working on a copy\n else:\n numerical_data = df.select_dtypes(include=['number']).copy()\n \n if numerical_data.empty:\n raise ValueError(\"No numerical data available for plotting. Please check your DataFrame or column selections.\")\n\n # Convert numerical columns to more memory-efficient data types\n for col in numerical_data.columns:\n if pd.api.types.is_float_dtype(numerical_data[col]):\n numerical_data.loc[:, col] = pd.to_numeric(numerical_data[col], downcast='float')\n elif pd.api.types.is_integer_dtype(numerical_data[col]):\n numerical_data.loc[:, col] = pd.to_numeric(numerical_data[col], downcast='integer')\n\n # Sample the data for faster plotting\n if len(numerical_data) > sample_size:\n numerical_data = numerical_data.sample(sample_size, random_state=42)\n if hue:\n hue_data = df[hue].sample(sample_size, random_state=42)\n numerical_data.loc[:, hue] = hue_data.values\n\n # Settings for the pair plot\n if not plot_kws:\n plot_kws = {'alpha': 0.6} # default transparency to differentiate overlapping points\n\n # Generating the pair plot\n pair_plot = sns.pairplot(numerical_data, hue=hue, diag_kind=diag_kind, markers=markers, plot_kws=plot_kws, height=3, aspect=0.75)\n pair_plot.fig.suptitle('Pair Plot of Numerical Variables', y=1.02) # Adjust title placement\n plt.show()\n\n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n\n # Clear variables from memory\n del numerical_data, pair_plot\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:16:08.945894Z","iopub.execute_input":"2024-06-01T09:16:08.946170Z","iopub.status.idle":"2024-06-01T09:16:08.957693Z","shell.execute_reply.started":"2024-06-01T09:16:08.946146Z","shell.execute_reply":"2024-06-01T09:16:08.956614Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"columns_for_pairplot = [\n 'Date Last Unpaid', \n 'Date Last Payment All States', 'First Due Date', 'Last Application Date', \n 'Last Approval Date', 'Last Rejection Date', '120 Days', '30 Days', \n '90 Days', 'Payments Sum', 'Annuity', 'Next Month Annuity', \n 'Credit Amount', 'Current Debt', 'Total Debt', 'Total Settled', \n 'Number of Payments', 'Total Outstanding Amount'\n]\n\nplot_pair_plots(df_train_refined, columns_for_pairplot, hue='Target', diag_kind='kde', plot_kws={'edgecolor': 'k'}, sample_size=5000)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:16:08.958818Z","iopub.execute_input":"2024-06-01T09:16:08.959070Z","iopub.status.idle":"2024-06-01T09:19:29.078316Z","shell.execute_reply.started":"2024-06-01T09:16:08.959047Z","shell.execute_reply":"2024-06-01T09:19:29.076579Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### PCA (Principal Component Analysis)","metadata":{}},{"cell_type":"code","source":"def perform_pca(df, n_components=2):\n \"\"\"\n Performs PCA on numerical data from the DataFrame and plots the first two principal components with enhanced visualizations.\n \n Args:\n df (DataFrame): The DataFrame containing the data.\n n_components (int): Number of principal components to retain.\n \n Returns:\n principal_df (DataFrame): DataFrame containing the principal components.\n \"\"\"\n # Check if the DataFrame has enough data\n if df.empty:\n raise ValueError(\"The input DataFrame is empty.\")\n \n # Selecting only numerical columns\n numerical_data = df.select_dtypes(include=['number']).copy()\n \n if numerical_data.empty:\n raise ValueError(\"No numerical data available to perform PCA.\")\n \n # Handling missing values by imputing with column mean\n numerical_data = numerical_data.fillna(numerical_data.mean())\n \n # Drop any remaining rows with NaN values\n numerical_data = numerical_data.dropna()\n \n # Clip extreme values to avoid overflow issues\n numerical_data = numerical_data.clip(lower=np.finfo(np.float64).min, upper=np.finfo(np.float64).max)\n\n # Standardizing the numerical data\n scaler = StandardScaler()\n scaled_data = scaler.fit_transform(numerical_data)\n \n # Applying PCA\n pca = PCA(n_components=n_components)\n principal_components = pca.fit_transform(scaled_data)\n principal_df = pd.DataFrame(data=principal_components,\n columns=[f'Principal Component {i+1}' for i in range(n_components)])\n \n # Plotting the results\n plt.figure(figsize=(8, 6))\n sns.scatterplot(x='Principal Component 1', y='Principal Component 2', data=principal_df)\n plt.axhline(0, color='grey', lw=1)\n plt.axvline(0, color='grey', lw=1)\n plt.xlabel('Principal Component 1')\n plt.ylabel('Principal Component 2')\n plt.title('PCA of Dataset')\n plt.grid(True)\n plt.show()\n \n # Clear plot from memory\n plt.clf()\n plt.close('all')\n \n # Print explained variance ratio\n explained_variance = pca.explained_variance_ratio_\n print(f\"Explained Variance Ratio: {explained_variance}\")\n \n # Clear variables from memory\n del numerical_data, scaled_data, principal_components, pca\n gc.collect()\n \n return principal_df","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:19:29.080618Z","iopub.execute_input":"2024-06-01T09:19:29.081409Z","iopub.status.idle":"2024-06-01T09:19:29.093944Z","shell.execute_reply.started":"2024-06-01T09:19:29.081378Z","shell.execute_reply":"2024-06-01T09:19:29.092840Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"perform_pca(df_train_refined)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:19:29.095394Z","iopub.execute_input":"2024-06-01T09:19:29.095699Z","iopub.status.idle":"2024-06-01T09:19:40.674410Z","shell.execute_reply.started":"2024-06-01T09:19:29.095674Z","shell.execute_reply":"2024-06-01T09:19:40.673406Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Advanced Visualizations","metadata":{}},{"cell_type":"markdown","source":"#### Heatmaps for Temporal Data","metadata":{}},{"cell_type":"code","source":"def plot_temporal_heatmap(df, time_col, category_col, value_col, aggfunc='mean', annot=False):\n \"\"\"\n Generates a heatmap for temporal data grouped by time and another categorical variable, aggregated over specified values.\n \n Args:\n df (DataFrame): The DataFrame containing the data.\n time_col (str): The name of the column containing time data.\n category_col (str): The name of the categorical column to group data.\n value_col (str): The name of the column whose values will be visualized.\n aggfunc (str or function): Aggregation function used to aggregate data.\n annot (bool): Flag to show annotations on the heatmap.\n \"\"\"\n # Ensure the specified columns exist in the DataFrame\n if time_col not in df.columns or category_col not in df.columns or value_col not in df.columns:\n raise ValueError(\"One or more columns specified do not exist in the DataFrame.\")\n \n # Convert to appropriate data types, avoiding conversion to 'category' for float16\n df[time_col] = pd.to_numeric(df[time_col], errors='coerce')\n if df[category_col].dtype != 'float16':\n df[category_col] = df[category_col].astype('category')\n df[value_col] = pd.to_numeric(df[value_col], errors='coerce')\n \n # Drop rows with NaNs in the specified columns\n df.dropna(subset=[time_col, category_col, value_col], inplace=True)\n \n # Convert float16 columns to float32 to avoid issues with aggregation\n for col in df.select_dtypes(include=['float16']).columns:\n df[col] = df[col].astype('float32')\n \n # Creating a pivot table\n pivot_table = pd.pivot_table(df, index=time_col, columns=category_col, values=value_col, aggfunc=aggfunc, observed=False)\n \n # Plotting the heatmap\n plt.figure(figsize=(12, 8))\n sns.heatmap(pivot_table, annot=annot, fmt=\".1f\", cmap='coolwarm', linewidths=.5)\n plt.title(f'Temporal Heatmap of {value_col} by {category_col} Over {time_col}')\n plt.xlabel(category_col)\n plt.ylabel(time_col)\n plt.show()\n \n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n\n # Collect garbage to free up memory\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:19:40.675905Z","iopub.execute_input":"2024-06-01T09:19:40.676667Z","iopub.status.idle":"2024-06-01T09:19:40.687559Z","shell.execute_reply.started":"2024-06-01T09:19:40.676628Z","shell.execute_reply":"2024-06-01T09:19:40.686674Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_temporal_heatmap(df_train_refined, 'Week Number', 'Payments Sum', 'Target', aggfunc='mean', annot=True)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:19:40.688571Z","iopub.execute_input":"2024-06-01T09:19:40.688889Z","iopub.status.idle":"2024-06-01T09:33:22.709492Z","shell.execute_reply.started":"2024-06-01T09:19:40.688848Z","shell.execute_reply":"2024-06-01T09:33:22.708471Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Interactive Dashboards with Ploty and Dash","metadata":{}},{"cell_type":"code","source":"# Set the ngrok authentication token\nngrok.set_auth_token(\"2gyUJfRiQGnRb0kX6aLG6kboygs_GZPc7eZH6w5yLpT6EngR\")\n\ndef create_interactive_dashboard(df):\n \"\"\"\n Creates an interactive dashboard using Dash, featuring dynamic graphs that allow users to explore data.