CodeMint 0.1.0__py3-none-any.whl

CodeMint/__init__.py

@@ -0,0 +1,2 @@
+"""CodeMint: offline-friendly experiment source display package."""
+__version__ = "0.1.0"

CodeMint/code1.py

@@ -0,0 +1,4 @@
+# CodeMint source holder: importing this module displays the stored experiment.
+SOURCE = 'import numpy as np\n\na = np.array([10, 20, 30, 40, 50])\n\nprint(a)\n\nprint(a.ndim)\n\n# 02. 2D Array\n\nimport numpy as np\n\na = np.array([[1, 2, 3],\n              [4, 5, 6]])\n\nprint(a)\n\nprint(a.ndim)\n\n# 03. 3D Array\n\nimport numpy as np\n\na = np.array([\n    [[1, 2], [3, 4]],\n    [[5, 6], [7, 8]]\n])\n\nprint(a.ndim)\n\n# 04. Shape\n\nimport numpy as np\n\na = np.array([[1, 2, 3],\n              [4, 5, 6]])\n\nprint(a.shape)\n\n# 05. Size\n\nprint(a.size)\n\n# 06. Data Type\n\nimport numpy as np\n\na = np.array([1, 2, 3])\n\nprint(a.dtype)\n\n# 07. arange()\n\nimport numpy as np\n\na = np.arange(1, 11)\n\nprint(a)\n\n# 08. linspace()\n\nimport numpy as np\n\na = np.linspace(0, 10, 5)\n\nprint(a)\n\n# 09. reshape()\n\nimport numpy as np\n\na = np.arange(1, 13)\n\nb = a.reshape(3, 4)\n\nprint(b)\n\n# 10. flatten()\n\nimport numpy as np\n\na = np.array([[1, 2], [3, 4]])\n\nprint(a.flatten())\n\n# 11. zeros and ones\n\nimport numpy as np\n\nprint(np.zeros((3, 3)))\n\nprint(np.ones((2, 4)))\n\n# 12. Identity Matrix and Random Numbers\n\nimport numpy as np\n\nprint(np.eye(3))\n\nprint(np.random.rand(3, 3))\n\n# 13. Array Indexing and Slicing\n\nimport numpy as np\n\na = np.array([10, 20, 30, 40])\n\nprint(a[2])\n\nprint(a[1:4])\n\n# 14. Aggregate Functions\n\nimport numpy as np\n\na = np.array([10, 20, 30, 40, 50])\n\nprint(np.sum(a))     # Sum of all elements\nprint(np.mean(a))    # Average of elements\nprint(np.max(a))     # Largest value\nprint(np.min(a))     # Smallest value\nprint(np.std(a))     # Standard deviation\n\n# 15. Matrix Addition & Multiplication\n\nimport numpy as np\n\nA = np.array([[1, 2],\n              [3, 4]])\n\nB = np.array([[5, 6],\n              [7, 8]])\n\nprint(A + B)\n\nprint(np.dot(A, B))\n\nprint(A.T)                     # Transpose\n\nprint(np.linalg.det(A))        # Determinant\n\nprint(np.linalg.inv(A))        # Inverse\n\nprint(np.linalg.eigvals(A))    # Eigen values\n\n# Descriptive Statistics\n\nimport numpy as np\nfrom scipy import stats\nfrom scipy.stats import pearsonr, linregress\nfrom scipy.optimize import minimize\nfrom sklearn.linear_model import LinearRegression\nimport pandas as pd\nimport statsmodels.api as sm\n\ndata = np.array([10, 20, 30, 40, 50, 60, 70, 80])\n\nprint("Mean =", stats.tmean(data))\n\nprint("Mode =", stats.mode(data))\n\nprint("Skewness =", stats.skew(data))\n\n# Correlation\n\nfrom scipy.stats import pearsonr\n\nx = [10, 20, 30, 40, 50]\ny = [15, 25, 35, 45, 55]\n\ncorr, p = pearsonr(x, y)\n\nprint(corr)\n\nprint(p)\n\n# Linear Regression using SciPy\n\nfrom scipy.stats import linregress\n\nx = [1, 2, 3, 4, 5]\ny = [2, 4, 5, 4, 5]\n\nresult = linregress(x, y)\n\nprint("Slope =", result.slope)\n\nprint("Intercept =", result.intercept)\n\nprint("R-Value =", result.rvalue)\n\nprint("P-Value =", result.pvalue)\n\n# Statsmodels (statsmodels.api)\n\nimport pandas as pd\nimport statsmodels.api as sm\n\nx = pd.DataFrame({\n    \'x\': [1, 2, 3, 4, 5]\n})\n\ny = [2, 4, 5, 4, 5]\n\nx = sm.add_constant(x)\n\nmodel = sm.OLS(y, x)\n\nresult = model.fit()\n\nprint(result.summary())\n\n# Statsmodels (statsmodels.formula.api)\n\nimport pandas as pd\nimport statsmodels.formula.api as smf\n\ndf = pd.DataFrame({\n    \'x\': [1, 2, 3, 4, 5],\n    \'y\': [2, 4, 5, 4, 5]\n})\n\nmodel = smf.ols(\'y ~ x\', data=df)\n\nresult = model.fit()\n\nprint(result.summary())\n\n# Scikit-Learn Linear Regression\n\nimport numpy as np\nfrom sklearn.linear_model import LinearRegression\n\nx = np.array([[1],\n              [2],\n              [3],\n              [4],\n              [5]])\n\ny = np.array([2, 4, 5, 4, 5])\n\nmodel = LinearRegression()\n\nmodel.fit(x, y)\n\nprint("Slope =", model.coef_)\n\nprint("Intercept =", model.intercept_)\n\nprediction = model.predict([[6]])\n\nprint("Prediction for x = 6 =", prediction)\n\nimport pandas as pd\nimport statsmodels.api as sm\nimport matplotlib.pyplot as plt\n\n# Data\nx = [1, 2, 3, 4, 5]\ny = [2, 4, 5, 4, 5]\n\n# Add constant for intercept\nX = sm.add_constant(x)\n\n# OLS Model\nmodel = sm.OLS(y, X)\nresult = model.fit()\n\n# Predicted values\ny_pred = result.predict(X)\n\n# Plot\nplt.scatter(x, y, label=\'Actual Data\')\nplt.plot(x, y_pred, label=\'OLS Regression Line\')\nplt.xlabel(\'X\')\nplt.ylabel(\'Y\')\nplt.title(\'OLS Regression\')\nplt.legend()\nplt.grid(True)\nplt.show()\n\n# Summary\nprint(result.summary())\n\n'
+
+print(SOURCE)

CodeMint/code2.py

@@ -0,0 +1,4 @@
+# CodeMint source holder: importing this module displays the stored experiment.
