{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":6431544,"sourceType":"datasetVersion","datasetId":3711247}],"dockerImageVersionId":30558,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"### Guideline Map: \n\n\n##### Next Notebook:\n###### [Project 2: Loan Credit Prediction](https://www.kaggle.com/code/rouzbeh/loan-credit-prediction)  \n\n##### Course Content:   \n###### [1. Dive into Data Science: Content -->](https://www.kaggle.com/rouzbeh/1-dive-into-ds-content)","metadata":{"_uuid":"7950c977-d3a2-4bba-878d-3038bef4b249","_cell_guid":"8649e7a7-7e1f-40a2-a305-663596015466","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2023-09-16T12:12:02.718861Z","iopub.execute_input":"2023-09-16T12:12:02.719658Z","iopub.status.idle":"2023-09-16T12:12:03.181232Z","shell.execute_reply.started":"2023-09-16T12:12:02.719611Z","shell.execute_reply":"2023-09-16T12:12:03.179957Z"}}},{"cell_type":"markdown","source":"# Drug Prescription Prediction Project Overview\nThe goal of this project is to provide a quick overview of the **CRISP-DM methodology** for data-driven problem-solving in a practical context. To achieve this, we leveraged **ChatGPT 3.5** to generate Python code through an interactive dialogue with **prompts**. The resulting code was then seamlessly **copied and pasted** into **Kaggle** code cells, simplifying the execution process.\n\nThis approach is particularly beneficial for those looking to navigate Data Science challenges and solve problems **without prior knowledge** of specific techniques, algorithms, or coding intricacies. The emphasis is on a **user-friendly experience**, making it accessible for individuals who may not have a deep understanding of the technical aspects of data analysis and modeling.","metadata":{"_uuid":"3bc0821c-af0e-439a-b48c-eea76c33a14e","_cell_guid":"dc0d4114-8f2c-4dfd-a19b-40dcf8ca18a8","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"markdown","source":"## Data Understanding (or EDA Process)","metadata":{"_uuid":"c4e03f1e-d94c-4063-8640-46a6be341fb6","_cell_guid":"7a3cdf0a-9436-4fc5-a05c-021f1888c646","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"markdown","source":"**Prompt:**  \nHey GPT, i have a dataset that file name is \"drug.csv\".  this file contain 7 features:  \n\n1: Age: Continues  \n2: Sex: Categorical  \n3: BP: Categorical (Blood Pressure)  \n4: Cholesterol: Cat.  \n5: Na: Cont.  \n6: K: Cont.  \nDrug: Cat. ( drug that was prescribed with doctor)  \n\nin first step, i want read this file in python and see data.\nplease tell me python coding with explain and comment.","metadata":{"_uuid":"9574ae40-b02e-48ac-8e61-3f629df97ed6","_cell_guid":"2f9d4235-abc3-439a-a8d8-c8b10d2f3607","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"# Step 1: Import the necessary libraries\nimport pandas as pd\n\n# Step 2: Define the file path\nfile_path = \"/kaggle/input/introds/Drug.csv\"\n\n# Step 3: Read the CSV file into a pandas DataFrame\n# The read_csv() function is used to read the CSV file and store it as a DataFrame.\n# You can specify various options such as delimiter, encoding, and header.\n# In this case, we assume that the first row contains column headers.\ndf = pd.read_csv(\"/kaggle/input/introds/Drug.csv\")\n# Step 4: View the first few rows of the DataFrame to get an overview of the data\n# The head() function is used to display the top rows of the DataFrame (default is 5 rows).\n# You can specify the number of rows to display within the parentheses.\n# This helps you get a quick look at the structure of your data.\nprint(df.head())\n\n# Step 5: Close the file (optional)\n# While not necessary, you can close the file explicitly when you're done with it.\n# pandas automatically closes the file when the DataFrame goes out of scope.\n# But it's good practice to close the file manually if needed.\n# df.close()  # Uncomment this line if you want to close the file explicitly\n\n# Step 6: Display basic information about the DataFrame\n# You can use the info() function to get information about the DataFrame, including data types and missing values.\nprint(df.info())","metadata":{"_uuid":"f1d1a9e7-353c-4629-8b52-1dd91ba00583","_cell_guid":"dd52476e-4bbc-4a2e-a890-8372f8c6b7d4","trusted":true,"collapsed":false,"execution":{"iopub.status.busy":"2024-12-29T17:25:23.608206Z","iopub.execute_input":"2024-12-29T17:25:23.608751Z","iopub.status.idle":"2024-12-29T17:25:23.666315Z","shell.execute_reply.started":"2024-12-29T17:25:23.608718Z","shell.execute_reply":"2024-12-29T17:25:23.665012Z"},"jupyter":{"outputs_hidden":false}},"outputs":[{"name":"stdout","text":"   Age Sex      BP Cholesterol        Na         K   Drug\n0   23   F    HIGH        HIGH  0.792535  0.031258  drugY\n1   47   M     LOW        HIGH  0.739309  0.056468  drugC\n2   47   M     LOW        HIGH  0.697269  0.068944  drugC\n3   28   F  NORMAL        HIGH  0.563682  0.072289  drugX\n4   61   F     LOW        HIGH  0.559294  0.030998  drugY\n<class 'pandas.core.frame.DataFrame'>\nRangeIndex: 200 entries, 0 to 199\nData columns (total 7 columns):\n #   Column       Non-Null Count  Dtype  \n---  ------       --------------  -----  \n 0   Age          200 non-null    int64  \n 1   Sex          200 non-null    object \n 2   BP           200 non-null    object \n 3   Cholesterol  200 non-null    object \n 4   Na           200 non-null    float64\n 5   K            200 non-null    float64\n 6   Drug         200 non-null    object \ndtypes: float64(2), int64(1), object(4)\nmemory usage: 11.1+ KB\nNone\n","output_type":"stream"}],"execution_count":4},{"cell_type":"markdown","source":"**Prompt:**  \nOK! Now, i want describe basic statistical measures for all quantitative and qualitative data features such as statistical summarize, data type, frequent table for each features and inside class.","metadata":{"_uuid":"d612f97d-03db-4707-9534-1f659b4d89de","_cell_guid":"8d067728-a3ca-4236-b286-853a6c9b5468","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"# Step 7: Descriptive statistics for quantitative features\n# Describe basic statistics for the quantitative features (Age, Na, K).