{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Example usage\n",
"\n",
"To use `mds_2025_helper_functions` in a project:"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Imports"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"from mds_2025_helper_functions.scores import compare_model_scores\n",
"from sklearn.datasets import load_iris, load_diabetes\n",
"from sklearn.dummy import DummyRegressor, DummyClassifier\n",
"from sklearn.tree import DecisionTreeRegressor, DecisionTreeClassifier\n",
"import warnings\n",
"warnings.filterwarnings('ignore')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Compare CV scores of multiple models"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"`compare_model_scores()` is a wrapper function for scikit learn's `cross_validate()` that allows you to compare the mean cross validation scores across multiple models. The only difference in calling this function compared to `cross_validate()` is that it takes multiple model objects rather than one.\n",
"\n",
"Note: The default scoring metric is R² for regression and accuracy for classification tasks.\n",
"\n",
"### Basic usage\n",
"To demonstrate, let's load a sample dataset and instantiate our model classes. We'll be using the Diabetes dataset from scikit learn. The Diabetes dataset contains 10 baseline variables and progression of diabetes after one year. To learn more about this dataset, visit its documentation: https://scikit-learn.org/stable/datasets/toy_dataset.html#diabetes-dataset"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"X, y = load_diabetes(return_X_y=True)\n",
"dummy_regressor = DummyRegressor()\n",
"tree_regressor = DecisionTreeRegressor()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This is already enough for our basic use of the function. Simply pass these to `compare_model_scores()`.\n",
"\n",
"Note: The default scoring metric is R² for regression tasks. Negative R² scores indicate the model performs worse than predicting the mean value."
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
fit_time
\n",
"
score_time
\n",
"
test_score
\n",
"
\n",
"
\n",
"
model
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
DummyRegressor
\n",
"
0.000114
\n",
"
0.000162
\n",
"
-0.027506
\n",
"
\n",
"
\n",
"
DecisionTreeRegressor
\n",
"
0.001786
\n",
"
0.000196
\n",
"
-0.175689
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" fit_time score_time test_score\n",
"model \n",
"DummyRegressor 0.000114 0.000162 -0.027506\n",
"DecisionTreeRegressor 0.001786 0.000196 -0.175689"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"compare_model_scores(dummy_regressor, tree_regressor, X=X, y=y)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As you can see, the function returns a dataframe with the performance statistics for each model. The model names are used for the index.\n",
"\n",
"### Using `cross_validate()` arguments\n",
"Like `cross_validate`, the function also works for classification models, and you can pass arguments to reutrn training scores, or use different scoring metrics.\n",
"\n",
"For classification, we'll be using the Iris dataset from scikit learn. The Iris dataset contains measurements of iris flowers with 3 different species. To learn more about this dataset, visit its documentation: https://scikit-learn.org/stable/datasets/toy_dataset.html#iris-dataset"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
fit_time
\n",
"
score_time
\n",
"
test_score
\n",
"
train_score
\n",
"
\n",
"
\n",
"
model
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
DummyClassifier
\n",
"
0.000093
\n",
"
0.000614
\n",
"
0.166667
\n",
"
0.166667
\n",
"
\n",
"
\n",
"
DecisionTreeClassifier
\n",
"
0.000246
\n",
"
0.000511
\n",
"
0.966583
\n",
"
1.000000
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" fit_time score_time test_score train_score\n",
"model \n",
"DummyClassifier 0.000093 0.000614 0.166667 0.166667\n",