+SOURCE = 'import pandas as pd\nimport statsmodels.api as sm\nfrom mlxtend.data import boston_housing_data\nimport time\n\nfrom sklearn.metrics import (\n    r2_score,\n    mean_absolute_error,\n    mean_squared_error\n)\nimport numpy as np\n\n# ============================================================\n# 1. Load dataset\n# ============================================================\n\nX, y = boston_housing_data()\n\n# Column names\ncolumns = [\n    \'CRIM\', \'ZN\', \'INDUS\', \'CHAS\', \'NOX\', \'RM\',\n    \'AGE\', \'DIS\', \'RAD\', \'TAX\', \'PTRATIO\',\n    \'B\', \'LSTAT\'\n]\n\n# Create DataFrame\ndataset = pd.DataFrame(X, columns=columns)\n\n# Add target variable\ndataset["MEDV"] = y\n\n# Display first 5 rows\nprint("First 5 rows:")\nprint(dataset.head())\n\n# Display column names\nprint("\\nColumn Names:")\nprint(dataset.columns)\n\n# ============================================================\n# 2. Select Predictor and Target\n# ============================================================\n\nX = dataset["RM"]       # Independent variable\ny = dataset["MEDV"]     # Dependent variable\n\n# Add intercept (constant)\nX = sm.add_constant(X)\n\n# ============================================================\n# 3. Create and Train OLS Regression Model\n# ============================================================\n\nstart_time = time.time()\n\n# Create model\nmodel = sm.OLS(y, X)\n\n# Train model\nresult = model.fit()\n\nend_time = time.time()\n\n# ============================================================\n# 4. Display Model Summary\n# ============================================================\n\nprint("\\nRegression Model Summary:")\nprint(result.summary())\n\n# ============================================================\n# 5. Display Coefficients\n# ============================================================\n\nprint("\\nCoefficients:")\nprint(result.params)\n\n# ============================================================\n# 6. Make Predictions\n# ============================================================\n\npredictions = result.predict(X)\n\nprint("\\nFirst 10 Predictions:")\nprint(predictions.head(10))\n\n# ============================================================\n# 7. Regression Evaluation Metrics\n# ============================================================\n\nr2 = r2_score(y, predictions)\n\nmae = mean_absolute_error(y, predictions)\n\nmse = mean_squared_error(y, predictions)\n\nrmse = np.sqrt(mse)\n\n# Adjusted R²\nn = len(y)\np = X.shape[1] - 1\n\nadjusted_r2 = 1 - ((1 - r2) * (n - 1) / (n - p - 1))\n\nprint("\\n==============================")\nprint("REGRESSION EVALUATION METRICS")\nprint("==============================")\n\nprint(f"R² Score       : {r2:.4f}")\nprint(f"Adjusted R²    : {adjusted_r2:.4f}")\nprint(f"MAE            : {mae:.4f}")\nprint(f"MSE            : {mse:.4f}")\nprint(f"RMSE           : {rmse:.4f}")\n\n# ============================================================\n# 8. Execution Time\n# ============================================================\n\nexecution_time = end_time - start_time\n\nprint("\\n==============================")\nprint("EXECUTION TIME")\nprint("==============================")\n\nprint(f"Start Time     : {start_time:.6f} seconds")\nprint(f"End Time       : {end_time:.6f} seconds")\nprint(f"Execution Time : {execution_time:.6f} seconds")\n\nimport pandas as pd\nimport statsmodels.formula.api as smf\nfrom mlxtend.data import boston_housing_data\nimport time\n\nfrom sklearn.metrics import (\n    r2_score,\n    mean_absolute_error,\n    mean_squared_error\n)\nimport numpy as np\n\n# ============================================================\n# 1. Load dataset\n# ============================================================\n\nX, y = boston_housing_data()\n\n# Feature names\ncolumns = [\n    \'CRIM\', \'ZN\', \'INDUS\', \'CHAS\', \'NOX\', \'RM\',\n    \'AGE\', \'DIS\', \'RAD\', \'TAX\', \'PTRATIO\',\n    \'B\', \'LSTAT\'\n]\n\n# Create DataFrame\ndataset = pd.DataFrame(X, columns=columns)\n\n# Add target variable\ndataset[\'MEDV\'] = y\n\n# ============================================================\n# 2. Build and Train Model\n# ============================================================\n\nstart_time = time.time()\n\n# Build model\nmodel = smf.ols(formula=\'MEDV ~ RM\', data=dataset)\n\n# Train model\nresult = model.fit()\n\nend_time = time.time()\n\n# ============================================================\n# 3. Model Summary\n# ============================================================\n\nprint("\\nRegression Model Summary:")\nprint(result.summary())\n\n# ============================================================\n# 4. Coefficients\n# ============================================================\n\nprint("\\nCoefficients:")\nprint(result.params)\n\n# ============================================================\n# 5. Predictions\n# ============================================================\n\npredictions = result.predict(dataset)\n\nprint("\\nFirst 10 Predictions:")\nprint(predictions.head(10))\n\n# ============================================================\n# 6. Regression Evaluation Metrics\n# ============================================================\n\nr2 = r2_score(dataset[\'MEDV\'], predictions)\n\nmae = mean_absolute_error(dataset[\'MEDV\'], predictions)\n\nmse = mean_squared_error(dataset[\'MEDV\'], predictions)\n\nrmse = np.sqrt(mse)\n\n# Number of observations\nn = len(dataset)\n\n# Number of predictors\np = 1\n\n# Adjusted R²\nadjusted_r2 = 1 - ((1 - r2) * (n - 1) / (n - p - 1))\n\nprint("\\n==============================")\nprint("REGRESSION EVALUATION METRICS")\nprint("==============================")\n\nprint(f"R² Score       : {r2:.4f}")\nprint(f"Adjusted R²    : {adjusted_r2:.4f}")\nprint(f"MAE            : {mae:.4f}")\nprint(f"MSE            : {mse:.4f}")\nprint(f"RMSE           : {rmse:.4f}")\n\n# ============================================================\n# 7. Execution Time\n# ============================================================\n\nexecution_time2 = end_time - start_time\n\nprint("\\n==============================")\nprint("EXECUTION TIME")\nprint("==============================")\n\nprint(f"Start Time     : {start_time:.6f} seconds")\nprint(f"End Time       : {end_time:.6f} seconds")\nprint(f"Execution Time : {execution_time2:.6f} seconds")\n\nimport pandas as pd\nfrom mlxtend.data import boston_housing_data\nfrom sklearn.linear_model import LinearRegression\nfrom sklearn.metrics import (\n    r2_score,\n    mean_absolute_error,\n    mean_squared_error\n)\nimport numpy as np\nimport time\n\n# ============================================================\n# 1. Load Boston Housing Dataset\n# ============================================================\n\nX, y = boston_housing_data()\n\n# Feature names\ncolumns = [\n    \'CRIM\', \'ZN\', \'INDUS\', \'CHAS\', \'NOX\',\n    \'RM\', \'AGE\', \'DIS\', \'RAD\', \'TAX\',\n    \'PTRATIO\', \'B\', \'LSTAT\'\n]\n\n# Create DataFrame\ndataset = pd.DataFrame(X, columns=columns)\n\n# Add target column\ndataset[\'MEDV\'] = y\n\n# Display first 5 rows\nprint("First 5 rows:")\nprint(dataset.head())\n\n# ============================================================\n# 2. Predictor and Target\n# ============================================================\n\nX = dataset[[\'RM\']]      # Independent variable\ny = dataset[\'MEDV\']      # Dependent variable\n\n# ============================================================\n# 3. Create and Train Model\n# ============================================================\n\nstart_time = time.time()\n\n# Create model\nmodel = LinearRegression()\n\n# Train model\nmodel.fit(X, y)\n\nend_time = time.time()\n\n# ============================================================\n# 4. Coefficient and Intercept\n# ============================================================\n\nprint("\\nCoefficient:", model.coef_[0])\n\nprint("Intercept:", model.intercept_)\n\n# ============================================================\n# 5. Regression Equation\n# ============================================================\n\nprint("\\nRegression Equation:")\nprint(\n    f"MEDV = {model.intercept_:.4f} + "\n    f"({model.coef_[0]:.4f} × RM)"\n)\n\n# ============================================================\n# 6. Predictions\n# ============================================================\n\nprediction = model.predict(X)\n\nprint("\\nFirst 10 Predictions:")\nprint(prediction[:10])\n\n# ============================================================\n# 7. Regression Evaluation Metrics\n# ============================================================\n\nr2 = r2_score(y, prediction)\n\nmae = mean_absolute_error(y, prediction)\n\nmse = mean_squared_error(y, prediction)\n\nrmse = np.sqrt(mse)\n\n# Number of observations\nn = len(y)\n\n# Number of predictors\np = X.shape[1]\n\n# Adjusted R²\nadjusted_r2 = 1 - (\n    (1 - r2) * (n - 1) / (n - p - 1)\n)\n\nprint("\\n==============================")\nprint("REGRESSION EVALUATION METRICS")\nprint("==============================")\n\nprint(f"R² Score       : {r2:.4f}")\nprint(f"Adjusted R²    : {adjusted_r2:.4f}")\nprint(f"MAE            : {mae:.4f}")\nprint(f"MSE            : {mse:.4f}")\nprint(f"RMSE           : {rmse:.4f}")\n\n# ============================================================\n# 8. Execution Time\n# ============================================================\n\nexecution_time3 = end_time - start_time\n\nprint("\\n==============================")\nprint("EXECUTION TIME")\nprint("==============================")\n\nprint(f"Start Time     : {start_time:.6f} seconds")\nprint(f"End Time       : {end_time:.6f} seconds")\nprint(f"Execution Time : {execution_time3:.6f} seconds")\n\n# Store execution times\nstatsmodels_api_time = execution_time1\nstatsmodels_formula_time = execution_time2\nsklearn_time = execution_time3\n\n# Create comparison table\ncomparison = pd.DataFrame({\n    "Model": [  \n        "statsmodels.api (OLS)",\n        "statsmodels.formula.api (OLS)",\n        "scikit-learn (LinearRegression)"\n    ],\n    "Execution Time (seconds)": [\n        statsmodels_api_time,\n        statsmodels_formula_time,\n        sklearn_time\n    ]\n})\n\nprint("\\nExecution Time Comparison")\nprint(comparison)\n\n\n\n'
+
+print(SOURCE)

CodeMint/code3.py

@@ -0,0 +1,4 @@
+# CodeMint source holder: importing this module displays the stored experiment.