\n# The describe() function provides statistics like count, mean, std (standard deviation), min, 25%, 50%, and max.\nquantitative_stats = df[['Age', 'Na', 'K']].describe()\n\n# Step 8: Frequency table for qualitative features\n# For qualitative features (Sex, BP, Cholesterol, Drug), we'll create frequency tables.\n# The value_counts() function counts the occurrences of each unique value in a column.\n# Let's create frequency tables for each qualitative feature.\n\n# Frequency table for Sex\nsex_counts = df['Sex'].value_counts()\n\n# Frequency table for BP (Blood Pressure)\nbp_counts = df['BP'].value_counts()\n\n# Frequency table for Cholesterol\ncholesterol_counts = df['Cholesterol'].value_counts()\n\n# Frequency table for Drug (target variable)\ndrug_counts = df['Drug'].value_counts()\n\n# Display the statistical summary for quantitative features\nprint(\"Statistical Summary for Quantitative Features:\")\nprint(quantitative_stats)\n\n# Display the frequency tables for qualitative features\nprint(\"\\nFrequency Table for Sex:\")\nprint(sex_counts)\n\nprint(\"\\nFrequency Table for Blood Pressure:\")\nprint(bp_counts)\n\nprint(\"\\nFrequency Table for Cholesterol:\")\nprint(cholesterol_counts)\n\nprint(\"\\nFrequency Table for Drug (Target Variable):\")\nprint(drug_counts)","metadata":{"_uuid":"81c291a9-b803-4e52-abf8-2d85bf3667b7","_cell_guid":"4afd37fb-c460-431f-9d0f-058254820a3f","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:26:17.329038Z","iopub.execute_input":"2024-12-29T17:26:17.329451Z","iopub.status.idle":"2024-12-29T17:26:17.363072Z","shell.execute_reply.started":"2024-12-29T17:26:17.329416Z","shell.execute_reply":"2024-12-29T17:26:17.361593Z"}},"outputs":[{"name":"stdout","text":"Statistical Summary for Quantitative Features:\n              Age          Na           K\ncount  200.000000  200.000000  200.000000\nmean    44.315000    0.697095    0.050174\nstd     16.544315    0.118907    0.017611\nmin     15.000000    0.500169    0.020022\n25%     31.000000    0.583887    0.035054\n50%     45.000000    0.721853    0.049663\n75%     58.000000    0.801494    0.066000\nmax     74.000000    0.896056    0.079788\n\nFrequency Table for Sex:\nSex\nM    104\nF     96\nName: count, dtype: int64\n\nFrequency Table for Blood Pressure:\nBP\nHIGH      77\nLOW       64\nNORMAL    59\nName: count, dtype: int64\n\nFrequency Table for Cholesterol:\nCholesterol\nHIGH      103\nNORMAL     97\nName: count, dtype: int64\n\nFrequency Table for Drug (Target Variable):\nDrug\ndrugY    91\ndrugX    54\ndrugA    23\ndrugC    16\ndrugB    16\nName: count, dtype: int64\n","output_type":"stream"}],"execution_count":6},{"cell_type":"markdown","source":"**Prompt:**  \nthanks GPT! can you tell me what can i show percent and count together for each frequent table above?","metadata":{"_uuid":"16adcd62-1a62-4fdc-96b3-d9c86a2c0a5a","_cell_guid":"7bf6d5a9-8e2f-488b-bedf-f7d4ffe1fd80","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"# Calculate the percentage for each category in the frequency tables\n# You can use the division operator to calculate the percentage.\n# (count of category / total count) * 100\n\n# Calculate percentage for Sex\nsex_percentage = (sex_counts / len(df)) * 100\n\n# Calculate percentage for BP (Blood Pressure)\nbp_percentage = (bp_counts / len(df)) * 100\n\n# Calculate percentage for Cholesterol\ncholesterol_percentage = (cholesterol_counts / len(df)) * 100\n\n# Calculate percentage for Drug (Target Variable)\ndrug_percentage = (drug_counts / len(df)) * 100\n\n# Create new DataFrames to display both count and percentage\n# Using pd.concat() to concatenate count and percentage along the columns (axis=1).\n# Using axis=1 ensures they are displayed side by side.\nsex_summary = pd.concat([sex_counts, sex_percentage], axis=1)\nsex_summary.columns = ['Count', 'Percentage']\n\nbp_summary = pd.concat([bp_counts, bp_percentage], axis=1)\nbp_summary.columns = ['Count', 'Percentage']\n\ncholesterol_summary = pd.concat([cholesterol_counts, cholesterol_percentage], axis=1)\ncholesterol_summary.columns = ['Count', 'Percentage']\n\ndrug_summary = pd.concat([drug_counts, drug_percentage], axis=1)\ndrug_summary.columns = ['Count', 'Percentage']\n\n# Display the updated frequency tables with count and percentage\nprint(\"\\nFrequency Table for Sex:\")\nprint(sex_summary)\n\nprint(\"\\nFrequency Table for Blood Pressure:\")\nprint(bp_summary)\n\nprint(\"\\nFrequency Table for Cholesterol:\")\nprint(cholesterol_summary)\n\nprint(\"\\nFrequency Table for Drug (Target Variable):\")\nprint(drug_summary)","metadata":{"_uuid":"5d41b0ba-a950-449d-9304-04ae09f5a7f4","_cell_guid":"48eb4376-854f-448e-ab02-285589f9ec39","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:26:25.124368Z","iopub.execute_input":"2024-12-29T17:26:25.124769Z","iopub.status.idle":"2024-12-29T17:26:25.143959Z","shell.execute_reply.started":"2024-12-29T17:26:25.124737Z","shell.execute_reply":"2024-12-29T17:26:25.142734Z"}},"outputs":[{"name":"stdout","text":"\nFrequency Table for Sex:\n     Count  Percentage\nSex                   \nM      104        52.0\nF       96        48.0\n\nFrequency Table for Blood Pressure:\n        Count  Percentage\nBP                       \nHIGH       77        38.5\nLOW        