"DecisionTreeClassifier 0.000246 0.000511 0.966583 1.000000"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"X, y = load_iris(return_X_y=True)\n",
"dummy_classifier = DummyClassifier()\n",
"tree_classifier = DecisionTreeClassifier()\n",
"scoring_metric = \"f1_macro\" # A scoring metric for multiclass classification\n",
"\n",
"compare_model_scores(dummy_classifier, tree_classifier, X=X, y=y, return_train_scores=True, scoring=scoring_metric)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Passing multiple models of the same type\n",
"\n",
"When you compare several models of the same type, each model is be given an index in the output table based on the order it was passed to `compare_model_scores()`."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
fit_time
\n",
"
score_time
\n",
"
test_score
\n",
"
\n",
"
\n",
"
model
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
DecisionTreeClassifier
\n",
"
0.000306
\n",
"
0.000197
\n",
"
0.966667
\n",
"
\n",
"
\n",
"
DecisionTreeClassifier_2
\n",
"
0.000221
\n",
"
0.000153
\n",
"
0.960000
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" fit_time score_time test_score\n",
"model \n",
"DecisionTreeClassifier 0.000306 0.000197 0.966667\n",
"DecisionTreeClassifier_2 0.000221 0.000153 0.960000"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"second_tree_classifier = DecisionTreeClassifier(max_depth=3)\n",
"\n",
"compare_model_scores(tree_classifier, second_tree_classifier, X=X, y=y)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Perform exploratory data analysis (EDA)\n",
"\n",
"The `perform_eda` function provides a comprehensive exploratory data analysis (EDA) framework for any dataset. It combines summary statistics and feature visualizations, making it a valuable tool for understanding and exploring data.\n",
"\n",
"## Function Signature\n",
"\n",
"```python\n",
"perform_eda(dataframe, rows=5, cols=2)\n",
"```\n",
"\n",
"## Parameters\n",
"\n",
"| Parameter | Type | Default | Description |\n",
"|------------|----------------|----------|---------------------------------------------------|\n",
"| dataframe | `pd.DataFrame` | Required | Input dataset for EDA. Must be a Pandas DataFrame. |\n",
"| rows | `int` | `5` | Number of rows in the grid layout for visualizations. |\n",
"| cols | `int` | `2` | Number of columns in the grid layout for visualizations. |\n",
"\n",
"## Returns\n",
"\n",
"This function does not return a value. Instead, it:\n",
"\n",
"1. Prints a summary of the dataset.\n",
"2. Generates plots for missing values, correlations, and feature distributions.\n",
"3. Outputs potential outliers and scatterplots for numeric features.\n",
"\n",
"## Key Features\n",
"\n",
"1. **Dataset Overview**\n",
" - Prints dataset structure, number of rows/columns, and column data types.\n",
"\n",
"2. **Basic Statistics**\n",
" - Descriptive statistics for all numeric and categorical columns.\n",
" - Handles datasets with mixed data types.\n",
"\n",
"3. **Missing Values Report**\n",
" - Highlights columns with missing values.\n",
" - Displays a heatmap of missing data if applicable.\n",
"\n",
"4. **Correlation Heatmap**\n",
" - For numeric columns, it computes and visualizes pairwise correlations.\n",
"\n",
"5. **Dynamic Feature Visualizations**\n",
" - Automatically generates appropriate visualizations:\n",
" - Histograms and KDE plots for numeric features.\n",
" - Count plots for categorical features.\n",
" - Line plots for datetime features.\n",
"\n",
"6. **Scatterplots**\n",
" - Scatterplots for numeric feature pairs (if more than one numeric column exists).\n",
"\n",
"7. **Outliers Detection**\n",
" - Identifies potential outliers using the Interquartile Range (IQR) method.\n",
"\n",
"---\n",
"\n",
"## Example Usage\n",
"\n",
"### Dataset\n"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"import numpy as np\n",
"\n",
"np.random.seed(42) \n",
"\n",