+SOURCE = 'import pandas as pd\nfrom sklearn.linear_model import LinearRegression\nfrom sklearn.metrics import (\n    r2_score,\n    mean_absolute_error,\n    mean_squared_error\n)\nimport numpy as np\nimport time\n\nfrom mlxtend.data import boston_housing_data\n\n# ============================================================\n# 1. Load dataset from library\n# ============================================================\n\nX_data, y_data = boston_housing_data()\n\n# Column names of Boston Housing dataset\ncolumns = [\n    \'CRIM\', \'ZN\', \'INDUS\', \'CHAS\', \'NOX\',\n    \'RM\', \'AGE\', \'DIS\', \'RAD\', \'TAX\',\n    \'PTRATIO\', \'B\', \'LSTAT\'\n]\n\n# Create DataFrame for input features\ndataset = pd.DataFrame(X_data, columns=columns)\n\n# Add target column\ndataset[\'MEDV\'] = y_data\n\n# Select required columns and remove missing values\ndataset = dataset[\n    [\'RM\', \'LSTAT\', \'PTRATIO\', \'MEDV\']\n].dropna()\n\n# ============================================================\n# 2. Independent and Dependent Variables\n# ============================================================\n\nX = dataset[[\'RM\', \'LSTAT\', \'PTRATIO\']]\ny = dataset[\'MEDV\']\n\n# ============================================================\n# 3. Create and Train Model\n# ============================================================\n\nstart_time = time.time()\n\n# Create model\nmodel = LinearRegression()\n\n# Train model\nmodel.fit(X, y)\n\nend_time = time.time()\n\n# ============================================================\n# 4. Coefficients and Intercept\n# ============================================================\n\nprint("Coefficients:", model.coef_)\nprint("Intercept:", model.intercept_)\n\n# ============================================================\n# 5. Regression Equation\n# ============================================================\n\nprint("\\nRegression Equation:")\n\nprint(\n    f"MEDV = {model.intercept_:.4f} + "\n    f"({model.coef_[0]:.4f} × RM) + "\n    f"({model.coef_[1]:.4f} × LSTAT) + "\n    f"({model.coef_[2]:.4f} × PTRATIO)"\n)\n\n# ============================================================\n# 6. Prediction for Test Input\n# ============================================================\n\nprediction = model.predict(\n    pd.DataFrame(\n        [[6, 10, 18]],\n        columns=[\'RM\', \'LSTAT\', \'PTRATIO\']\n    )\n)\n\nprint("\\nPredicted Price for RM=6, LSTAT=10, PTRATIO=18:")\nprint(prediction[0])\n\n# ============================================================\n# 7. Predictions for Entire Dataset\n# ============================================================\n\ny_pred = model.predict(X)\n\n# ============================================================\n# 8. Regression Evaluation Metrics\n# ============================================================\n\nr2 = r2_score(y, y_pred)\n\nmae = mean_absolute_error(y, y_pred)\n\nmse = mean_squared_error(y, y_pred)\n\nrmse = np.sqrt(mse)\n\n# Number of observations\nn = len(y)\n\n# Number of predictors\np = X.shape[1]\n\n# Adjusted R²\nadjusted_r2 = 1 - (\n    (1 - r2) * (n - 1) / (n - p - 1)\n)\n\nprint("\\n==============================")\nprint("REGRESSION EVALUATION METRICS")\nprint("==============================")\n\nprint(f"R² Score       : {r2:.4f}")\nprint(f"Adjusted R²    : {adjusted_r2:.4f}")\nprint(f"MAE            : {mae:.4f}")\nprint(f"MSE            : {mse:.4f}")\nprint(f"RMSE           : {rmse:.4f}")\n\n# ============================================================\n# 9. Execution Time\n# ============================================================\n\nexecution_time1 = end_time - start_time\n\nprint("\\n==============================")\nprint("EXECUTION TIME")\nprint("==============================")\n\nprint(f"Start Time     : {start_time:.6f} seconds")\nprint(f"End Time       : {end_time:.6f} seconds")\nprint(f"Execution Time : {execution_time1:.6f} seconds")\n\nimport pandas as pd\nimport statsmodels.api as sm\nfrom sklearn.metrics import (\n    r2_score,\n    mean_absolute_error,\n    mean_squared_error\n)\nimport numpy as np\nimport time\n\nfrom mlxtend.data import boston_housing_data\n\n# ============================================================\n# 1. Load dataset from library\n# ============================================================\n\nX_data, y_data = boston_housing_data()\n\n# Column names of Boston Housing dataset\ncolumns = [\n    \'CRIM\', \'ZN\', \'INDUS\', \'CHAS\', \'NOX\',\n    \'RM\', \'AGE\', \'DIS\', \'RAD\', \'TAX\',\n    \'PTRATIO\', \'B\', \'LSTAT\'\n]\n\n# Create DataFrame\ndataset = pd.DataFrame(X_data, columns=columns)\n\n# Add target column\ndataset[\'MEDV\'] = y_data\n\n# Select required