64        32.0\nNORMAL     59        29.5\n\nFrequency Table for Cholesterol:\n             Count  Percentage\nCholesterol                   \nHIGH           103        51.5\nNORMAL          97        48.5\n\nFrequency Table for Drug (Target Variable):\n       Count  Percentage\nDrug                    \ndrugY     91        45.5\ndrugX     54        27.0\ndrugA     23        11.5\ndrugC     16         8.0\ndrugB     16         8.0\n","output_type":"stream"}],"execution_count":7},{"cell_type":"markdown","source":"**Prompt:**  \nNext, i want checking quality of quantitative data for outlier detection.","metadata":{"_uuid":"3377052f-c3ef-4d35-9ad2-4095a72031c2","_cell_guid":"c8aae200-5b8a-491f-bb0c-d9ab0b1e2c25","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\n# Box plot for Age\nplt.figure(figsize=(8, 4))\nplt.boxplot(df['Age'], vert=False)\nplt.title('Box Plot for Age')\nplt.show()\n\n# Histogram for Na\nplt.figure(figsize=(8, 4))\nplt.hist(df['Na'], bins=20, edgecolor='k')\nplt.title('Histogram for Na')\nplt.xlabel('Na')\nplt.ylabel('Frequency')\nplt.show()\n\n# Histogram for K\nplt.figure(figsize=(8, 4))\nplt.hist(df['K'], bins=20, edgecolor='k')\nplt.title('Histogram for K')\nplt.xlabel('K')\nplt.ylabel('Frequency')\nplt.show()","metadata":{"_uuid":"e66d3e17-159b-45c9-8958-faa571d14815","_cell_guid":"3253bc98-d86d-4c99-a1c5-68b4888f942d","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:26:26.783011Z","iopub.execute_input":"2024-12-29T17:26:26.783368Z","iopub.status.idle":"2024-12-29T17:26:27.594099Z","shell.execute_reply.started":"2024-12-29T17:26:26.783343Z","shell.execute_reply":"2024-12-29T17:26:27.592915Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x400 with 1 Axes>","image/png":"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"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 800x400 with 1 Axes>","image/png":"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"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 800x400 with 1 Axes>","image/png":"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"},"metadata":{}}],"execution_count":8},{"cell_type":"markdown","source":"**Prompt:**  \nFor next, i want graph between each features vs last feature (Drug) to understand that if correlation between them or not.","metadata":{"_uuid":"8663fc20-3918-4d25-9cb4-c786ff68a8a0","_cell_guid":"1997053f-0586-459c-8c72-a52da82c9412","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"markdown","source":"1. Bar Plot for Categorical Features:","metadata":{"_uuid":"3bce5012-5076-4a83-a47c-ad2009aeef4f","_cell_guid":"31bb39e4-7d50-4347-944c-f96761e8d4ca","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"import seaborn as sns\nimport matplotlib.pyplot as plt\n\n# Bar plot for Sex vs. Drug\nsns.countplot(x='Sex', hue='Drug', data=df)\nplt.title('Distribution of Drug by Sex')\nplt.xlabel('Sex')\nplt.ylabel('Count')\nplt.show()\n\n# Bar plot for BP vs. Drug\nsns.countplot(x='BP', hue='Drug', data=df)\nplt.title('Distribution of Drug by Blood Pressure')\nplt.xlabel('Blood Pressure')\nplt.ylabel('Count')\nplt.show()\n\n# Bar plot for Cholesterol vs. Drug\nsns.countplot(x='Cholesterol', hue='Drug', data=df)\nplt.title('Distribution of Drug by Cholesterol')\nplt.xlabel('Cholesterol')\nplt.ylabel('Count')\nplt.show()","metadata":{"_uuid":"c5d6d86f-b10b-4d56-8c82-9519c742eb7a","_cell_guid":"9f880eb7-3266-4911-9078-2f4ac12c0702","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:26:27.596282Z","iopub.execute_input":"2024-12-29T17:26:27.596634Z","iopub.status.idle":"2024-12-29T17:26:29.379633Z","shell.execute_reply.started":"2024-12-29T17:26:27.596606Z","shell.execute_reply":"2024-12-29T17:26:29.377996Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}}],"execution_count":9},{"cell_type":"markdown","source":"2. Histograms for Numeric Features:","metadata":{"_uuid":"78c6ed68-d6dc-490b-85ba-5174f2069b2d","_cell_guid":"68fc0eb3-2dcc-4604-aa9e-27ea3123e4c7","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"# Histogram for Age vs. Drug\ng = sns.FacetGrid(df, col='Drug', height=4)\ng.map(plt.hist, 'Age', bins=15)\ng.set_axis_labels('Age', 'Count')\ng.set_titles('Drug = {col_name}')\nplt.show()\n\n# Histogram for Na vs. Drug\ng = sns.FacetGrid(df, col='Drug', height=4)\ng.map(plt.hist, 'Na', bins=15)\ng.set_axis_labels('Na', 'Count')\ng.set_titles('Drug = {col_name}')\nplt.show()\n\n# Histogram for K vs. Drug\ng = sns.FacetGrid(df, col='Drug', height=4)\ng.map(plt.hist, 'K', bins=15)\ng.set_axis_labels('K', 'Count')\ng.set_titles('Drug = {col_name}')\nplt.show()","metadata":{"_uuid":"9c05578e-4f32-4cbd-9b75-d89b8d015a5a","_cell_guid":"37673cf1-c548-4217-849b-9cf7b647fdda","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:26:46.843283Z","iopub.execute_input":"2024-12-29T17:26:46.843759Z","iopub.status.idle":"2024-12-29T17:26:50.930402Z","shell.execute_reply.started":"2024-12-29T17:26:46.843723Z","shell.execute_reply":"2024-12-29T17:26:50.928926Z"}},"outputs":[{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/seaborn/axisgrid.py:118: UserWarning: The figure layout has changed to tight\n  self._figure.tight_layout(*args, **kwargs)\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 2000x400 with 5 Axes>","image/png":"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"},"metadata":{}},{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/seaborn/axisgrid.py:118: UserWarning: The figure layout has changed to tight\n  self._figure.tight_layout(*args, **kwargs)\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 2000x400 with 5 Axes>","image/png":"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"},"metadata":{}},{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/seaborn/axisgrid.py:118: UserWarning: The figure layout has changed to tight\n  self._figure.tight_layout(*args, **kwargs)\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 2000x400 with 5 Axes>","image/png":"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"},"metadata":{}}],"execution_count":10},{"cell_type":"markdown","source":"3. Correlation Matrix:","metadata":{"_uuid":"684d0117-af98-475c-b57c-766bcf0b3136","_cell_guid":"f9b384f8-6d1e-4dea-8f3a-2faf932bc666","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"# Calculate the correlation matrix\ncorrelation_matrix = df[['Age', 'Na', 'K']].corr()\n\n# Create a heatmap of the correlation matrix\nsns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', fmt=\".2f\")\nplt.title('Correlation Matrix')\nplt.show()","metadata":{"_uuid":"a7345369-d04f-465b-8786-e152caea5fa8","_cell_guid":"339bfddb-3b52-4174-af93-697f2d5d564d","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:32:46.143625Z","iopub.execute_input":"2024-12-29T17:32:46.144074Z","iopub.status.idle":"2024-12-29T17:32:46.434382Z","shell.execute_reply.started":"2024-12-29T17:32:46.144044Z","shell.execute_reply":"2024-12-29T17:32:46.433101Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 2 Axes>","image/png":"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"},"metadata":{}}],"execution_count":11},{"cell_type":"markdown","source":"**Prompt:**  \nI want statistical test for that correlation between all combination between numerical features Na vs K vs Age.","metadata":{"_uuid":"d9e036bd-0057-4a7f-802c-2652a540723d","_cell_guid":"114f317e-3616-48d8-8109-1b46e8092e42","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"import scipy.stats as stats\n\n# List of numerical features\nnumerical_features = ['Na', 'K', 'Age']\n\n# Perform correlation tests for all combinations of numerical features\nfor i in range(len(numerical_features)):\n    for j in range(i + 1, len(numerical_features)):\n        feature1 = numerical_features[i]\n        feature2 = numerical_features[j]\n        \n        # Calculate Pearson's correlation coefficient and p-value\n        pearson_corr, pearson_p_value = stats.pearsonr(df[feature1], df[feature2])\n        \n        # Calculate Spearman's rank correlation coefficient and p-value\n        spearman_corr, spearman_p_value = stats.spearmanr(df[feature1], df[feature2])\n        \n        print(f\"Correlation between {feature1} and {feature2}:\")\n        print(f\"Pearson's correlation coefficient: {pearson_corr:.4f} (p-value: {pearson_p_value:.4f})\")\n        print(f\"Spearman's rank correlation coefficient: {spearman_corr:.4f} (p-value: {spearman_p_value:.4f})\")\n        print(\"\\n\")","metadata":{"_uuid":"fcfd11a6-38b7-418e-94cc-c55dce8cc299","_cell_guid":"127d4d32-0eb2-46d5-b2b6-8bd3770bce1a","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:37:36.978820Z","iopub.execute_input":"2024-12-29T17:37:36.979232Z","iopub.status.idle":"2024-12-29T17:37:37.002322Z","shell.execute_reply.started":"2024-12-29T17:37:36.979199Z","shell.execute_reply":"2024-12-29T17:37:37.000371Z"}},"outputs":[{"name":"stdout","text":"Correlation between Na and K:\nPearson's correlation coefficient: 0.0173 (p-value: 0.8075)\nSpearman's rank correlation coefficient: -0.0052 (p-value: 0.9417)\n\n\nCorrelation between Na and Age:\nPearson's correlation coefficient: 0.1004 (p-value: 0.1573)\nSpearman's rank correlation coefficient: 0.0931 (p-value: 0.1900)\n\n\nCorrelation between K and Age:\nPearson's correlation coefficient: 0.1115 (p-value: 0.1160)\nSpearman's rank correlation coefficient: 0.1094 (p-value: 0.1230)\n\n\n","output_type":"stream"}],"execution_count":12},{"cell_type":"markdown","source":"**Prompt:**  \nCan i have correlation plot between Na vs K","metadata":{"_uuid":"73431de0-70cd-4732-b79f-5bcb94ad67f2","_cell_guid":"c532cf46-650b-4676-a40c-c9e2856e9de4","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"import seaborn as sns\nimport matplotlib.pyplot as plt\n\n# Create a scatterplot between \"Na\" and \"K\"\nsns.set(style=\"whitegrid\")\nplt.figure(figsize=(8, 6))\n\n# Plot the correlation between \"Na\" and \"K\"\nsns.scatterplot(x=\"Na\", y=\"K\", data=df)\n\nplt.title(\"Correlation between Na and K\")\nplt.xlabel(\"Na\")\nplt.ylabel(\"K\")\nplt.show()","metadata":{"_uuid":"8a03e970-ab52-4e62-9d8c-34ae028b711c","_cell_guid":"93419e3a-ea18-4ceb-be2e-9175196167b7","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:40:59.963302Z","iopub.execute_input":"2024-12-29T17:40:59.963677Z","iopub.status.idle":"2024-12-29T17:41:00.390940Z","shell.execute_reply.started":"2024-12-29T17:40:59.963635Z","shell.execute_reply":"2024-12-29T17:41:00.389425Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x600 with 1 Axes>","image/png":"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"},"metadata":{}}],"execution_count":13},{"cell_type":"markdown","source":"**Prompt:**  \nOK, now i want that plot with Drug layer in color code.","metadata":{"_uuid":"751bcebd-cd72-449a-a297-7f49e86df47a","_cell_guid":"04edb583-bc09-433a-b438-cef9be6e1340","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"import seaborn as sns\nimport matplotlib.pyplot as plt\n\n# Create a scatterplot of \"Na\" vs. \"K\" with color-coded points based on \"Drug\"\nsns.set(style=\"whitegrid\")\nplt.figure(figsize=(8, 6))\n\n# Plot \"Na\" vs. \"K\" with color-coded points for different \"Drug\" categories\nsns.scatterplot(x=\"Na\", y=\"K\", hue=\"Drug\", data=df, palette=\"viridis\")\n\nplt.title(\"Scatterplot of Na vs. K (Color-Coded by