"data = {\n",
" 'age': np.random.randint(20, 60, size=50), # Random ages between 20 and 60\n",
" 'salary': np.random.randint(30000, 120000, size=50), # Salaries between 30k and 120k\n",
" 'department': np.random.choice(['HR', 'Finance', 'IT', 'Marketing', 'Operations'], size=50), # Random departments\n",
" 'joining_date': pd.to_datetime(np.random.choice(pd.date_range('2010-01-01', '2022-01-01'), size=50)), # Random dates\n",
" 'experience': np.random.randint(1, 30, size=50), # Years of experience between 1 and 30\n",
" 'performance_score': np.random.uniform(1, 5, size=50), # Performance score between 1 and 5\n",
" 'bonus': np.random.choice([True, False], size=50, p=[0.3, 0.7]) # Random True/False for bonuses\n",
"}\n",
"\n",
"df = pd.DataFrame(data)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Run perform_eda"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"===== Dataset Overview =====\n",
"\n",
"RangeIndex: 50 entries, 0 to 49\n",
"Data columns (total 7 columns):\n",
" # Column Non-Null Count Dtype \n",
"--- ------ -------------- ----- \n",
" 0 age 50 non-null int64 \n",
" 1 salary 50 non-null int64 \n",
" 2 department 50 non-null object \n",
" 3 joining_date 50 non-null datetime64[ns]\n",
" 4 experience 50 non-null int64 \n",
" 5 performance_score 50 non-null float64 \n",
" 6 bonus 50 non-null bool \n",
"dtypes: bool(1), datetime64[ns](1), float64(1), int64(3), object(1)\n",
"memory usage: 2.5+ KB\n",
"None\n",
"\n",
"===== Basic Statistics =====\n",
" count unique top freq mean \\\n",
"age 50.0 NaN NaN NaN 39.04 \n",
"salary 50.0 NaN NaN NaN 77464.08 \n",
"department 50 5 Operations 12 NaN \n",
"joining_date 50 NaN NaN NaN 2016-06-11 07:12:00 \n",
"experience 50.0 NaN NaN NaN 14.74 \n",
"performance_score 50.0 NaN NaN NaN 3.370538 \n",
"bonus 50 2 False 31 NaN \n",
"\n",
" min 25% \\\n",
"age 21.0 30.0 \n",
"salary 31016.0 53510.25 \n",
"department NaN NaN \n",
"joining_date 2010-07-17 00:00:00 2013-10-28 00:00:00 \n",
"experience 1.0 7.0 \n",
"performance_score 1.072301 2.521356 \n",
"bonus NaN NaN \n",
"\n",
" 50% 75% \\\n",
"age 40.0 46.75 \n",
"salary 78587.0 99000.0 \n",
"department NaN NaN \n",
"joining_date 2016-01-08 12:00:00 2019-06-17 00:00:00 \n",
"experience 16.0 23.0 \n",
"performance_score 3.341115 4.399446 \n",
"bonus NaN NaN \n",
"\n",
" max std \n",
"age 59.0 11.347858 \n",
"salary 119812.0 27874.066306 \n",
"department NaN NaN \n",
"joining_date 2021-09-22 00:00:00 NaN \n",
"experience 29.0 9.222444 \n",
"performance_score 4.978202 1.080807 \n",
"bonus NaN NaN \n",
"\n",
"===== Missing Values Report =====\n",
"Series([], dtype: int64)\n",
"No missing values in the dataset.\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"===== Outliers Report =====\n",
"age: 0 potential outliers\n",
"salary: 0 potential outliers\n",
"experience: 0 potential outliers\n",
"performance_score: 0 potential outliers\n"
]
}
],
"source": [
"from mds_2025_helper_functions.eda import perform_eda\n",
"\n",
"perform_eda(df, rows=3, cols=3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Summarize a dataset"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The `dataset_summary` function provides a comprehensive summary of any dataset. It is designed to give a quick yet detailed overview of the dataset, focusing on missing values, feature types, duplicate rows, and descriptive statistics for both numerical and categorical features.\n",
"\n",
"### Function Signature\n",
"\n",
"```python\n",
"dataset_summary(data)\n",
"```\n",
"\n",
"### Parameters\n",
"\n",
" **`data`** \n",
"- **Type:** `pd.DataFrame` \n",
"- **Default:** Required \n",
"- **Description:** \n",
" - Input dataset to analyze and summarize. \n",
" - Must be a Pandas DataFrame. \n",
" - Supports both single-index and multi-index DataFrames (automatically flattened for processing). \n",
" - Sparse DataFrames are supported and converted to dense format during analysis. \n",