\n \n Args:\n df (DataFrame): The DataFrame containing the data.\n \"\"\"\n app = Dash(__name__)\n\n # Prepare a dropdown options list for categorical data selection\n categorical_columns = df.select_dtypes(include=['object', 'category']).columns\n if len(categorical_columns) < 2:\n raise ValueError(\"DataFrame must have at least two categorical columns.\")\n\n options = [{'label': col, 'value': col} for col in categorical_columns]\n\n # Dash layout with dropdown and graph components\n app.layout = html.Div([\n html.H1('Data Visualization Dashboard', style={'text-align': 'center'}),\n \n html.Div([\n dcc.Dropdown(\n id='x-axis-col',\n options=options,\n value=options[0]['value'], # default value\n style={'width': '48%', 'display': 'inline-block'}\n ),\n dcc.Dropdown(\n id='y-axis-col',\n options=options,\n value=options[1]['value'] if len(options) > 1 else options[0]['value'], # default value\n style={'width': '48%', 'float': 'right', 'display': 'inline-block'}\n )\n ]),\n \n dcc.Graph(id='interactive-scatter-plot')\n ])\n\n # Callback for updating the graph based on dropdowns\n @app.callback(\n Output('interactive-scatter-plot', 'figure'),\n [Input('x-axis-col', 'value'),\n Input('y-axis-col', 'value')]\n )\n def update_graph(x_col, y_col):\n # Check for valid column selection\n if not x_col or not y_col:\n return px.scatter(title=\"Select both x and y axes to see the plot.\")\n \n # Generate a new figure based on selected columns for x and y axes\n fig = px.scatter(df, x=x_col, y=y_col, color='Week Number',\n title=f'Interactive Scatter Plot of {x_col} vs {y_col}')\n return fig\n\n # Start ngrok tunnel\n public_url = ngrok.connect(8050)\n print(f'Public URL: {public_url}')\n\n def run_dash():\n app.run_server(port=8050, debug=True, use_reloader=False) # Turn off reloader if inside Jupyter\n\n # Run Dash in a separate thread\n dash_thread = threading.Thread(target=run_dash)\n dash_thread.start()\n\n # Clean up to free memory\n del categorical_columns, options\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:22.711089Z","iopub.execute_input":"2024-06-01T09:33:22.711996Z","iopub.status.idle":"2024-06-01T09:33:23.306618Z","shell.execute_reply.started":"2024-06-01T09:33:22.711964Z","shell.execute_reply":"2024-06-01T09:33:23.305254Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**THE FUNCTION CAN ONLY BE RUN A MAXIMUM NUMBER OF 3 TIMES. NGROK FREE TIER ONLY ALLOWS 3 ENDPOINTS. PLEASE GO TO THE LINK GENERATED TO ACCESS THE WEBSITE.**","metadata":{}},{"cell_type":"code","source":"create_interactive_dashboard(df_train_refined)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.307634Z","iopub.status.idle":"2024-06-01T09:33:23.307962Z","shell.execute_reply.started":"2024-06-01T09:33:23.307804Z","shell.execute_reply":"2024-06-01T09:33:23.307817Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Feature Engineering","metadata":{}},{"cell_type":"markdown","source":"#### Aggregation","metadata":{}},{"cell_type":"code","source":"# Define a function to convert all values in the DataFrame to strings\ndef convert_to_string(df):\n for col in df.columns:\n if pd.api.types.is_integer_dtype(df[col]):\n df[col] = df[col].astype('int64').astype(str)\n elif pd.api.types.is_float_dtype(df[col]):\n df[col] = df[col].astype('float64').astype(str)\n else:\n df[col] = df[col].astype(str)\n return df\n\n# Aggregator class\nclass Aggregator:\n \"\"\"\n A class for aggregating DataFrame columns based on specific rules and suffix conventions.\n \"\"\"\n\n @staticmethod\n def get_exprs(df):\n \"\"\"\n Combines all defined aggregation expressions into a single list applicable to a DataFrame.\n \n Args:\n df (pl.DataFrame): The DataFrame to process.\n\n Returns:\n list: List of combined aggregation expressions.\n \"\"\"\n exprs = []\n\n # Numeric columns ending with 'P' or 'A'\n num_cols = [col for col in df.columns if col.endswith(('P', 'A'))]\n exprs.extend([pl.col(col).max().alias(f\"max_{col}\") for col in num_cols])\n\n # Date columns ending with 'D'\n date_cols = [col for col in df.columns if col.endswith('D')]\n exprs.extend([pl.col(col).max().alias(f\"max_{col}\") for col in date_cols])\n\n # String columns ending with 'M'\n str_cols = [col for col in df.columns if col.endswith('M')]\n exprs.extend([pl.col(col).max().alias(f\"max_{col}\") for col in str_cols])\n\n # Other columns ending with 'T' or 'L'\n other_cols = [col for col in df.columns if col.endswith(('T', 'L'))]\n exprs.extend([pl.col(col).max().alias(f\"max_{col}\") for col in other_cols])\n\n # Columns containing 'num_group'\n count_cols = [col for col in df.columns if 'num_group' in col]\n exprs.extend([pl.col(col).max().alias(f\"max_{col}\") for col in count_cols])\n\n return exprs\n\n @staticmethod\n def aggregate_features(df, key):\n \"\"\"\n Aggregates the dataframe based on specified key using Polars.\n\n Args:\n df (pl.DataFrame): The DataFrame to aggregate.\n key (str): The column to group by.\n\n Returns:\n pl.DataFrame: Aggregated DataFrame.\n \"\"\"\n # Ensure the DataFrame is in Polars format\n if not isinstance(df, pl.DataFrame):\n df = pl.from_pandas(df)\n\n # Get the aggregation expressions\n exprs = Aggregator.get_exprs(df)\n\n # Perform the aggregation\n return df.group_by(key).agg(exprs)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.310002Z","iopub.status.idle":"2024-06-01T09:33:23.310483Z","shell.execute_reply.started":"2024-06-01T09:33:23.310243Z","shell.execute_reply":"2024-06-01T09:33:23.310264Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Main function to handle conversion and aggregation in chunks\ndef main(df, chunk_size=10000):\n aggregated_chunks = []\n\n # Process the DataFrame in chunks\n for start in range(0, len(df), chunk_size):\n chunk = df.iloc[start:start + chunk_size].copy() # Use .copy() to avoid SettingWithCopyWarning\n \n # Convert the chunk to string\n chunk_str = convert_to_string(chunk)\n \n # Convert to Polars DataFrame\n chunk_polars = pl.from_pandas(chunk_str)\n \n # Aggregate the chunk\n aggregated_chunk = Aggregator.aggregate_features(chunk_polars, 'Case ID')\n aggregated_chunks.append(aggregated_chunk)\n \n # Free memory of the current chunk\n del chunk, chunk_str, chunk_polars, aggregated_chunk\n gc.collect()\n\n # Combine all aggregated chunks\n combined_aggregated_data = pl.concat(aggregated_chunks)\n \n # Free memory of the aggregated chunks list\n del aggregated_chunks\n gc.collect()\n \n return combined_aggregated_data\n\naggregated_data = main(df_train_refined)\nprint(aggregated_data)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.311650Z","iopub.status.idle":"2024-06-01T09:33:23.312110Z","shell.execute_reply.started":"2024-06-01T09:33:23.311868Z","shell.execute_reply":"2024-06-01T09:33:23.311896Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Feature Interactions","metadata":{}},{"cell_type":"code","source":"def create_feature_interactions(df):\n \"\"\"\n Enhances a DataFrame by creating new features based on interactions between existing features.\n This includes interactions between numeric and categorical features, and among numeric features themselves.\n\n Args:\n df (DataFrame): The DataFrame to manipulate.