Drug)\")\nplt.xlabel(\"Na\")\nplt.ylabel(\"K\")\nplt.legend(title=\"Drug\")\nplt.show()","metadata":{"_uuid":"bc021f59-016f-4f52-a692-be289261a183","_cell_guid":"23cf3e9d-dd99-47ff-a29f-eac221c7ebbb","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:42:19.303811Z","iopub.execute_input":"2024-12-29T17:42:19.304233Z","iopub.status.idle":"2024-12-29T17:42:19.886153Z","shell.execute_reply.started":"2024-12-29T17:42:19.304201Z","shell.execute_reply":"2024-12-29T17:42:19.884731Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x600 with 1 Axes>","image/png":"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"},"metadata":{}}],"execution_count":14},{"cell_type":"markdown","source":"**Prompt:**  \nI want statistical test between Drug and Sex that has 2 classes, and checking its assumptions.","metadata":{"_uuid":"94e4a0f6-bff6-4957-a4f6-407117663fdd","_cell_guid":"5dc0789b-1fa4-4507-bc81-8ee873c2c4fa","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"import pandas as pd\nimport scipy.stats as stats\n\n# Create a contingency table between \"Drug\" and \"Sex\"\ncontingency_table = pd.crosstab(df['Drug'], df['Sex'])\n\n# Perform the chi-squared test of independence\nchi2_stat, p_value, dof, expected = stats.chi2_contingency(contingency_table)\n\n# Display the results\nprint(\"Contingency Table:\")\nprint(contingency_table)\nprint(\"\\nChi-Squared Statistic:\", chi2_stat)\nprint(\"Degrees of Freedom:\", dof)\nprint(\"p-value:\", p_value)\nprint(\"\\nExpected Frequencies:\")\nprint(expected)","metadata":{"_uuid":"534f3ea0-dd62-49f0-ad4a-98516f2f3a86","_cell_guid":"c5263329-87e0-497e-8ca5-6d0b7d7f6f8d","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:46:35.093772Z","iopub.execute_input":"2024-12-29T17:46:35.094191Z","iopub.status.idle":"2024-12-29T17:46:35.123167Z","shell.execute_reply.started":"2024-12-29T17:46:35.094159Z","shell.execute_reply":"2024-12-29T17:46:35.121613Z"}},"outputs":[{"name":"stdout","text":"Contingency Table:\nSex     F   M\nDrug         \ndrugA   9  14\ndrugB   6  10\ndrugC   7   9\ndrugX  27  27\ndrugY  47  44\n\nChi-Squared Statistic: 2.119248418109203\nDegrees of Freedom: 4\np-value: 0.7138369773987128\n\nExpected Frequencies:\n[[11.04 11.96]\n [ 7.68  8.32]\n [ 7.68  8.32]\n [25.92 28.08]\n [43.68 47.32]]\n","output_type":"stream"}],"execution_count":15},{"cell_type":"markdown","source":"## Data Preparation","metadata":{"_uuid":"26ba815e-ba2e-4d22-844f-e2a185303ae4","_cell_guid":"25f542e7-1edc-42f0-b20f-bf8dcb7b5cb5","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"markdown","source":"**Prompt:**  \nI want create Na/K as a ratio index, and add to my data.","metadata":{"_uuid":"d58174e6-8b67-4ac9-bf64-a465a7f99d2d","_cell_guid":"9e1697a1-51fc-45ea-9a75-dcac4032fe5d","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"# Create a new column \"Na/K\" by dividing \"Na\" by \"K\"\ndf['Na/K'] = df['Na'] / df['K']\n\n# Display the updated DataFrame with the new column\nprint(df.head())","metadata":{"_uuid":"32f0aca9-00d8-4740-8efa-e50dcfb530e4","_cell_guid":"32ed4de4-3f5c-4ade-999d-75fc4bdeab03","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:57:52.919513Z","iopub.execute_input":"2024-12-29T17:57:52.920853Z","iopub.status.idle":"2024-12-29T17:57:52.932209Z","shell.execute_reply.started":"2024-12-29T17:57:52.920798Z","shell.execute_reply":"2024-12-29T17:57:52.930751Z"}},"outputs":[{"name":"stdout","text":"   Age Sex      BP Cholesterol        Na         K   Drug       Na/K\n0   23   F    HIGH        HIGH  0.792535  0.031258  drugY  25.354629\n1   47   M     LOW        HIGH  0.739309  0.056468  drugC  13.092530\n2   47   M     LOW        HIGH  0.697269  0.068944  drugC  10.113556\n3   28   F  NORMAL        HIGH  0.563682  0.072289  drugX   7.797618\n4   61   F     LOW        HIGH  0.559294  0.030998  drugY  18.042906\n","output_type":"stream"}],"execution_count":16},{"cell_type":"markdown","source":"**Prompt:**  \nMy categorical features and their classes are:  \n\nSex: {F, M}  \nBP: {LOW, NORMAL, HIGH}  \nCholesterol: {NORMAL, HIGH}  \nDrug: {drugA, drugB, drugC, drugX, dugY}.  \n\nNow, please use ordinal encoding for Sex, BP, Cholesterol and lable encoding for Drug.","metadata":{"_uuid":"39597440-265c-4ce2-8ca6-aca54f13d71c","_cell_guid":"506514d3-7682-452a-bd6f-7c94849a1bef","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"from sklearn.preprocessing import OrdinalEncoder, LabelEncoder\n\n# Ordinal encoding for Sex\nsex_mapping = {'F': 0, 'M': 1}\ndf['Sex_encoded'] = df['Sex'].map(sex_mapping)\n\n# Ordinal encoding for BP (Blood Pressure)\nbp_mapping = {'LOW': 0, 'NORMAL': 1, 'HIGH': 2}\ndf['BP_encoded'] = df['BP'].map(bp_mapping)\n\n# Ordinal encoding for Cholesterol\ncholesterol_mapping = {'NORMAL': 0, 'HIGH': 1}\ndf['Cholesterol_encoded'] = df['Cholesterol'].map(cholesterol_mapping)\n\n# Label encoding for Drug\nlabel_encoder = LabelEncoder()\ndf['Drug_encoded'] = label_encoder.fit_transform(df['Drug'])\n\n# Display the updated DataFrame with encoded columns\nprint(df.head())","metadata":{"_uuid":"acc0d1ee-1c79-4b08-916a-da9b81ca5051","_cell_guid":"62d23a15-001d-486c-be71-2c0142f6b487","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T17:59:18.564398Z","iopub.execute_input":"2024-12-29T17:59:18.564811Z","iopub.status.idle":"2024-12-29T17:59:18.662874Z","shell.execute_reply.started":"2024-12-29T17:59:18.564780Z","shell.execute_reply":"2024-12-29T17:59:18.661713Z"}},"outputs":[{"name":"stdout","text":"   Age Sex      BP Cholesterol        Na         K   Drug       Na/K  \\\n0   23   F    HIGH        HIGH  0.792535  0.031258  drugY  25.354629   \n1   47   M     LOW        HIGH  0.739309  0.056468  drugC  13.092530   \n2   47   M     LOW        HIGH  0.697269  0.068944  drugC  10.113556   \n3   28   F  NORMAL        HIGH  0.563682  0.072289  drugX   