"\n",
"### Returns\n",
"\n",
"The function returns a dictionary containing the following keys:\n",
"\n",
"| Key | Type | Description |\n",
"|--------------------|----------------|--------------------------------------------------------------------------|\n",
"| `missing_values` | `pd.DataFrame` | A DataFrame summarizing the number and percentage of missing values. |\n",
"| `feature_types` | `dict` | Counts of numerical and categorical features: |\n",
"| | | `{'numerical_features': int, 'categorical_features': int}`. |\n",
"| `duplicates` | `int` | The number of duplicate rows in the dataset. |\n",
"| `numerical_summary`| `pd.DataFrame` | Descriptive statistics for numerical features. |\n",
"| `categorical_summary`| `pd.DataFrame`| Unique value counts for categorical features. |\n",
"\n",
"---\n",
"\n",
"## Example Usage\n",
"\n",
"### Imports and Dataset\n",
"\n",
"We use the Titanic dataset from the seaborn library for this demonstration. First, we import the necessary libraries and load the dataset."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
survived
\n",
"
pclass
\n",
"
sex
\n",
"
age
\n",
"
sibsp
\n",
"
parch
\n",
"
fare
\n",
"
embarked
\n",
"
class
\n",
"
who
\n",
"
adult_male
\n",
"
deck
\n",
"
embark_town
\n",
"
alive
\n",
"
alone
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
0
\n",
"
3
\n",
"
male
\n",
"
22.0
\n",
"
1
\n",
"
0
\n",
"
7.2500
\n",
"
S
\n",
"
Third
\n",
"
man
\n",
"
True
\n",
"
NaN
\n",
"
Southampton
\n",
"
no
\n",
"
False
\n",
"
\n",
"
\n",
"
1
\n",
"
1
\n",
"
1
\n",
"
female
\n",
"
38.0
\n",
"
1
\n",
"
0
\n",
"
71.2833
\n",
"
C
\n",
"
First
\n",
"
woman
\n",
"
False
\n",
"
C
\n",
"
Cherbourg
\n",
"
yes
\n",
"
False
\n",
"
\n",
"
\n",
"
2
\n",
"
1
\n",
"
3
\n",
"
female
\n",
"
26.0
\n",
"
0
\n",
"
0
\n",
"
7.9250
\n",
"
S
\n",
"
Third
\n",
"
woman
\n",
"
False
\n",
"
NaN
\n",
"
Southampton
\n",
"
yes
\n",
"
True
\n",
"
\n",
"
\n",
"
3
\n",
"
1
\n",
"
1
\n",
"
female
\n",
"
35.0
\n",
"
1
\n",
"
0
\n",
"
53.1000
\n",
"
S
\n",
"
First
\n",
"
woman
\n",
"
False
\n",
"
C
\n",
"
Southampton
\n",
"
yes
\n",
"
False
\n",
"
\n",
"
\n",
"
4
\n",
"
0
\n",
"
3
\n",
"
male
\n",
"
35.0
\n",
"
0
\n",
"
0
\n",
"
8.0500
\n",
"
S
\n",
"
Third
\n",
"
man
\n",
"
True
\n",
"
NaN
\n",
"
Southampton
\n",
"
no
\n",
"
True
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" survived pclass sex age sibsp parch fare embarked class \\\n",
"0 0 3 male 22.0 1 0 7.2500 S Third \n",
"1 1 1 female 38.0 1 0 71.2833 C First \n",
"2 1 3 female 26.0 0 0 7.9250 S Third \n",
"3 1 1 female 35.0 1 0 53.1000 S First \n",
"4 0 3 male 35.0 0 0 8.0500 S Third \n",
"\n",
" who adult_male deck embark_town alive alone \n",
"0 man True NaN Southampton no False \n",
"1 woman False C Cherbourg yes False \n",
"2 woman False NaN Southampton yes True \n",
"3 woman False C Southampton yes False \n",
"4 man True NaN Southampton no True "
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import seaborn as sns\n",
"from mds_2025_helper_functions import dataset_summary\n",
"\n",
"# Load the Titanic dataset\n",
"titanic = sns.load_dataset(\"titanic\")\n",
"\n",
"# Display the first few rows of the dataset\n",
"titanic.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Execute dataset_summary functioin\n",
"\n",
"To start, we pass the Titanic dataset to the `dataset_summary` function. This generates a detailed summary that includes missing values, feature types, duplicate rows, and summaries for numerical and categorical features."
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"from mds_2025_helper_functions.dataset_summary import dataset_summary\n",
"\n",
"# Generate the dataset summary\n",
"summary = dataset_summary(titanic)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Analyze the Results\n",
"\n",
"#### **Missing Values**\n",
"\n",
"The function identifies missing values in each column, providing counts and percentages."