\n\n Returns:\n DataFrame: The DataFrame with additional interaction features.\n \"\"\"\n\n # Convert relevant columns to numeric if they are not already\n numeric_columns = ['Payments Sum', 'Annuity', 'Credit Amount', 'Total Debt', 'Total Settled']\n for col in numeric_columns:\n if col in df.columns:\n df[col] = pd.to_numeric(df[col], errors='coerce')\n\n # Numeric-Numeric Interaction: Creating a ratio\n if 'Payments Sum' in df.columns and 'Annuity' in df.columns:\n df['Payments Sum to Annuity Ratio'] = df['Payments Sum'] / df['Annuity']\n print(\"Added Numeric-Numeric Interaction: Payments Sum to Annuity Ratio\")\n\n # Additional Numeric-Numeric Interaction\n if 'Payments Sum' in df.columns and 'Credit Amount' in df.columns:\n df['Payments Sum to Credit Amount Ratio'] = df['Payments Sum'] / df['Credit Amount']\n print(\"Added Numeric-Numeric Interaction: Payments Sum to Credit Amount Ratio\")\n\n # Numeric-Categorical Interaction: Mean encoding\n if 'Education Level' in df.columns and 'Payments Sum' in df.columns:\n df['Average Payments Sum by Education'] = df.groupby('Education Level', observed=False)['Payments Sum'].transform('mean')\n print(\"Added Numeric-Categorical Interaction: Average Payments Sum by Education\")\n\n if 'Request Type' in df.columns and 'Total Debt' in df.columns:\n df['Average Total Debt by Request Type'] = df.groupby('Request Type', observed=False)['Total Debt'].transform('mean')\n print(\"Added Numeric-Categorical Interaction: Average Total Debt by Request Type\")\n\n # Categorical-Categorical Interaction: Combining categories\n if 'Education Level' in df.columns and 'Marital Status' in df.columns:\n df['Education and Marital Status'] = df['Education Level'].astype(str) + \"_\" + df['Marital Status'].astype(str)\n print(\"Added Categorical-Categorical Interaction: Education and Marital Status\")\n\n if 'Request Type' in df.columns and 'Last Rejection Reason' in df.columns:\n df['Request and Rejection Reason'] = df['Request Type'].astype(str) + \"_\" + df['Last Rejection Reason'].astype(str)\n print(\"Added Categorical-Categorical Interaction: Request and Rejection Reason\")\n\n return df","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.313360Z","iopub.status.idle":"2024-06-01T09:33:23.313680Z","shell.execute_reply.started":"2024-06-01T09:33:23.313527Z","shell.execute_reply":"2024-06-01T09:33:23.313540Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"interacted_data = create_feature_interactions(df_train_refined)\nprint(interacted_data.head())","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.314599Z","iopub.status.idle":"2024-06-01T09:33:23.314936Z","shell.execute_reply.started":"2024-06-01T09:33:23.314768Z","shell.execute_reply":"2024-06-01T09:33:23.314783Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Temporal Analysis","metadata":{}},{"cell_type":"markdown","source":"#### Time Series Analysis","metadata":{}},{"cell_type":"code","source":"def plot_time_series(df, date_col, target_col, line_color='blue', line_style='-', show_grid=True):\n \"\"\"\n Enhanced plot function for time series data to visualize trends, seasonality, and handle missing data.\n \n Args:\n df (DataFrame): The DataFrame containing the data.\n date_col (str): The name of the datetime column.\n target_col (str): The target variable to plot.\n line_color (str): Color of the line plot.\n line_style (str): Style of the line plot.\n show_grid (bool): Whether to display grid lines in the plot.\n \"\"\"\n # Check if the necessary columns are in the DataFrame\n if date_col not in df.columns or target_col not in df.columns:\n raise ValueError(\"Specified columns are not in the DataFrame\")\n \n # Create a copy of the DataFrame to avoid modifying the original\n df_copy = df[[date_col, target_col]].copy()\n\n # Ensure the date column is in datetime format\n df_copy[date_col] = pd.to_datetime(df_copy[date_col], errors='coerce')\n\n # Handling missing values in date and target columns\n if df_copy[date_col].isnull().any() or df_copy[target_col].isnull().any():\n df_copy.dropna(subset=[date_col, target_col], inplace=True)\n print(\"Missing data in the specified columns has been dropped.\")\n \n # Set date column as index\n df_copy.set_index(date_col, inplace=True)\n \n # Plotting the target variable over time\n plt.figure(figsize=(14, 7))\n df_copy[target_col].plot(color=line_color, style=line_style)\n plt.title(f'Time Series Plot of {target_col}')\n plt.xlabel('Date')\n plt.ylabel(target_col)\n if show_grid:\n plt.grid(True)\n plt.show()\n\n # Free memory\n del df_copy\n \n # Clear the plot to free up memory\n plt.clf()\n plt.close('all')\n\n # Collect garbage to free up memory\n gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.316547Z","iopub.status.idle":"2024-06-01T09:33:23.317033Z","shell.execute_reply.started":"2024-06-01T09:33:23.316794Z","shell.execute_reply":"2024-06-01T09:33:23.316816Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_time_series(df_train_refined, 'Last Application Date', 'Payments Sum')","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.318442Z","iopub.status.idle":"2024-06-01T09:33:23.318761Z","shell.execute_reply.started":"2024-06-01T09:33:23.318604Z","shell.execute_reply":"2024-06-01T09:33:23.318618Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**Description of the DataFrame** \n\nThe `df_train_refined` DataFrame consists of 1,526,659 entries, each representing a unique loan application case identified by the `Case ID`. The dataset is enriched with 37 columns that capture a broad spectrum of information about each application, including chronological details, financial metrics, and categorical data. \n\nChronological Information:\n- `Week Number`, `Month Decision`, `Weekday Decision` indicate the temporal aspects of the loan decisions.\n- Date columns such as `Date of Birth`, `Date First Offer`, `Date Last Unpaid`, `Date Last Payment All States`, `First Due Date`, `Last Application Date`, `Last Approval Date`, and `Last Rejection Date` provide key timestamps marking various stages in the loan lifecycle.\n\nFinancial Metrics:\n- Columns like `Payments Sum`, `Annuity`, `Next Month Annuity`, `Amount Installments Paid Before Due Last 24 Months`, `Maximum Annuity`, `Credit Amount`, `Current Debt`, `Total Debt`, `Total Settled`, `Number of Payments`, `Total Outstanding Amount`, and `Total Outstanding Amount Estimated` offer detailed insights into the financial aspects and performance of each loan application.\n\nCategorical Data:\n- Educational background and marital status are captured in columns such as `Education Level`, `Education Level 2`, `Marital Status`, and `Marital Status 2`.\n- Other categorical data include `Request Type`, `Last Rejection Reason`, and `Last Client Rejection Reason`.\n\nDays Overdue:\n- Specific columns like `120 Days`, `180 Days`, `30 Days`, `360 Days`, and `90 Days` provide a detailed breakdown of the number of days payments are overdue.\n\nThis refined DataFrame is prepared for an in-depth exploratory data analysis, allowing for a comprehensive understanding of the factors influencing loan defaults, which is crucial for developing robust predictive models.\n\n**Graphical Plots of Data**\n\nGraphical plots are indispensable tools in exploratory data analysis (EDA), offering visual insights that are often difficult to capture through numerical summaries alone. The use of histograms and density plots for numerical data allows us to observe the distribution patterns, identify skewness, and detect the presence of multiple modes within the data. These visualizations help in understanding the spread and central tendencies of the variables, providing a clear picture of their overall behavior. Boxplots are particularly useful for identifying outliers, as they highlight the interquartile range and the extremities of the data, facilitating the detection of anomalies that might skew the analysis or affect model performance. Scatter plots, on the other hand, are invaluable for visualizing relationships between pairs of numerical variables, making it easier to spot trends, clusters, and potential correlations. Heatmaps, used for both correlation analysis and temporal data visualization, offer a comprehensive view of how variables interact with each other and change over time. These advanced visualizations help in uncovering hidden patterns and dependencies that are crucial for building robust predictive models.