7.797618   \n4   61   F     LOW        HIGH  0.559294  0.030998  drugY  18.042906   \n\n   Sex_encoded  BP_encoded  Cholesterol_encoded  Drug_encoded  \n0            0           2                    1             4  \n1            1           0                    1             2  \n2            1           0                    1             2  \n3            0           1                    1             3  \n4            0           0                    1             4  \n","output_type":"stream"}],"execution_count":17},{"cell_type":"markdown","source":"**Prompt:**  \nPlease split my dataset with all features for training and testing in 50-50 for Modeling and evaluation.","metadata":{"_uuid":"d090d03d-d252-4be3-89a9-cc8952abdff5","_cell_guid":"3b387170-da1c-473c-88a1-e45282c10cab","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"from sklearn.model_selection import train_test_split\n\n# Define your feature columns (X) and target column (y)\nX = df[['Age', 'Sex_encoded', 'BP_encoded', 'Cholesterol_encoded', 'Na', 'K', 'Na/K']]\ny = df['Drug_encoded']  # Assuming 'Target' is the column name for the target variable\n\n# Split the dataset into training and testing sets with a 50-50 ratio\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5, random_state=42)\n\n# Print the shapes of the training and testing sets\nprint(\"Training set shape (X, y):\", X_train.shape, y_train.shape)\nprint(\"Testing set shape (X, y):\", X_test.shape, y_test.shape)","metadata":{"_uuid":"e67bb760-733f-4d33-95f9-ca502f96a2a6","_cell_guid":"f2651899-48ff-4f9c-87c1-a8af55942405","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T18:00:57.684856Z","iopub.execute_input":"2024-12-29T18:00:57.685260Z","iopub.status.idle":"2024-12-29T18:00:57.833494Z","shell.execute_reply.started":"2024-12-29T18:00:57.685229Z","shell.execute_reply":"2024-12-29T18:00:57.832231Z"}},"outputs":[{"name":"stdout","text":"Training set shape (X, y): (100, 7) (100,)\nTesting set shape (X, y): (100, 7) (100,)\n","output_type":"stream"}],"execution_count":18},{"cell_type":"markdown","source":"**Prompt:**  \nI want doing normalization on Age and Na/K features in dataset based on training data and add to data set.","metadata":{"_uuid":"89dc442e-ccde-4838-bc9c-bf38e58811b4","_cell_guid":"3b76b87f-49b1-4a1e-be3d-04db8e768ade","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"from sklearn.preprocessing import StandardScaler\n\n# Initialize the StandardScaler\nscaler = StandardScaler()\n\n# Fit the scaler on the training data and transform both training and testing data\nX_train[['Age', 'Na/K']] = scaler.fit_transform(X_train[['Age', 'Na/K']])\nX_test[['Age', 'Na/K']] = scaler.transform(X_test[['Age', 'Na/K']])\n\n# Print the first few rows of the updated training data\nprint(\"Updated Training Data:\")\nprint(X_train.head())","metadata":{"_uuid":"13278323-8f4e-46ff-bbbe-1d4b5de87fda","_cell_guid":"18771820-e6c0-4e80-ba02-1e27f493bfb1","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T18:11:39.484786Z","iopub.execute_input":"2024-12-29T18:11:39.485906Z","iopub.status.idle":"2024-12-29T18:11:39.503079Z","shell.execute_reply.started":"2024-12-29T18:11:39.485868Z","shell.execute_reply":"2024-12-29T18:11:39.501871Z"}},"outputs":[{"name":"stdout","text":"Updated Training Data:\n          Age  Sex_encoded  BP_encoded  Cholesterol_encoded        Na  \\\n4    0.921032            0           0                    1  0.559294   \n32   0.213453            1           0                    0  0.625889   \n142  0.862067            1           2                    0  0.635762   \n145  0.921032            1           1                    1  0.745123   \n109 -1.319635            1           1                    1  0.543355   \n\n            K      Na/K  \n4    0.030998  0.299149  \n32   0.056828 -0.762623  \n142  0.073744 -1.124022  \n145  0.078906 -0.999862  \n109  0.032247  0.118925  \n","output_type":"stream"}],"execution_count":19},{"cell_type":"markdown","source":"## Modeling and Evaluation","metadata":{"_uuid":"cf39005b-b2ac-4c1d-9e4b-a8070c782661","_cell_guid":"c9aef0f5-f54d-4d45-b8a7-faee277aac73","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"markdown","source":"**Prompt:**  \nPlease train my training dataset with decision tree model and show my tree model, after removing  Sex_encoded, Na and K.","metadata":{"_uuid":"f739d3d6-bfef-4fe9-bcae-f911f23f1c18","_cell_guid":"779f36e6-293e-48ad-95ef-fd4c317ec2bb","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"from sklearn.tree import DecisionTreeClassifier\nfrom sklearn.tree import export_text\nimport graphviz\nfrom sklearn import tree\n\n# Define the features (excluding 'Sex_encoded', 'Na', and 'K')\nfeatures_to_use = ['Age', 'BP_encoded', 'Cholesterol_encoded', 'Na/K']\n\n# Select the relevant columns from the training dataset\nX_train_subset = X_train[features_to_use]\n\n# Initialize and train the Decision Tree classifier\nclf = DecisionTreeClassifier(random_state=42)\nclf.fit(X_train_subset, y_train)\n\n# Display the Decision Tree model as text\ntree_rules = export_text(clf, feature_names=features_to_use)\nprint(\"Decision Tree Model Rules:\")\nprint(tree_rules)\n\n# Visualize the Decision Tree model\ndot_data = tree.export_graphviz(clf, out_file=None, \n                                feature_names=features_to_use, \n                                class_names=sorted(df['Drug'].unique()), \n                                filled=True, rounded=True, special_characters=True) \ngraph = graphviz.Source(dot_data)\ngraph.render(\"decision_tree\")  # Save the tree visualization as a PDF file\ngraph.view(\"decision_tree\")    # Open the PDF file (you may need Graphviz