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Missing Values:\n",
" column missing_count missing_percentage\n",
"0 survived 0 0.000000\n",
"1 pclass 0 0.000000\n",
"2 sex 0 0.000000\n",
"3 age 177 19.865320\n",
"4 sibsp 0 0.000000\n",
"5 parch 0 0.000000\n",
"6 fare 0 0.000000\n",
"7 embarked 2 0.224467\n",
"8 class 0 0.000000\n",
"9 who 0 0.000000\n",
"10 adult_male 0 0.000000\n",
"11 deck 688 77.216611\n",
"12 embark_town 2 0.224467\n",
"13 alive 0 0.000000\n",
"14 alone 0 0.000000\n"
]
}
],
"source": [
"print(\"Missing Values:\")\n",
"print(summary[\"missing_values\"])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### **Feature Types**\n",
"\n",
"The summary includes counts of numerical and categorical features in the dataset."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Feature Types:\n",
"{'numerical_features': 6, 'categorical_features': 9}\n"
]
}
],
"source": [
"print(\"Feature Types:\")\n",
"print(summary[\"feature_types\"])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### **Duplicate Rows**\n",
"\n",
"The function identifies the total number of duplicate rows in the dataset."
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Duplicate Rows:\n",
"107\n"
]
}
],
"source": [
"print(\"Duplicate Rows:\")\n",
"print(summary[\"duplicates\"])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### **Numerical Summary**\n",
"\n",
"The `numerical_summary` key contains descriptive statistics for all numerical features in the dataset."
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Numerical Summary:\n",
" count mean std min 25% 50% 75% max\n",
"survived 891.0 0.383838 0.486592 0.00 0.0000 0.0000 1.0 1.0000\n",
"pclass 891.0 2.308642 0.836071 1.00 2.0000 3.0000 3.0 3.0000\n",
"age 714.0 29.699118 14.526497 0.42 20.1250 28.0000 38.0 80.0000\n",
"sibsp 891.0 0.523008 1.102743 0.00 0.0000 0.0000 1.0 8.0000\n",
"parch 891.0 0.381594 0.806057 0.00 0.0000 0.0000 0.0 6.0000\n",
"fare 891.0 32.204208 49.693429 0.00 7.9104 14.4542 31.0 512.3292\n"
]
}
],
"source": [
"print(\"Numerical Summary:\")\n",
"print(summary[\"numerical_summary\"])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### **Categorical Summary**\n",
"\n",
"Unique value counts for all categorical features are summarized under the `categorical_summary` key."
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Categorical Summary:\n",
" column unique_values\n",
"0 sex 2\n",
"1 embarked 3\n",
"2 class 3\n",
"3 who 3\n",
"4 deck 7\n",
"5 embark_town 3\n",
"6 alive 2\n"
]
}
],
"source": [
"print(\"Categorical Summary:\")\n",
"print(summary[\"categorical_summary\"])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Comparing Multiple Datasets\n",
"\n",
"The function can be used to analyze and compare multiple datasets simultaneously. Let’s compare the Titanic dataset with the Iris dataset from `seaborn`."
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Titanic Numerical Summary:\n",
" count mean std min 25% 50% 75% max\n",
"survived 891.0 0.383838 0.486592 0.00 0.0000 0.0000 1.0 1.0000\n",
"pclass 891.0 2.308642 0.836071 1.00 2.0000 3.0000 3.0 3.0000\n",
"age 714.0 29.699118 14.526497 0.42 20.1250 28.0000 38.0 80.0000\n",
"sibsp 891.0 0.523008 1.102743 0.00 0.0000 0.0000 1.0 8.0000\n",
"parch 891.0 0.381594 0.806057 0.00 0.0000 0.0000 0.0 6.0000\n",
"fare 891.0 32.204208 49.693429 0.00 7.9104 14.4542 31.0 512.3292\n",
"\n",
"Iris Numerical Summary:\n",
" count mean std min 25% 50% 75% max\n",
"sepal_length 150.0 5.843333 0.828066 4.3 5.1 5.80 6.4 7.9\n",
"sepal_width 150.0 3.057333 0.435866 2.0 2.8 3.00 3.3 4.4\n",
"petal_length 150.0 3.758000 1.765298 1.0 1.6 4.35 5.1 6.9\n",
"petal_width 150.0 1.199333 0.762238 0.1 0.3 1.30 1.8 2.5\n"
]
}
],
"source": [
"# Load another dataset\n",
"iris = sns.load_dataset(\"iris\")\n",
"\n",
"# Generate summaries for both datasets\n",
"titanic_summary = dataset_summary(titanic)\n",