\n\nInteractive dashboards further enhance the analytical process by providing a dynamic platform for data exploration. These dashboards allow users to filter, drill down, and visualize data in real-time, offering an intuitive way to interact with the dataset. This interactivity not only makes the analysis more engaging but also enables the identification of insights that might be missed in static plots. For temporal data, time series plots are particularly effective, as they illustrate the evolution of variables over time, revealing trends, seasonality, and temporal anomalies. By leveraging a combination of these graphical tools, the EDA process becomes more comprehensive, providing a deeper understanding of the dataset's structure and relationships, which is essential for informed decision-making and effective model building.\n\n**Descriptive Statistics of Data**\n\nDescriptive statistics play a pivotal role in summarizing and understanding the fundamental characteristics of the dataset. These statistics provide concise numerical summaries that describe the central tendency, dispersion, and shape of the data distribution. Measures such as mean, median, and mode offer insights into the central location of the data, highlighting the average or most common values. Standard deviation, variance, and interquartile range are critical for assessing the spread and variability, indicating how much the data deviates from the central value. Skewness and kurtosis further describe the asymmetry and peakedness of the distribution, respectively, providing a more nuanced understanding of the data's shape.\n\nAdditionally, for categorical data, frequency distributions and proportions are calculated to understand the distribution of different categories within each variable. Cross-tabulations, combined with Chi-square tests, are employed to examine associations between categorical variables, revealing potential dependencies and interactions. Descriptive statistics also extend to the analysis of missing data, where the percentage of missing values is calculated to assess the completeness of the dataset. By identifying patterns and proportions of missing data, we can make informed decisions on data imputation or exclusion strategies. Overall, descriptive statistics offer a foundational understanding of the dataset, enabling us to identify key characteristics, detect anomalies, and guide the subsequent steps in the data analysis and modeling process.","metadata":{}},{"cell_type":"markdown","source":"###4. Model Building
","metadata":{}},{"cell_type":"markdown","source":"The overall process involves data preparation, model definition, and evaluation. Initially, data features are engineered, and memory usage is optimized by converting data types. The data is then converted to a Pandas DataFrame with categorical columns appropriately typed. A custom voting model class is defined to average predictions from multiple estimators. The model is trained and evaluated using stratified group k-fold cross-validation, ensuring stratified and grouped splits. LightGBM models with specified hyperparameters are trained on each fold, and their predictions are averaged. Model performance is assessed through cross-validation scores, feature importance analysis, ROC curves, precision-recall curves, and confusion matrices, ultimately combining the models to produce stable and reliable scores.","metadata":{}},{"cell_type":"code","source":"# Define a custom voting model class\nclass VotingModel(BaseEstimator, RegressorMixin):\n def __init__(self, estimators):\n super().__init__()\n self.estimators = estimators\n\n # The fit method does not perform any training, as the estimators are already trained\n def fit(self, X, y=None):\n return self\n\n # Predict by averaging the predictions of each estimator\n def predict(self, X):\n y_preds = [estimator.predict(X) for estimator in self.estimators]\n return np.mean(y_preds, axis=0)\n\n # Predict probabilities by averaging the probabilities predicted by each estimator\n def predict_proba(self, X):\n y_preds = [estimator.predict_proba(X) for estimator in self.estimators]\n return np.mean(y_preds, axis=0)\n \ndef build_model(params):\n \"\"\"\n Builds a LightGBM model with the given parameters.\n \n Parameters:\n - params (dict): A dictionary of LightGBM parameters.\n \n Returns:\n - model (lgb.LGBMClassifier): The LightGBM model.\n \"\"\"\n model = lgb.LGBMClassifier(**params)\n return model","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.320201Z","iopub.status.idle":"2024-06-01T09:33:23.320553Z","shell.execute_reply.started":"2024-06-01T09:33:23.320388Z","shell.execute_reply":"2024-06-01T09:33:23.320409Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X = df_train.drop(columns=[\"target\", \"case_id\",\"WEEK_NUM\"])\ny = df_train[\"target\"]\nweeks = df_train[\"WEEK_NUM\"]\ncategorical_features = df_train.select_dtypes(include=['object', 'category']).columns.tolist()","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.322175Z","iopub.status.idle":"2024-06-01T09:33:23.322521Z","shell.execute_reply.started":"2024-06-01T09:33:23.322368Z","shell.execute_reply":"2024-06-01T09:33:23.322381Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Define parameters for the LightGBM model\nparams_initial = {\n \"boosting_type\": \"gbdt\",\n \"objective\": \"binary\",\n \"metric\": \"auc\",\n \"max_depth\": 10,\n \"learning_rate\": 0.05,\n \"max_bin\": 255,\n \"n_estimators\": 1200,\n \"colsample_bytree\": 0.8,\n \"colsample_bynode\": 0.8,\n \"verbose\": -1,\n \"random_state\": 42,\n \"reg_alpha\": 0.1,\n \"reg_lambda\": 10,\n \"extra_trees\": True,\n 'num_leaves': 64,\n \"device\": \"gpu\", \n}\n\n# Parameters for cross-validated model\nparams_cv = params_initial\n\n# Define cross-validation strategy with stratified groups\ncv = StratifiedGroupKFold(n_splits=5, shuffle=False)\n\n# Build and evaluate the initial model\ninitial_model = build_model(params_initial)\ninitial_model.fit(X, y)","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.325202Z","iopub.status.idle":"2024-06-01T09:33:23.325909Z","shell.execute_reply.started":"2024-06-01T09:33:23.325663Z","shell.execute_reply":"2024-06-01T09:33:23.325682Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_test = df_test.drop(columns=[\"WEEK_NUM\"])\nX_test = X_test.set_index(\"case_id\")\n\nfor col in X_test.select_dtypes(include=['float16']).columns:\n X_test[col] = X_test[col].astype(np.float32)\n\n# Align categorical features\nfor cat_col in categorical_features:\n X_test[cat_col] = pd.Categorical(X_test[cat_col], categories=X[cat_col].cat.categories)\n \ncommon_columns = X.columns\n# Ensure columns are aligned\nX_test = X_test[common_columns]\n\n# Predict with the initial model\ninitial_pred = initial_model.predict_proba(X_test)[:, 1]","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.327292Z","iopub.status.idle":"2024-06-01T09:33:23.327757Z","shell.execute_reply.started":"2024-06-01T09:33:23.327514Z","shell.execute_reply":"2024-06-01T09:33:23.327534Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"1. Data Partitioning: The data is partitioned using StratifiedGroupKFold cross-validation, which ensures that each fold is stratified, maintaining the same distribution of target classes, and grouped by a specific identifier (e.g., weeks) to prevent data leakage.\n\n2. Model Selection: The model chosen is the LightGBM classifier, a gradient boosting framework that uses tree-based learning algorithms. It's selected for its efficiency and high performance on structured datasets.\n\n3. Model Training: The LightGBM model is trained using the training data in each fold of the cross-validation. The model parameters, such as learning rate, max depth, and number of estimators, are set beforehand. During training, early stopping and evaluation metrics like AUC are used to monitor and tune the model's performance.\n\n4. Feature Importance: The attributes that have the greatest effect on the results are identified through feature importance analysis. Features are ranked based on their average importance across all folds, with the most important features contributing significantly to the model's predictive power. The top features are plotted to visually assess their impact.","metadata":{}},{"cell_type":"markdown","source":"###5. Model Validation