installed)","metadata":{"_uuid":"f12cc709-5198-4541-8e6d-835e9c0660b4","_cell_guid":"e4761eb9-36cf-4398-a3a7-ae447ca7a736","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false},"execution":{"iopub.status.busy":"2024-12-29T18:23:19.573111Z","iopub.execute_input":"2024-12-29T18:23:19.573525Z","iopub.status.idle":"2024-12-29T18:23:20.159549Z","shell.execute_reply.started":"2024-12-29T18:23:19.573488Z","shell.execute_reply":"2024-12-29T18:23:20.158356Z"}},"outputs":[{"name":"stdout","text":"Decision Tree Model Rules:\n|--- Na/K <= -0.19\n|   |--- BP_encoded <= 1.50\n|   |   |--- BP_encoded <= 0.50\n|   |   |   |--- Cholesterol_encoded <= 0.50\n|   |   |   |   |--- class: 3\n|   |   |   |--- Cholesterol_encoded >  0.50\n|   |   |   |   |--- class: 2\n|   |   |--- BP_encoded >  0.50\n|   |   |   |--- class: 3\n|   |--- BP_encoded >  1.50\n|   |   |--- Age <= 0.30\n|   |   |   |--- class: 0\n|   |   |--- Age >  0.30\n|   |   |   |--- class: 1\n|--- Na/K >  -0.19\n|   |--- class: 4\n\n","output_type":"stream"},{"execution_count":20,"output_type":"execute_result","data":{"text/plain":"'decision_tree.pdf'"},"metadata":{}},{"name":"stderr","text":"Error: no \"view\" mailcap rules found for type \"application/pdf\"\n/usr/bin/xdg-open: 869: www-browser: not found\n/usr/bin/xdg-open: 869: links2: not found\n/usr/bin/xdg-open: 869: elinks: not found\n/usr/bin/xdg-open: 869: links: not found\n/usr/bin/xdg-open: 869: lynx: not found\n/usr/bin/xdg-open: 869: w3m: not found\nxdg-open: no method available for opening 'decision_tree.pdf'\n","output_type":"stream"}],"execution_count":20},{"cell_type":"markdown","source":"**Prompt:**  \nFor next, i want evaluate my model for training and testing dataset","metadata":{"_uuid":"8b8a5903-4823-4ec4-9049-f755cebfd05c","_cell_guid":"7025e638-5e12-4ec6-9ec2-ac98d415ac81","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"from sklearn.metrics import accuracy_score, classification_report, confusion_matrix\n\n# Make predictions on the training and testing datasets\ny_train_pred = clf.predict(X_train_subset)\ny_test_pred = clf.predict(X_test[features_to_use])\n\n# Evaluate the model on the training dataset\naccuracy_train = accuracy_score(y_train, y_train_pred)\nconfusion_matrix_train = confusion_matrix(y_train, y_train_pred)\nclassification_report_train = classification_report(y_train, y_train_pred, target_names=df['Drug'].unique())\n\n# Evaluate the model on the testing dataset\naccuracy_test = accuracy_score(y_test, y_test_pred)\nconfusion_matrix_test = confusion_matrix(y_test, y_test_pred)\nclassification_report_test = classification_report(y_test, y_test_pred, target_names=df['Drug'].unique())\n\n# Print evaluation metrics for training dataset\nprint(\"Evaluation Metrics for Training Dataset:\")\nprint(f\"Accuracy: {accuracy_train:.2f}\")\nprint(\"Confusion Matrix:\")\nprint(confusion_matrix_train)\nprint(\"Classification Report:\")\nprint(classification_report_train)\n\n# Print evaluation metrics for testing dataset\nprint(\"\\nEvaluation Metrics for Testing Dataset:\")\nprint(f\"Accuracy: {accuracy_test:.2f}\")\nprint(\"Confusion Matrix:\")\nprint(confusion_matrix_test)\nprint(\"Classification Report:\")\nprint(classification_report_test)","metadata":{"_uuid":"e1873973-f38c-4e58-bf6c-b6035774ee66","_cell_guid":"76975d48-f97d-4f35-a9cb-fc77a7d31e3b","trusted":true,"collapsed":false,"execution":{"iopub.status.busy":"2024-12-29T18:28:24.373563Z","iopub.execute_input":"2024-12-29T18:28:24.374758Z","iopub.status.idle":"2024-12-29T18:28:24.406677Z","shell.execute_reply.started":"2024-12-29T18:28:24.374711Z","shell.execute_reply":"2024-12-29T18:28:24.405388Z"},"jupyter":{"outputs_hidden":false}},"outputs":[{"name":"stdout","text":"Evaluation Metrics for Training Dataset:\nAccuracy: 1.00\nConfusion Matrix:\n[[ 7  0  0  0  0]\n [ 0  8  0  0  0]\n [ 0  0  9  0  0]\n [ 0  0  0 27  0]\n [ 0  0  0  0 49]]\nClassification Report:\n              precision    recall  f1-score   support\n\n       drugY       1.00      1.00      1.00         7\n       drugC       1.00      1.00      1.00         8\n       drugX       1.00      1.00      1.00         9\n       drugA       1.00      1.00      1.00        27\n       drugB       1.00      1.00      1.00        49\n\n    accuracy                           1.00       100\n   macro avg       1.00      1.00      1.00       100\nweighted avg       1.00      1.00      1.00       100\n\n\nEvaluation Metrics for Testing Dataset:\nAccuracy: 1.00\nConfusion Matrix:\n[[16  0  0  0  0]\n [ 0  8  0  0  0]\n [ 0  0  7  0  0]\n [ 0  0  0 27  0]\n [ 0  0  0  0 42]]\nClassification Report:\n              precision    recall  f1-score   support\n\n       drugY       1.00      1.00      1.00        16\n       drugC       1.00      1.00      1.00         8\n       drugX       1.00      1.00      1.00         7\n       drugA       1.00      1.00      1.00        27\n       drugB       1.00      1.00      1.00        42\n\n    accuracy                           1.00       100\n   macro avg       1.00      1.00      1.00       100\nweighted avg       1.00      1.00      1.00       100\n\n","output_type":"stream"}],"execution_count":21},{"cell_type":"markdown","source":"**Prompt:**  \nNow, please replace decision tree with logistic regression, and evaluate that for comparison with last model.","metadata":{"_uuid":"6b1d20ad-eb4c-41df-8663-a76c396b85b9","_cell_guid":"ee2b1260-d552-48c3-bbd2-9aa686f9963d","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"from sklearn.linear_model import LogisticRegression\n\n# Initialize and train the Logistic Regression classifier\nlogistic_reg = LogisticRegression(random_state=42, max_iter=1000)\nlogistic_reg.fit(X_train_subset, y_train)\n\n# Make