columns\ndataset = dataset[\n    [\'RM\', \'LSTAT\', \'PTRATIO\', \'MEDV\']\n].dropna()\n\n# ============================================================\n# 2. Independent and Dependent Variables\n# ============================================================\n\nX = dataset[[\'RM\', \'LSTAT\', \'PTRATIO\']]\ny = dataset[\'MEDV\']\n\n# Add constant\nX = sm.add_constant(X)\n\n# ============================================================\n# 3. Build and Train Model\n# ============================================================\n\nstart_time = time.time()\n\n# Build model\nmodel = sm.OLS(y, X)\n\n# Fit model\nresult = model.fit()\n\nend_time = time.time()\n\n# ============================================================\n# 4. Summary\n# ============================================================\n\nprint(result.summary())\n\n# ============================================================\n# 5. Coefficients\n# ============================================================\n\nprint("\\nCoefficients:")\nprint(result.params)\n\n# ============================================================\n# 6. Predictions for Entire Dataset\n# ============================================================\n\npredictions = result.predict(X)\n\nprint("\\nFirst 10 Predictions:")\nprint(predictions.head(10))\n\n# ============================================================\n# 7. Regression Evaluation Metrics\n# ============================================================\n\nr2 = r2_score(y, predictions)\n\nmae = mean_absolute_error(y, predictions)\n\nmse = mean_squared_error(y, predictions)\n\nrmse = np.sqrt(mse)\n\n# Number of observations\nn = len(y)\n\n# Number of predictors\np = 3\n\n# Adjusted R²\nadjusted_r2 = 1 - (\n    (1 - r2) * (n - 1) / (n - p - 1)\n)\n\nprint("\\n==============================")\nprint("REGRESSION EVALUATION METRICS")\nprint("==============================")\n\nprint(f"R² Score       : {r2:.4f}")\nprint(f"Adjusted R²    : {adjusted_r2:.4f}")\nprint(f"MAE            : {mae:.4f}")\nprint(f"MSE            : {mse:.4f}")\nprint(f"RMSE           : {rmse:.4f}")\n\n# ============================================================\n# 8. New Data Prediction\n# ============================================================\n\nnew_data = pd.DataFrame({\n    \'const\': [1],\n    \'RM\': [6],\n    \'LSTAT\': [10],\n    \'PTRATIO\': [18]\n})\n\nprediction = result.predict(new_data)\n\nprint("\\nPredicted Price:")\nprint(f"{prediction.iloc[0]:.4f}")\n\n# ============================================================\n# 9. Execution Time\n# ============================================================\n\nexecution_time2 = end_time - start_time\n\nprint("\\n==============================")\nprint("EXECUTION TIME")\nprint("==============================")\n\nprint(f"Start Time     : {start_time:.6f} seconds")\nprint(f"End Time       : {end_time:.6f} seconds")\nprint(f"Execution Time : {execution_time2:.6f} seconds")\n\nimport pandas as pd\nimport statsmodels.formula.api as smf\nfrom sklearn.metrics import (\n    r2_score,\n    mean_absolute_error,\n    mean_squared_error\n)\nimport numpy as np\nimport time\n\nfrom mlxtend.data import boston_housing_data\n\n# ============================================================\n# 1. Load dataset from library\n# ============================================================\n\nX_data, y_data = boston_housing_data()\n\n# Column names of Boston Housing dataset\ncolumns = [\n    \'CRIM\', \'ZN\', \'INDUS\', \'CHAS\', \'NOX\',\n    \'RM\', \'AGE\', \'DIS\', \'RAD\', \'TAX\',\n    \'PTRATIO\', \'B\', \'LSTAT\'\n]\n\n# Create DataFrame\ndataset = pd.DataFrame(X_data, columns=columns)\n\n# Add target column\ndataset[\'MEDV\'] = y_data\n\n# Select required columns\ndataset = dataset[\n    [\'RM\', \'LSTAT\', \'PTRATIO\', \'MEDV\']\n].dropna()\n\n# ============================================================\n# 2. Build and Train Model\n# ============================================================\n\nstart_time = time.time()\n\n# Build model\nmodel = smf.ols(\n    formula=\'MEDV ~ RM + LSTAT + PTRATIO\',\n    data=dataset\n)\n\n# Fit model\nresult = model.fit()\n\nend_time = time.time()\n\n# ============================================================\n# 3. Model Summary\n# ============================================================\n\nprint(result.summary())\n\n# ============================================================\n# 4. Coefficients\n# ============================================================\n\nprint("\\nCoefficients:")\nprint(result.params)\n\n# ============================================================\n# 5. Predictions for Entire Dataset\n# ============================================================\n\npredictions = result.predict(dataset)\n\nprint("\\nFirst 10 Predictions:")\nprint(predictions.head(10))\n\n# ============================================================\n# 6. Regression Evaluation Metrics\n# ============================================================\n\nr2 = r2_score(dataset[\'MEDV\'], predictions)\n\nmae = mean_absolute_error(\n    dataset[\'MEDV\'],\n    predictions\n)\n\nmse = mean_squared_error(\n    dataset[\'MEDV\'],\n    predictions\n)\n\nrmse = np.sqrt(mse)\n\n# Number of observations\nn = len(dataset)\n\n# Number of predictors\np = 3\n\n# Adjusted R²\nadjusted_r2 = 1 - (\n    (1 - r2) * (n - 1) / (n - p - 1)\n)\n\nprint("\\n==============================")\nprint("REGRESSION EVALUATION METRICS")\nprint("==============================")\n\nprint(f"R² Score       : {r2:.4f}")\nprint(f"Adjusted R²    : {adjusted_r2:.4f}")\nprint(f"MAE            : {mae:.4f}")\nprint(f"MSE            : {mse:.4f}")\nprint(f"RMSE           : {rmse:.4f}")\n\n# ============================================================\n# 7. New Data Prediction\n# ============================================================\n\nnew_data = pd.DataFrame({\n    \'RM\': [6],\n    \'LSTAT\': [10],\n    \'PTRATIO\': [18]\n})\n\nprediction = result.predict(new_data)\n\nprint("\\nPredicted Price:")\nprint(f"{prediction.iloc[0]:.4f}")\n\n# ============================================================\n# 8. Execution Time\n# ============================================================\n\nexecution_time3 = end_time - start_time\n\nprint("\\n==============================")\nprint("EXECUTION TIME")\nprint("==============================")\n\nprint(f"Start Time     : {start_time:.6f} seconds")\nprint(f"End Time       : {end_time:.6f} seconds")\nprint(f"Execution Time : {execution_time3:.6f} seconds")\n\n# Store execution times\nstatsmodels_api_time = execution_time2\nstatsmodels_formula_time = execution_time3\nsklearn_time = execution_time1\n\n# Create comparison table\ncomparison = pd.DataFrame({\n    "Model": [\n        "statsmodels.api (OLS)",\n        "statsmodels.formula.api (OLS)",\n        "scikit-learn (LinearRegression)"\n    ],\n    "Execution Time (seconds)": [\n        statsmodels_api_time,\n        statsmodels_formula_time,\n        sklearn_time\n    ]\n})\n\nprint("\\nExecution Time Comparison")\nprint(comparison)\n\n'
+
+print(SOURCE)

CodeMint/code4.py

@@ -0,0 +1,4 @@
+# CodeMint source holder: importing this module displays the stored experiment.
+SOURCE = '# Import libraries\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\n\nfrom sklearn.datasets import fetch_openml\nfrom sklearn.linear_model import LinearRegression\nfrom sklearn.preprocessing import PolynomialFeatures\nfrom sklearn.pipeline import make_pipeline\nfrom sklearn.metrics import r2_score, mean_absolute_error, mean_squared_error\n\n# Load Boston Housing Dataset\nboston = fetch_openml(name="boston", version=1, as_frame=True)\n\n# Create DataFrame\ndataset = boston.frame\n\n# Rename target column\ndataset.rename(columns={"MEDV": "target"}, inplace=True)\n\n# Convert target column to numeric\ndataset["target"] = pd.to_numeric(dataset["target"])\n\n# Predictor and Target\npredictor = "LSTAT"\nX = dataset[[predictor]].values\ny = dataset["target"].values\n\n# Create Polynomial Regression Models\nquadratic_predictor = make_pipeline(\n    PolynomialFeatures(degree=2, include_bias=False),\n    LinearRegression()\n)\n\ncubic_predictor = make_pipeline(\n    PolynomialFeatures(degree=3, include_bias=False),\n    LinearRegression()\n)\n\ndegree4_predictor = make_pipeline(\n    PolynomialFeatures(degree=4, include_bias=False),\n    LinearRegression(fit_intercept=True)\n)\n\n# Generate smooth X values for plotting\nxt = np.arange(0, 40, 0.1).reshape(-1, 1)\n\n# Fit the models\nquadratic_predictor.fit(X, y)\ncubic_predictor.fit(X, y)\n\n# Scatter Plot with Polynomial Regression Curves\nplt.figure(figsize=(10,6))\n\nplt.scatter(X, y, color=\'black\', alpha=0.6, label=\'Actual Data\')\n\nplt.plot(\n    xt,\n    quadratic_predictor.predict(xt),\n    color=\'blue\',\n    linewidth=2,\n    label=\'Degree 2\'\n)\n\nplt.plot(\n    xt,\n    cubic_predictor.predict(xt),\n    color=\'red\',\n    linewidth=2,\n    label=\'Degree 3\'\n)\n\nplt.plot(\n    xt,\n    degree4_predictor.fit(X, y).predict(xt),\n    color=\'purple\',\n    linewidth=2,\n    label=\'Degree 4\'\n)\n\nplt.xlabel("LSTAT")\nplt.ylabel("House Price (MEDV)")\nplt.title("Polynomial Regression on Boston Housing Dataset")\nplt.legend()\nplt.grid(True)\nplt.show()\n\n# -----------------------------\n# Degree 2 Performance\n# -----------------------------\ny_pred2 = quadratic_predictor.predict(X)\n\nprint("\\nDegree 