","metadata":{}},{"cell_type":"markdown","source":"The cross-validation process involves splitting the dataset into several folds to assess the performance of the model more robustly. Specifically, a StratifiedGroupKFold cross-validation strategy is employed, which ensures that each fold is representative of the overall distribution of the target variable and that the grouping variable (e.g., weeks) is respected, preventing data leakage between folds. For each split, the model is trained on the training set and validated on the validation set, capturing the variability of the model's performance across different subsets of data. This iterative process helps to mitigate overfitting and provides a more accurate estimate of the model's predictive capabilities. After fitting the model on each fold, the predictions are averaged to create an ensemble model, which generally improves the robustness and performance of the final predictions. This approach ensures that the model is evaluated comprehensively before being applied to unseen test data.","metadata":{}},{"cell_type":"code","source":"def cross_validate_model(X, y, weeks, model, cv):\n \"\"\"\n Performs cross-validation on the given model using the provided data and cross-validation strategy.\n \n Parameters:\n - X (pd.DataFrame): The feature matrix.\n - y (pd.Series): The target variable.\n - weeks (pd.Series): The week numbers for stratified group k-fold.\n - model (lgb.LGBMClassifier): The LightGBM model to be cross-validated.\n - cv (StratifiedGroupKFold): The cross-validation strategy.\n \n Returns:\n - fitted_models (list): List of fitted models.\n - cv_scores (list): List of cross-validation scores.\n - feature_importance_df (pd.DataFrame): DataFrame of feature importance.\n \"\"\"\n fitted_models = []\n cv_scores = []\n feature_importance_df = pd.DataFrame()\n \n for idx_train, idx_valid in cv.split(X, y, groups=weeks):\n X_train, y_train = X.iloc[idx_train], y.iloc[idx_train]\n X_valid, y_valid = X.iloc[idx_valid], y.iloc[idx_valid]\n\n # Ensure the validation set has the same columns as the training set\n common_columns = X_train.columns\n X_valid = X_valid[common_columns]\n \n print(\"Valid week range: \", (weeks.iloc[idx_valid].min(), weeks.iloc[idx_valid].max()))\n\n # Train the model\n model.fit(\n X_train, y_train,\n eval_set=[(X_valid, y_valid)],\n callbacks=[lgb.log_evaluation(50), lgb.early_stopping(50)]\n )\n\n # Append the trained model to the list of fitted models\n fitted_models.append(model)\n \n # Predict probabilities for the validation set\n y_pred_valid = model.predict_proba(X_valid)[:, 1]\n \n # Calculate the AUC score for the validation set\n auc_score = roc_auc_score(y_valid, y_pred_valid)\n cv_scores.append(auc_score)\n\n # Collect feature importance data\n fold_importance_df = pd.DataFrame()\n fold_importance_df[\"feature\"] = model.feature_name_\n fold_importance_df[\"importance\"] = model.feature_importances_\n fold_importance_df[\"fold\"] = len(cv_scores)\n feature_importance_df = pd.concat([feature_importance_df, fold_importance_df], axis=0)\n \n return fitted_models, cv_scores, feature_importance_df","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.329570Z","iopub.status.idle":"2024-06-01T09:33:23.329909Z","shell.execute_reply.started":"2024-06-01T09:33:23.329740Z","shell.execute_reply":"2024-06-01T09:33:23.329754Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Perform cross-validation and evaluate the cross-validated ensemble model\nfitted_models, cv_scores, feature_importance_df = cross_validate_model(X, y, weeks, build_model(params_cv), cv)\nensemble_model = VotingModel(fitted_models)\n\n# Predictions for cross-validated ensemble model\nensemble_pred = ensemble_model.predict_proba(X_test)[:, 1]","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.331068Z","iopub.status.idle":"2024-06-01T09:33:23.331401Z","shell.execute_reply.started":"2024-06-01T09:33:23.331236Z","shell.execute_reply":"2024-06-01T09:33:23.331253Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Aggregate feature importance\nfeature_importance_df = feature_importance_df.groupby(\"feature\").mean().sort_values(by=\"importance\", ascending=False)\n\n# Adjust the figsize to increase the height\nplt.figure(figsize=(14, 70))\nplt.title('LightGBM Features (avg over folds)')\nsns.barplot(x=\"importance\", y=\"feature\", data=feature_importance_df.reset_index())\nplt.show()\n\n# Print the most important features\nprint(\"Top 10 features:\")\nprint(feature_importance_df.head(10))","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.333128Z","iopub.status.idle":"2024-06-01T09:33:23.333583Z","shell.execute_reply.started":"2024-06-01T09:33:23.333351Z","shell.execute_reply":"2024-06-01T09:33:23.333369Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"1. lastapprcommoditycat_1041M (Importance: 882.5):\n\nThis feature likely represents the category of the last approved commodity, encoded as a string and then converted to a categorical type.\nHigh importance suggests that the type of commodity recently approved plays a crucial role in predicting the target variable, possibly indicating customer behavior or preferences.\n\n2. max_relationshiptoclient_642T (Importance: 760.5):\n\nThis feature could describe a specific relationship type to a client, encoded as a numerical or categorical value.\nThe model finds this relationship significant, perhaps reflecting customer loyalty, engagement, or other relational attributes that impact the prediction.\n\n3. max_incometype_1044T (Importance: 751.0):\n\nIndicates the maximum value of a specific income type for a client.\nThis feature's importance suggests that the nature and amount of income influence the likelihood of the target event, likely related to financial stability or purchasing power.\n\n4. max_relationshiptoclient_415T (Importance: 719.5):\n\nAnother relationship to client metric, reinforcing the importance of client relationships in model predictions.\nDifferent from the previous relationship metric, indicating that various aspects of client relationships are essential.\n\n5. price_1097A (Importance: 656.5):\n\nRepresents a specific price attribute, likely related to a product or service.\nThe importance of price suggests that cost-related factors significantly affect the target variable, highlighting the economic considerations in the model.","metadata":{}},{"cell_type":"markdown","source":"###6. Model Evaluation and Validation