predictions on the training and testing datasets\ny_train_pred_lr = logistic_reg.predict(X_train_subset)\ny_test_pred_lr = logistic_reg.predict(X_test[features_to_use])\n\n# Evaluate the Logistic Regression model on the training dataset\naccuracy_train_lr = accuracy_score(y_train, y_train_pred_lr)\nconfusion_matrix_train_lr = confusion_matrix(y_train, y_train_pred_lr)\nclassification_report_train_lr = classification_report(y_train, y_train_pred_lr, target_names=df['Drug'].unique())\n\n# Evaluate the Logistic Regression model on the testing dataset\naccuracy_test_lr = accuracy_score(y_test, y_test_pred_lr)\nconfusion_matrix_test_lr = confusion_matrix(y_test, y_test_pred_lr)\nclassification_report_test_lr = classification_report(y_test, y_test_pred_lr, target_names=df['Drug'].unique())\n\n# Print evaluation metrics for Logistic Regression on training dataset\nprint(\"Evaluation Metrics for Logistic Regression on Training Dataset:\")\nprint(f\"Accuracy: {accuracy_train_lr:.2f}\")\nprint(\"Confusion Matrix:\")\nprint(confusion_matrix_train_lr)\nprint(\"Classification Report:\")\nprint(classification_report_train_lr)\n\n# Print evaluation metrics for Logistic Regression on testing dataset\nprint(\"\\nEvaluation Metrics for Logistic Regression on Testing Dataset:\")\nprint(f\"Accuracy: {accuracy_test_lr:.2f}\")\nprint(\"Confusion Matrix:\")\nprint(confusion_matrix_test_lr)\nprint(\"Classification Report:\")\nprint(classification_report_test_lr)","metadata":{"_uuid":"218f03a6-fc25-401b-96a1-31cb9c9d8b85","_cell_guid":"773ee909-eb46-4b13-863a-1d6a448596d7","trusted":true,"collapsed":false,"execution":{"iopub.status.busy":"2024-12-29T18:31:10.754426Z","iopub.execute_input":"2024-12-29T18:31:10.754860Z","iopub.status.idle":"2024-12-29T18:31:10.809009Z","shell.execute_reply.started":"2024-12-29T18:31:10.754827Z","shell.execute_reply":"2024-12-29T18:31:10.807811Z"},"jupyter":{"outputs_hidden":false}},"outputs":[{"name":"stdout","text":"Evaluation Metrics for Logistic Regression on Training Dataset:\nAccuracy: 0.94\nConfusion Matrix:\n[[ 6  0  0  0  1]\n [ 0  8  0  0  0]\n [ 0  0  8  1  0]\n [ 0  0  0 23  4]\n [ 0  0  0  0 49]]\nClassification Report:\n              precision    recall  f1-score   support\n\n       drugY       1.00      0.86      0.92         7\n       drugC       1.00      1.00      1.00         8\n       drugX       1.00      0.89      0.94         9\n       drugA       0.96      0.85      0.90        27\n       drugB       0.91      1.00      0.95        49\n\n    accuracy                           0.94       100\n   macro avg       0.97      0.92      0.94       100\nweighted avg       0.94      0.94      0.94       100\n\n\nEvaluation Metrics for Logistic Regression on Testing Dataset:\nAccuracy: 0.98\nConfusion Matrix:\n[[16  0  0  0  0]\n [ 0  8  0  0  0]\n [ 0  0  7  0  0]\n [ 0  0  0 26  1]\n [ 0  1  0  0 41]]\nClassification Report:\n              precision    recall  f1-score   support\n\n       drugY       1.00      1.00      1.00        16\n       drugC       0.89      1.00      0.94         8\n       drugX       1.00      1.00      1.00         7\n       drugA       1.00      0.96      0.98        27\n       drugB       0.98      0.98      0.98        42\n\n    accuracy                           0.98       100\n   macro avg       0.97      0.99      0.98       100\nweighted avg       0.98      0.98      0.98       100\n\n","output_type":"stream"}],"execution_count":22},{"cell_type":"markdown","source":"**Prompt:**  \nPlease show the equation and thier coefficients of logistic model.","metadata":{"_uuid":"95760cfb-7eaa-482c-ac33-c5bc2d48ef20","_cell_guid":"77c90b16-d320-4ded-af60-735d5ab84c3b","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}},{"cell_type":"code","source":"# Access the coefficients (weights) of the Logistic Regression model\ncoefficients = logistic_reg.coef_\n\n# Access the intercept of the Logistic Regression model\nintercept = logistic_reg.intercept_\n\n# Display the coefficients and intercept\nprint(\"Logistic Regression Coefficients:\")\nfor feature, coef in zip(features_to_use, coefficients[0]):\n    print(f\"{feature}: {coef:.4f}\")\nprint(f\"Intercept: {intercept[0]:.4f}\")","metadata":{"_uuid":"30424c86-9076-4b2c-bb1b-1ee12d1b083c","_cell_guid":"894f5fa5-37e6-4c29-8f89-4685722db3cb","trusted":true,"collapsed":false,"execution":{"iopub.status.busy":"2024-12-29T18:32:14.209013Z","iopub.execute_input":"2024-12-29T18:32:14.209436Z","iopub.status.idle":"2024-12-29T18:32:14.216505Z","shell.execute_reply.started":"2024-12-29T18:32:14.209405Z","shell.execute_reply":"2024-12-29T18:32:14.215267Z"},"jupyter":{"outputs_hidden":false}},"outputs":[{"name":"stdout","text":"Logistic Regression Coefficients:\nAge: -0.5577\nBP_encoded: 1.7187\nCholesterol_encoded: -0.1914\nNa/K: -0.9667\nIntercept: -2.7108\n","output_type":"stream"}],"execution_count":23},{"cell_type":"markdown","source":"### Guideline Map: \n\n\n##### Next Notebook:\n###### [Project 2: Loan Credit Prediction](https://www.kaggle.com/code/rouzbeh/loan-credit-prediction)  \n\n##### Course Content:   \n###### [1. Dive into Data Science: Content -->](https://www.kaggle.com/rouzbeh/1-dive-into-ds-content)  \n\n##### Next Courses:   \n###### [2. Statistics for Data Science -->](https://www.kaggle.com/code/rouzbeh/2-statistics-for-ds-content)  \n###### [3. Data Mining and Applied Machine Learning -->](https://www.kaggle.com/code/rouzbeh/3-data-mining-applied-machine-learning-content)","metadata":{"_uuid":"a7489a70-e207-4915-b705-fd7f58847e75","_cell_guid":"a21585da-256f-43b9-9c10-166404f3861e","trusted":true,"collapsed":false,"jupyter":{"outputs_hidden":false}}}]}