From 267a2b8013439d45978a81db5e296610c9b329d3 Mon Sep 17 00:00:00 2001 From: dorian Date: Wed, 19 Jun 2024 09:17:43 +0200 Subject: [PATCH 1/2] Adding managing missing values --- src/back/managing_missing_values.py | 83 +++++++++++++++++++++++++++++ 1 file changed, 83 insertions(+) create mode 100644 src/back/managing_missing_values.py diff --git a/src/back/managing_missing_values.py b/src/back/managing_missing_values.py new file mode 100644 index 0000000..745641d --- /dev/null +++ b/src/back/managing_missing_values.py @@ -0,0 +1,83 @@ +import pandas as pd +from sklearn.impute import KNNImputer +from sklearn.linear_model import LinearRegression +import numpy as np +import load_csv as l + +def convert_categorical_to_numeric(data): + for column in data.columns: + if data[column].nunique() <= 15: + data[column] = data[column].astype('category') + data[column] = data[column].cat.codes.replace(-1, np.nan) + 1 + else: + data = data.drop(column, axis=1) + return data + +def drop_high_null_percentage(data, threshold=0.5): + missing_percentage = data.isnull().mean() + data = data.loc[:, missing_percentage <= threshold] + return data + + + +def replace_with_mean(data, column): + data[column] = data[column].fillna(data[column].mean()) + return data + +def replace_with_median(data, column): + data[column] = data[column].fillna(data[column].median()) + return data + +def replace_with_mode(data, column): + mode_value = data[column].mode() + if not mode_value.empty: + data[column] = data[column].fillna(mode_value[0]) + return data + +def impute_with_knn(data, column, n_neighbors=5): + imputer = KNNImputer(n_neighbors=n_neighbors) + data[[column]] = imputer.fit_transform(data[[column]]) + return data + +def impute_with_regression(data, column): + if data[column].isnull().sum() > 0: + train_data = data[data[column].notna()] + test_data = data[data[column].isna()] + if not train_data.empty and not test_data.empty: + regressor = LinearRegression() + regressor.fit(train_data.drop(column, axis=1), train_data[column]) + data.loc[data[column].isna(), column] = regressor.predict(test_data.drop(column, axis=1)) + return data + + +""" + Parameters: + - data: Pandas DataFrame with the data + - method: Method to handle missing values ('drop', 'mean', 'median', 'mode', 'knn', 'regression') + - n_neighbors: Number of neighbors to use for KNN imputation (only used if method='knn') +""" +def handle_missing_values(data, method, column, n_neighbors=5): + + data = drop_high_null_percentage(data) + data = convert_categorical_to_numeric(data) + + if method == 'mean': + return replace_with_mean(data, column) + elif method == 'median': + return replace_with_median(data, column) + elif method == 'mode': + return replace_with_mode(data, column) + elif method == 'knn': + return impute_with_knn(data, column, n_neighbors=n_neighbors) + elif method == 'regression': + return impute_with_regression(data, column) + elif method == 'drop_high_null': + return drop_high_null_percentage(data) + else: + raise ValueError("Unknown method") + + + +data = l.return_csv('./data.csv') +cleaned_data = handle_missing_values(data, method='mode', column='Route Type') +print(cleaned_data) From 022437dad86dabf2371569dec55284c34150e506 Mon Sep 17 00:00:00 2001 From: dorian Date: Wed, 19 Jun 2024 09:22:48 +0200 Subject: [PATCH 2/2] Mean median and other function effect all data and now only column, removing useless columns --- src/back/managing_missing_values.py | 62 +++++++++++------------------ 1 file changed, 24 insertions(+), 38 deletions(-) diff --git a/src/back/managing_missing_values.py b/src/back/managing_missing_values.py index 745641d..8360cf1 100644 --- a/src/back/managing_missing_values.py +++ b/src/back/managing_missing_values.py @@ -19,34 +19,28 @@ def drop_high_null_percentage(data, threshold=0.5): return data +def replace_with_mean(data): + return data.apply(lambda col: col.fillna(col.mean()) if col.dtype.kind in 'biufc' else col) -def replace_with_mean(data, column): - data[column] = data[column].fillna(data[column].mean()) - return data - -def replace_with_median(data, column): - data[column] = data[column].fillna(data[column].median()) - return data +def replace_with_median(data): + return data.apply(lambda col: col.fillna(col.median()) if col.dtype.kind in 'biufc' else col) -def replace_with_mode(data, column): - mode_value = data[column].mode() - if not mode_value.empty: - data[column] = data[column].fillna(mode_value[0]) - return data +def replace_with_mode(data): + return data.apply(lambda col: col.fillna(col.mode()[0]) if col.mode().size > 0 else col) -def impute_with_knn(data, column, n_neighbors=5): +def impute_with_knn(data, n_neighbors=5): imputer = KNNImputer(n_neighbors=n_neighbors) - data[[column]] = imputer.fit_transform(data[[column]]) - return data + return pd.DataFrame(imputer.fit_transform(data), columns=data.columns) -def impute_with_regression(data, column): - if data[column].isnull().sum() > 0: - train_data = data[data[column].notna()] - test_data = data[data[column].isna()] - if not train_data.empty and not test_data.empty: - regressor = LinearRegression() - regressor.fit(train_data.drop(column, axis=1), train_data[column]) - data.loc[data[column].isna(), column] = regressor.predict(test_data.drop(column, axis=1)) +def impute_with_regression(data): + for column in data.columns: + if data[column].isnull().sum() > 0: + train_data = data[data[column].notna()] + test_data = data[data[column].isna()] + if not train_data.empty and not test_data.empty: + regressor = LinearRegression() + regressor.fit(train_data.drop(column, axis=1), train_data[column]) + data.loc[data[column].isna(), column] = regressor.predict(test_data.drop(column, axis=1)) return data @@ -56,28 +50,20 @@ def impute_with_regression(data, column): - method: Method to handle missing values ('drop', 'mean', 'median', 'mode', 'knn', 'regression') - n_neighbors: Number of neighbors to use for KNN imputation (only used if method='knn') """ -def handle_missing_values(data, method, column, n_neighbors=5): +def handle_missing_values(data, method, n_neighbors=5): data = drop_high_null_percentage(data) - data = convert_categorical_to_numeric(data) - + data = convert_categorical_to_numeric(data) if method == 'mean': - return replace_with_mean(data, column) + return replace_with_mean(data) elif method == 'median': - return replace_with_median(data, column) + return replace_with_median(data) elif method == 'mode': - return replace_with_mode(data, column) + return replace_with_mode(data) elif method == 'knn': - return impute_with_knn(data, column, n_neighbors=n_neighbors) + return impute_with_knn(data, n_neighbors=n_neighbors) elif method == 'regression': - return impute_with_regression(data, column) - elif method == 'drop_high_null': - return drop_high_null_percentage(data) + return impute_with_regression(data) else: raise ValueError("Unknown method") - - -data = l.return_csv('./data.csv') -cleaned_data = handle_missing_values(data, method='mode', column='Route Type') -print(cleaned_data)