","metadata":{}},{"cell_type":"markdown","source":"The model evaluation process involves several steps to comprehensively assess the performance of the trained models. After fitting the models using cross-validation, predictions are generated on the validation sets. The true labels and predicted probabilities from all validation sets are aggregated to evaluate the model's performance across multiple metrics. Specifically, the evaluation includes plotting the Receiver Operating Characteristic (ROC) curve to visualize the trade-off between the true positive rate and false positive rate, and calculating the Area Under the ROC Curve (AUC) to quantify the overall performance. The Precision-Recall curve is also plotted to illustrate the trade-off between precision and recall, with the Average Precision (AP) score providing a summary metric. Additionally, a confusion matrix is constructed to display the counts of true positive, true negative, false positive, and false negative predictions, giving insights into the model's classification accuracy. Finally, a calibration curve is plotted to assess the reliability of predicted probabilities, comparing the fraction of positives to the mean predicted values. These visualizations and metrics collectively provide a detailed understanding of the model's strengths and areas for improvement.","metadata":{}},{"cell_type":"code","source":"def plot_roc_curve(y_true, y_pred_proba, title=\"ROC Curve\"):\n fpr, tpr, _ = roc_curve(y_true, y_pred_proba)\n roc_auc = auc(fpr, tpr)\n\n plt.figure(figsize=(8, 6))\n plt.plot(fpr, tpr, color='darkorange', lw=2, label=f'ROC curve (area = {roc_auc:.2f})')\n plt.plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--')\n plt.xlim([0.0, 1.0])\n plt.ylim([0.0, 1.05])\n plt.xlabel('False Positive Rate')\n plt.ylabel('True Positive Rate')\n plt.title(title)\n plt.legend(loc=\"lower right\")\n plt.show()\n \ndef plot_precision_recall_curve(y_true, y_pred_proba, title=\"Precision-Recall Curve\"):\n precision, recall, _ = precision_recall_curve(y_true, y_pred_proba)\n avg_precision = average_precision_score(y_true, y_pred_proba)\n\n plt.figure(figsize=(8, 6))\n plt.plot(recall, precision, color='b', lw=2, label=f'Precision-Recall curve (AP = {avg_precision:.2f})')\n plt.xlim([0.0, 1.0])\n plt.ylim([0.0, 1.05])\n plt.xlabel('Recall')\n plt.ylabel('Precision')\n plt.title(title)\n plt.legend(loc=\"lower left\")\n plt.show()\n \ny_true_all = np.array([])\ny_pred_proba_all = np.array([])\n\nfor idx, (idx_train, idx_valid) in enumerate(cv.split(X, y, groups=weeks)):\n X_valid, y_valid = X.iloc[idx_valid], y.iloc[idx_valid]\n y_pred_valid = fitted_models[idx].predict_proba(X_valid)[:, 1]\n \n y_true_all = np.concatenate([y_true_all, y_valid])\n y_pred_proba_all = np.concatenate([y_pred_proba_all, y_pred_valid])\n\n# Plot ROC curve\nplot_roc_curve(y_true_all, y_pred_proba_all, title=\"Cross-Validation ROC Curve\")\n\n# Plot Precision-Recall curve\nplot_precision_recall_curve(y_true_all, y_pred_proba_all, title=\"Cross-Validation Precision-Recall Curve\")","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.335126Z","iopub.status.idle":"2024-06-01T09:33:23.335627Z","shell.execute_reply.started":"2024-06-01T09:33:23.335372Z","shell.execute_reply":"2024-06-01T09:33:23.335395Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def plot_confusion_matrix(y_true, y_pred, title=\"Confusion Matrix\"):\n cm = confusion_matrix(y_true, y_pred)\n disp = ConfusionMatrixDisplay(confusion_matrix=cm)\n disp.plot(cmap=plt.cm.Blues)\n plt.title(title)\n plt.show()\n\n# Convert probabilities to binary predictions with a threshold of 0.5\ny_pred_all = (y_pred_proba_all >= 0.5).astype(int)\nplot_confusion_matrix(y_true_all, y_pred_all, title=\"Cross-Validation Confusion Matrix\")","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.336941Z","iopub.status.idle":"2024-06-01T09:33:23.337426Z","shell.execute_reply.started":"2024-06-01T09:33:23.337162Z","shell.execute_reply":"2024-06-01T09:33:23.337182Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def plot_calibration_curve(y_true, y_pred_proba, title=\"Calibration Curve\"):\n fraction_of_positives, mean_predicted_value = calibration_curve(y_true, y_pred_proba, n_bins=10)\n \n plt.figure(figsize=(8, 6))\n plt.plot(mean_predicted_value, fraction_of_positives, \"s-\", label=\"Calibration curve\")\n plt.plot([0, 1], [0, 1], \"k--\", label=\"Perfectly calibrated\")\n plt.xlim([0.0, 1.0])\n plt.ylim([0.0, 1.0])\n plt.xlabel('Mean predicted value')\n plt.ylabel('Fraction of positives')\n plt.title(title)\n plt.legend(loc=\"lower right\")\n plt.show()\n\nplot_calibration_curve(y_true_all, y_pred_proba_all, title=\"Cross-Validation Calibration Curve\")\n","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.339044Z","iopub.status.idle":"2024-06-01T09:33:23.339498Z","shell.execute_reply.started":"2024-06-01T09:33:23.339266Z","shell.execute_reply":"2024-06-01T09:33:23.339285Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**Performance of the Model**","metadata":{}},{"cell_type":"markdown","source":"1. ROC Curve and AUC Score:\nThe ROC curve plots the true positive rate (TPR) against the false positive rate (FPR) at various threshold settings.\nThe AUC (Area Under the Curve) score is 0.85, indicating that the model has good discriminative ability. An AUC score closer to 1.0 suggests excellent performance, while a score around 0.5 would imply random guessing.\nTherefore, an AUC of 0.85 signifies that the model is effective at distinguishing between the positive and negative classes.\n\n2. Precision-Recall Curve and Average Precision Score:\nThe Precision-Recall curve plots precision (positive predictive value) against recall (true positive rate).\nThe average precision score of 0.19 suggests that the model has moderate performance in terms of balancing precision and recall.\nGiven that precision-recall curves are particularly useful for imbalanced datasets, this score might indicate that while the model has a reasonable ability to identify positive instances, there are challenges in maintaining high precision across all thresholds.\n\n3. Confusion Matrix:\nThe confusion matrix provides a breakdown of true positives, true negatives, false positives, and false negatives:\nTrue Negatives (TN): 1,500,000\nFalse Positives (FP): 814\nFalse Negatives (FN): 47,126\nTrue Positives (TP): 868\nThis indicates that while the model correctly identifies a large number of negative instances (high TN), it also misses a significant number of positive instances (high FN), and the number of false positives (FP) is relatively low.","metadata":{}},{"cell_type":"markdown","source":"**Producing Stable Scores**\n\n1. Cross-Validation:\nUsing StratifiedGroupKFold for cross-validation, as done in this model, ensures that the model is trained and evaluated on different subsets of the data. This helps assess the model's generalizability and reduces the risk of overfitting.\n\n2. Ensemble Methods:\nThe model uses an ensemble of LightGBM classifiers combined through a voting mechanism. Ensemble methods help reduce variance and increase stability by averaging the predictions of multiple models. This approach leverages the strengths of individual models while mitigating their weaknesses.","metadata":{}},{"cell_type":"markdown","source":"**Submission to Kaggle Competition**","metadata":{}},{"cell_type":"code","source":"def make_submission(test_df, models, submission_file):\n \"\"\"\n Create a submission file using the test set and a list of trained models.\n \n Parameters:\n - test_df (pd.DataFrame): The test set feature matrix.\n - models (list): List of trained models.\n - submission_file (str): The path where the submission file will be saved.\n \"\"\"\n # Initialize an array to store predictions\n predictions = np.zeros(len(test_df))\n \n # Make predictions using each model and aggregate them (e.g., by averaging)\n for model in models:\n predictions += model.predict_proba(test_df)[:, 1]\n predictions /= len(models)\n \n # Create a DataFrame for submission\n submission_df = pd.DataFrame({\n 'case_id': test_df.index,\n 'score': predictions\n })\n \n # Save the submission DataFrame to a CSV file\n submission_df.to_csv(submission_file, index=False)\n\n # Debugging: Check if the file has been created successfully\n print(f\"Submission file saved to {submission_file}\")\n print(pd.read_csv(submission_file).head())\n\n\nmake_submission(X_test, fitted_models, \"/kaggle/working/submission.csv\")","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.341057Z","iopub.status.idle":"2024-06-01T09:33:23.341393Z","shell.execute_reply.started":"2024-06-01T09:33:23.341229Z","shell.execute_reply":"2024-06-01T09:33:23.341246Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"###7. Discussion