"iris_summary = dataset_summary(iris)\n",
"\n",
"# Compare numerical feature summaries\n",
"print(\"Titanic Numerical Summary:\")\n",
"print(titanic_summary[\"numerical_summary\"])\n",
"\n",
"print(\"\\nIris Numerical Summary:\")\n",
"print(iris_summary[\"numerical_summary\"])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Visualize hypothesis tests"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We'll continue to demonstrate the functionality of htv() using the toy data created in perform_eda(),"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
age
\n",
"
salary
\n",
"
department
\n",
"
joining_date
\n",
"
experience
\n",
"
performance_score
\n",
"
bonus
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
58
\n",
"
97121
\n",
"
Operations
\n",
"
2013-09-25
\n",
"
6
\n",
"
3.801431
\n",
"
False
\n",
"
\n",
"
\n",
"
1
\n",
"
48
\n",
"
99479
\n",
"
Finance
\n",
"
2015-06-05
\n",
"
16
\n",
"
4.386645
\n",
"
False
\n",
"
\n",
"
\n",
"
2
\n",
"
34
\n",
"
119475
\n",
"
Finance
\n",
"
2014-07-22
\n",
"
29
\n",
"
4.425297
\n",
"
False
\n",
"
\n",
"
\n",
"
3
\n",
"
27
\n",
"
49457
\n",
"
HR
\n",
"
2014-03-10
\n",
"
3
\n",
"
2.618033
\n",
"
False
\n",
"
\n",
"
\n",
"
4
\n",
"
40
\n",
"
96557
\n",
"
Marketing
\n",
"
2015-08-01
\n",
"
20
\n",
"
4.551080
\n",
"
True
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" age salary department joining_date experience performance_score bonus\n",
"0 58 97121 Operations 2013-09-25 6 3.801431 False\n",
"1 48 99479 Finance 2015-06-05 16 4.386645 False\n",
"2 34 119475 Finance 2014-07-22 29 4.425297 False\n",
"3 27 49457 HR 2014-03-10 3 2.618033 False\n",
"4 40 96557 Marketing 2015-08-01 20 4.551080 True"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from mds_2025_helper_functions.htv import htv\n",
"df.head()\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Based on our observations of the data, we can set up our hypothesis testing question as Are the average salaries of employees in the research department (e.g., HR) significantly higher than the average salaries of employees in all departments? The hypothesis test will be set as one-tail z-test with significant level a=0.05 \n",
"\n",
"__Hypothesis (H0): Average salary of departmental HR ≤ average salary of all departments__\n",
"\n",
"__Alternative hypothesis (H1): Average salary of departmental HR > average salary of all departments__\n"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(77464.08, 67306.75, 27874.066306200268, 8)"
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Extracting relevant data for hypothesis testing\n",
"all_salary = df['salary'] # All department salaries\n",
"hr_salary = df[df['department'] == 'HR']['salary'] # Salaries for HR department\n",
"\n",
"# Calculating parameters\n",
"mu0 = all_salary.mean() # Overall mean salary\n",
"mu1 = hr_salary.mean() # Mean salary for HR\n",
"sigma = all_salary.std() # Using overall standard deviation as an approximation\n",
"sample_size = len(hr_salary) # Sample size for HR\n",
"\n",
"# Define test parameters\n",
"test_params = {\n",
" \"mu0\": mu0,\n",
" \"mu1\": mu1,\n",
" \"sigma\": sigma,\n",
" \"sample_size\": sample_size\n",
"}\n",
"\n",
"# Visualize hypothesis testing using the htv function\n",
"fig, ax = htv(test_output=test_params, test_type=\"z\", alpha=0.05, tail=\"one-tailed\")\n",
"\n",
"(mu0, mu1, sigma, sample_size) # Display the calculated values\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We can clearly see the type1 error and type2 error from the above figure, which provides image help to those who can't understand the hypothesis test quickly to make them understand."
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python (mds-2025-helper)",
"language": "python",
"name": "mds-2025-helper"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.11"
}
},
"nbformat": 4,
"nbformat_minor": 4
}