","metadata":{}},{"cell_type":"markdown","source":"#### Identifying Influential Factors on Credit Risk and Stability\n\n1. **Factors Influencing Credit Risk and Its Stability**\n\n The assessment of credit risk involves evaluating various factors that significantly influence a client's likelihood of defaulting on a loan. These factors can be broadly categorized into demographic, financial, and behavioral aspects. \n\n - **Demographic Factors:**\n - **Age:** Younger applicants or those without an established credit history may present higher risks.\n - **Marital Status:** Married individuals might have more stable financial situations compared to single applicants.\n - **Education Level:** Higher education levels often correlate with better employment opportunities and financial literacy, reducing credit risk.\n\n - **Financial Factors:**\n - **Credit Amount:** Larger loan amounts increase the risk as they pose greater financial burdens on the borrower.\n - **Current Debt:** Higher levels of existing debt relative to income can indicate higher risk.\n - **Annuity and Payment Ratios:** Metrics like the ratio of total payments to annuity provide insight into the borrower’s financial commitment and repayment capacity.\n\n - **Behavioral Factors:**\n - **Payment History:** Timely payments and a lack of defaults in the past are strong indicators of lower credit risk.\n - **Number of Credit Inquiries:** Frequent inquiries might suggest financial distress or a higher propensity to take on more debt.\n - **Credit Utilization:** The proportion of available credit being used is crucial; higher utilization rates often signal higher risk.\n\n\n **Stability of Credit Risk:**\n The stability of credit risk is influenced by the consistency of these factors over time. Stable income, regular employment, and consistent repayment behaviors contribute to maintaining a low-risk profile. Volatile financial behaviors, frequent job changes, or significant variations in income can lead to fluctuations in credit risk, thereby affecting its stability.\n\n2. **Interesting Observations from the Challenge**\n\n - **Impact of Economic Conditions:** External economic factors, such as changes in the job market or economic downturns, can significantly impact credit risk across a broad segment of the population. During economic downturns, even reliable borrowers may face challenges, highlighting the importance of macroeconomic indicators in credit risk assessment.\n\n - **Importance of Data Integration:** Combining internal data (e.g., payment history, loan amounts) with external data sources (e.g., credit bureau reports, tax registry information) provides a more comprehensive view of the borrower's risk profile. This integration helps in identifying patterns that might not be apparent from a single data source.\n\n - **Predictive Power of Behavioral Data:** Behavioral data, such as payment patterns and the frequency of credit inquiries, often have a strong predictive power for credit risk. These behaviors can change rapidly and provide real-time insights into the borrower’s financial health, making them valuable for dynamic risk assessment models.\n\n - **Feature Engineering and Model Performance:** Effective feature engineering, including the creation of interaction features and aggregated metrics, significantly enhances model performance. Features that capture nuanced relationships between variables (e.g., the ratio of payments to annuity) provide deeper insights and improve the accuracy of credit risk predictions.\n\n - **Need for Model Stability:** In the real world, the stability of predictive models is as crucial as their performance. Models that require frequent retraining due to shifts in borrower behavior or economic conditions can be resource-intensive. Hence, developing models that balance accuracy with stability over time is essential for practical applications.\n\n These observations underline the complexity of credit risk assessment and the necessity for robust, multi-faceted approaches in developing reliable predictive models. By considering a wide range of factors and their interactions, as well as the external economic environment, financial institutions can better manage and mitigate credit risk.","metadata":{}},{"cell_type":"markdown","source":"#### Explaining Limitations and Weaknesses of the Modeling Approach\n\n1. **Limitations and Weaknesses of the Modeling Approach**\n\n - **Data Quality and Missing Values:** One of the primary limitations is the presence of missing values and inconsistencies in the dataset. Missing data can introduce biases and affect the accuracy of the model. Handling missing values through imputation or deletion might not always capture the true underlying patterns.\n\n - **Feature Selection and Engineering:** While feature engineering is crucial for improving model performance, it also introduces the risk of overfitting. Creating too many features, especially interaction terms, can lead to a model that performs well on the training data but fails to generalize to unseen data.\n\n - **Model Complexity:** More complex models, such as ensemble methods or deep learning approaches, can capture intricate patterns in the data but also require more computational resources and time. Additionally, these models can be difficult to interpret, making it challenging to understand the factors driving credit risk predictions.\n\n - **Temporal Stability:** Models trained on historical data might not perform well when applied to future data if the underlying patterns change over time. Economic conditions, regulatory changes, and shifts in borrower behavior can all impact the stability of the model’s performance.\n\n - **Bias and Fairness:** There is a risk that the model may inadvertently incorporate biases present in the training data, leading to unfair predictions for certain groups of borrowers. Ensuring fairness and mitigating bias is a significant challenge in credit risk modeling.\n\n - **Evaluation Metrics:** While traditional metrics like AUC-ROC are useful, they might not fully capture the practical implications of the model’s predictions. For instance, they do not account for the financial impact of false positives and false negatives, which is crucial in credit risk assessment.\n\n2. **Steps Taken to Improve Matching Accuracy in the Modeling Approach**\n\n - **Comprehensive Data Cleaning:** Extensive data cleaning procedures were implemented to address missing values and inconsistencies. Techniques such as mean imputation, median imputation, and even more sophisticated methods like K-Nearest Neighbors imputation were used to ensure that the data quality was as high as possible.\n\n - **Robust Feature Engineering:** Careful feature engineering was carried out to create meaningful interaction terms and aggregated features. Techniques such as mean encoding for categorical variables and creating ratios for numerical variables were employed to enhance the model’s predictive power.\n\n - **Cross-Validation:** To ensure the model’s robustness and prevent overfitting, cross-validation techniques were used during model training. This approach helps in assessing the model’s performance on different subsets of the data and ensures that it generalizes well to unseen data.\n\n - **Ensemble Methods:** Ensemble methods like Random Forest, Gradient Boosting Machines, and XGBoost were used to improve prediction accuracy. These methods combine the strengths of multiple models, reducing the risk of overfitting and capturing a broader range of patterns in the data.\n\n - **Hyperparameter Tuning:** Systematic hyperparameter tuning using techniques like grid search and random search was conducted to find the optimal parameters for the models. This process helps in improving the model’s performance by fine-tuning the settings that control the learning process.\n\n - **Addressing Temporal Changes:** To account for potential changes over time, the model was regularly retrained with recent data. Additionally, techniques like time series cross-validation were used to ensure that the model’s performance remains stable across different time periods.\n\n - **Bias Mitigation:** Steps were taken to identify and mitigate biases in the model. This included analyzing model predictions for different demographic groups and implementing fairness constraints where necessary to ensure equitable treatment of all borrowers.\n\n - **Advanced Evaluation Metrics:** Beyond traditional metrics, the model’s performance was also evaluated using metrics that consider financial impact, such as the expected cost of false positives and false negatives. This approach ensures that the model’s predictions are not only accurate but also practically relevant for financial decision-making.\n\nBy addressing these limitations and implementing these steps, the modeling approach aimed to achieve a balance between accuracy, interpretability, and fairness, ensuring a robust and reliable credit risk assessment model.","metadata":{}},{"cell_type":"markdown","source":"#### Elaborating on the Experience in Participating in a Kaggle Challenge\n\n1. **The Experience in Participating in a Kaggle Challenge**\n\n Participating in a Kaggle challenge is an enriching experience that offers a blend of learning, competition, and collaboration. The process begins with understanding the problem statement and correctly collecting data from the provided dataset. The test and train data, which has already been provided in this competition, is then stored into their separate variables. Furthermore, the train data is then refined, which involves trimming the dataset down to only the important columns that will be used for exploratory data analysis (EDA).\n\n The initial steps involve extensive exploratory data analysis (EDA) to gain insights into the data and identify potential features that could enhance model performance. This phase is crucial for understanding data distributions, relationships between variables, and any data quality issues. Kaggle's discussion forums and kernels (notebooks) are invaluable resources during this phase, offering insights from other participants and best practices for data preprocessing and feature engineering.\n\n The modeling phase involves selecting appropriate algorithms, fine-tuning hyperparameters, and iterating over different model configurations to improve performance. Kaggle's platform provides real-time feedback through leaderboard updates, which is both motivating and challenging. It pushes participants to constantly refine their models and approaches.\n\n Collaboration is another key aspect of the Kaggle experience. Participants often form teams to combine their skills and knowledge, resulting in more robust solutions. The community is supportive, with experienced Kagglers frequently sharing their knowledge through notebooks that provide basic guidelines on handling the dataset and submitting results, thereby helping others improve.\n\n Overall, participating in a Kaggle challenge is a journey of continuous learning, problem-solving, and skill enhancement, providing an opportunity to work on real-world data science problems in a competitive yet collaborative environment.\n\n2. **The Discussion and Submission Score on Kaggle**\n\n Throughout the Kaggle challenge, active participation in discussions and sharing insights with the community is crucial. Engaging in discussions helps clarify doubts, gain new perspectives, and learn from the experiences of others. Sharing intermediate results and approaches can lead to valuable feedback and further refinement of the solution.\n\n Submission scores on Kaggle are updated in real-time on the leaderboard, providing a measure of how well the model performs compared to others. Here is a link to the submission, providing an overview of the performance and ranking: [Kaggle Submission Link](https://www.kaggle.com/code/thadted/cos20083-advanced-data-analytics-group-3?scriptVersionId=180073170). This link leads to the submission page, where detailed scores and the leaderboard position can be viewed.\n\n\n**Public Leaderboard**\n \n \n \n**Private Leaderboard**\n \n\n3. **The Improvements That Need to Be Done in Order to Win the Challenge**\n\n To win the Kaggle challenge, several improvements can be implemented, focusing on both machine learning techniques and metric optimization strategies.\n\n#### Machine Learning Improvements\n\n1. **Feature Engineering**: Enhance feature engineering efforts to create more informative features that capture hidden patterns and relationships in the data. This can include creating interaction features, polynomial features, and using domain knowledge to derive new features.\n\n2. **Model Ensemble**: Experiment with different ensemble methods, such as stacking, boosting, and bagging, to combine the strengths of multiple models. This approach can help improve the model's robustness and generalization ability.\n\n3. **Hyperparameter Tuning**: Utilize advanced hyperparameter tuning techniques like Bayesian Optimization, Random Search, and Grid Search to fine-tune the model parameters for optimal performance.\n\n4. **Handling Imbalanced Data**: Implement strategies to handle class imbalance, such as using Synthetic Minority Over-sampling Technique (SMOTE), adjusting class weights, or employing anomaly detection methods to better identify rare events.\n\n5. **Model Interpretability**: Use model interpretability tools such as SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations) to understand the impact of each feature on the model's predictions. This can provide insights into feature importance and guide further feature engineering efforts.\n\n#### Metric Optimization (Metric Hack)\n\n1. **Custom Objective Function**: Implement a custom objective function that directly optimizes the competition metric. This can help align the model training process with the evaluation criteria used in the challenge.\n\n2. **Threshold Tuning**: Optimize the decision threshold for classification to balance precision and recall based on the specific requirements of the competition metric.\n\n3. **Cross-Validation Strategy**: Use a robust cross-validation strategy, such as Stratified K-Fold or Time Series Split, to ensure that the model performs well across different subsets of the data. This helps in achieving stable and reliable performance.\n\n4. **Post-Processing**: Apply post-processing techniques to the model predictions to improve the final scores. This can include techniques like calibration, adjusting probability thresholds, and using ensemble predictions for final submission.\n\n5. **Model Blending**: Blend the predictions from multiple models using weighted averaging or other blending techniques to leverage the strengths of each model and reduce variance.\n\n By implementing these improvements, one can enhance the performance of their machine learning models and optimize their approach to align better with the competition metrics, thereby increasing the chances of winning the Kaggle challenge.","metadata":{},"attachments":{"07e7005e-4378-4fdf-b2ae-34f401d07a08.png":{"image/png":"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"},"0beb5b73-aca6-40c9-bef6-a311c3d594ad.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"# Team contribution","metadata":{}},{"cell_type":"raw","source":"# Please describe the contribution of each team member e.g. participation percentage and tasks \n(1) Thaddeus Chong Zhuo Liang (50% and 2,4,5,6)\n(2) Wai Hpone (50% and 1,2,3,7)","metadata":{"vscode":{"languageId":"raw"}}},{"cell_type":"code","source":"# References\nhttps://www.kaggle.com/code/zulqarnainalipk/explained-home-credit-pipeline\nhttps://www.kaggle.com/code/harrychan123/lgb-cat-ensemble-stacking/notebook","metadata":{"execution":{"iopub.status.busy":"2024-06-01T09:33:23.343141Z","iopub.status.idle":"2024-06-01T09:33:23.343594Z","shell.execute_reply.started":"2024-06-01T09:33:23.343366Z","shell.execute_reply":"2024-06-01T09:33:23.343385Z"},"trusted":true},"execution_count":null,"outputs":[]}]}