2")\nprint("R2 Score :", r2_score(y, y_pred2))\nprint("MAE      :", mean_absolute_error(y, y_pred2))\nprint("MSE      :", mean_squared_error(y, y_pred2))\nprint("RMSE     :", np.sqrt(mean_squared_error(y, y_pred2)))\n\n# -----------------------------\n# Degree 3 Performance\n# -----------------------------\ny_pred3 = cubic_predictor.predict(X)\n\nprint("\\nDegree 3")\nprint("R2 Score :", r2_score(y, y_pred3))\nprint("MAE      :", mean_absolute_error(y, y_pred3))\nprint("MSE      :", mean_squared_error(y, y_pred3))\nprint("RMSE     :", np.sqrt(mean_squared_error(y, y_pred3)))\n\ny_pred4 = degree4_predictor.predict(X)\n\nprint("\\nDegree 4")\nprint("R2 Score :", r2_score(y, y_pred3))\nprint("MAE      :", mean_absolute_error(y, y_pred3))\nprint("MSE      :", mean_squared_error(y, y_pred3))\nprint("RMSE     :", np.sqrt(mean_squared_error(y, y_pred3)))\n\nprint("\\nConclusion:")\n\nr2_degree2 = r2_score(y, y_pred2)\nr2_degree3 = r2_score(y, y_pred3)\nr2_degree4 = r2_score(y, y_pred4)\n\nif r2_degree2 > r2_degree3 and r2_degree2 > r2_degree4:\n    print("Degree 2 is the better polynomial regression model because it has the highest R² score and the lowest prediction error.")\nelif r2_degree3 > r2_degree2 and r2_degree3 > r2_degree4:\n    print("Degree 3 is the better polynomial regression model because it has the highest R² score and the lowest prediction error.")\nelse:\n    print("Degree 4 is the better polynomial regression model because it has the highest R² score and the lowest prediction error.")\n\n\n'
+
+print(SOURCE)

CodeMint/code5.py

@@ -0,0 +1,4 @@
+# CodeMint source holder: importing this module displays the stored experiment.
+SOURCE = 'import pandas as pd\nfrom sklearn.datasets import load_iris\nfrom sklearn.model_selection import train_test_split\n\n# Load Iris dataset\niris = load_iris()\n\ndf = pd.DataFrame(\n    iris.data,\n    columns=iris.feature_names\n)\n\ndf["target"] = iris.target\n\n# Binary classification\n# 0 = Setosa\n# 1 = Versicolor\n# Remove Virginica (2)\ndf = df[df["target"] != 2]\n\n# Features and target\nX = df[iris.feature_names]\ny = df["target"]\n\n# Same train-test split for all 3 packages\nX_train, X_test, y_train, y_test = train_test_split(\n    X,\n    y,\n    test_size=0.2,\n    random_state=42,\n    stratify=y\n)\n\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.metrics import accuracy_score, confusion_matrix\n\n# Create model\nmodel_sklearn = LogisticRegression()\n\n# Train\nmodel_sklearn.fit(X_train, y_train)\n\n# Prediction\ny_pred_sklearn = model_sklearn.predict(X_test)\n\n# Accuracy\naccuracy_sklearn = accuracy_score(y_test, y_pred_sklearn)\n\nprint("SKLEARN")\nprint("Accuracy:", accuracy_sklearn)\n\nprint("\\nConfusion Matrix:")\nprint(confusion_matrix(y_test, y_pred_sklearn))\n\nprint("\\nIntercept:")\nprint(model_sklearn.intercept_)\n\nprint("\\nCoefficients:")\nprint(model_sklearn.coef_)\n\nimport statsmodels.api as sm\nfrom sklearn.metrics import accuracy_score, confusion_matrix\n\n# Add intercept manually\nX_train_sm = sm.add_constant(X_train)\nX_test_sm = sm.add_constant(X_test)\n\n# Create model\nmodel_sm = sm.Logit(y_train, X_train_sm)\n\n# Train\nresult_sm = model_sm.fit()\n\n# Probability prediction\ny_prob_sm = result_sm.predict(X_test_sm)\n\n# Convert probability to class\ny_pred_sm = (y_prob_sm >= 0.5).astype(int)\n\n# Accuracy\naccuracy_sm = accuracy_score(y_test, y_pred_sm)\n\nprint("STATSMODELS.API")\nprint("Accuracy:", accuracy_sm)\n\nprint("\\nConfusion Matrix:")\nprint(confusion_matrix(y_test, y_pred_sm))\n\nprint("\\nCoefficients:")\nprint(result_sm.params)\n\n# Full statistical summary\nprint("\\nStatistical Summary:")\nprint(result_sm.summary())\n\nimport statsmodels.formula.api as smf\nfrom sklearn.metrics import accuracy_score, confusion_matrix\n\n# Create a dataframe for formula API\ntrain_data = X_train.copy()\ntrain_data["target"] = y_train\n\ntest_data = X_test.copy()\ntest_data["target"] = y_test\n\n# Rename columns for formula\ntrain_data.columns = [\n    "sepal_length",\n    "sepal_width",\n    "petal_length",\n    "petal_width",\n    "target"\n]\n\ntest_data.columns = [\n    "sepal_length",\n    "sepal_width",\n    "petal_length",\n    "petal_width",\n    "target"\n]\n\n# Create model using formula\nmodel_formula = smf.logit(\n    "target ~ sepal_length + sepal_width + petal_length + petal_width",\n    data=train_data\n)\n\n# Train\nresult_formula = model_formula.fit()\n\n# Probability prediction\ny_prob_formula = result_formula.predict(test_data)\n\n# Convert probability to class\ny_pred_formula = (y_prob_formula >= 0.5).astype(int)\n\n# Accuracy\naccuracy_formula = accuracy_score(y_test, y_pred_formula)\n\nprint("STATSMODELS.FORMULA.API")\nprint("Accuracy:", accuracy_formula)\n\nprint("\\nConfusion Matrix:")\nprint(confusion_matrix(y_test, y_pred_formula))\n\nprint("\\nCoefficients:")\nprint(result_formula.params)\n\n# Full statistical summary\nprint("\\nStatistical Summary:")\nprint(result_formula.summary())\n\n""\n\n# ============================================================\n# FINAL COMPARISON OF ALL 3 PACKAGES\n# ============================================================\n\nfrom sklearn.metrics import (\n    accuracy_score,\n    precision_score,\n    recall_score,\n    f1_score,\n    mean_squared_error,\n    mean_absolute_error,\n    log_loss\n)\n\n# ------------------------------------------------------------\n# Metrics for SKLEARN\n# ------------------------------------------------------------\n\naccuracy_sklearn = accuracy_score(y_test, y_pred_sklearn)\nprecision_sklearn = precision_score(y_test, y_pred_sklearn)\nrecall_sklearn = recall_score(y_test, y_pred_sklearn)\nf1_sklearn = f1_score(y_test, y_pred_sklearn)\nmse_sklearn = mean_squared_error(y_test, y_pred_sklearn)\nmae_sklearn = mean_absolute_error(y_test, y_pred_sklearn)\ny_prob_sklearn = model_sklearn.predict_proba(X_test)\nlogloss_sklearn = log_loss(y_test, y_prob_sklearn[:, 1])\n\n\n# ------------------------------------------------------------\n# Metrics for STATSMODELS.API\n# ------------------------------------------------------------\n\naccuracy_sm = accuracy_score(y_test, y_pred_sm)\nprecision_sm = precision_score(y_test, y_pred_sm)\nrecall_sm = recall_score(y_test, y_pred_sm)\nf1_sm = f1_score(y_test, y_pred_sm)\nmse_sm = mean_squared_error(y_test, y_pred_sm)\nmae_sm = mean_absolute_error(y_test, y_pred_sm)\nlogloss_sm = log_loss(y_test, y_prob_sm)\n\n\n# ------------------------------------------------------------\n# Metrics for STATSMODELS.FORMULA.API\n# ------------------------------------------------------------\n\naccuracy_formula = accuracy_score(y_test, y_pred_formula)\nprecision_formula = precision_score(y_test, y_pred_formula)\nrecall_formula = recall_score(y_test, y_pred_formula)\nf1_formula = f1_score(y_test, y_pred_formula)\nmse_formula = mean_squared_error(y_test, y_pred_formula)\nmae_formula = mean_absolute_error(y_test, y_pred_formula)\nlogloss_formula = log_loss(y_test, y_prob_formula)\n\n\n# ============================================================\n# CREATE FINAL COMPARISON TABLE\n# ============================================================\n\ncomparison = pd.DataFrame({\n\n    "Package": [\n        "sklearn",\n        "statsmodels.api",\n        "statsmodels.formula.api"\n    ],\n\n    "Accuracy": [\n        accuracy_sklearn,\n        accuracy_sm,\n        accuracy_formula\n    ],\n\n    "Precision": [\n        precision_sklearn,\n        precision_sm,\n        precision_formula\n    ],\n\n    "Recall": [\n        recall_sklearn,\n        recall_sm,\n        recall_formula\n    ],\n\n    "F1 Score": [\n        f1_sklearn,\n        f1_sm,\n        f1_formula\n    ],\n\n    "MSE": [\n        mse_sklearn,\n        mse_sm,\n        mse_formula\n    ],\n\n    "MAE": [\n        mae_sklearn,\n        mae_sm,\n        mae_formula\n    ],\n\n    "Log Loss": [\n        logloss_sklearn,\n        logloss_sm,\n        logloss_formula\n    ]\n})\n\n\n# ------------------------------------------------------------\n# Display rounded values\n# ------------------------------------------------------------\n\nprint("\\n================ FINAL COMPARISON ================\\n")\n\nprint(comparison.round(4))\n\nimport pandas as pd\n\nfrom sklearn.datasets import load_iris\nfrom sklearn.model_selection import train_test_split\n\n# Load dataset\niris = load_iris()\n\ndf = pd.DataFrame(\n    iris.data,\n    columns=iris.feature_names\n)\n\ndf["target"] = iris.target\n\n# Features and Target\nX = df[iris.feature_names]\ny = df["target"]\n\n# Train-Test Split\nX_train, X_test, y_train, y_test = train_test_split(\n    X,\n    y,\n    test_size=0.2,\n    random_state=42,\n    stratify=y\n)\n\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.metrics import accuracy_score, confusion_matrix\n\n# Create Model\nmodel_sklearn = LogisticRegression(\n    multi_class=\'multinomial\',\n    max_iter=1000\n)\n\n# Train\nmodel_sklearn.fit(X_train, y_train)\n\n# Predict\ny_pred_sklearn = model_sklearn.predict(X_test)\n\n# Probabilities\ny_prob_sklearn = model_sklearn.predict_proba(X_test)\n\n# Accuracy\naccuracy_sklearn = accuracy_score(\n    y_test,\n    y_pred_sklearn\n)\n\nprint("SKLEARN")\n\nprint("Accuracy:", accuracy_sklearn)\n\nprint("\\nConfusion Matrix:")\nprint(confusion_matrix(y_test, y_pred_sklearn))\n\nimport statsmodels.api as sm\nfrom sklearn.metrics import accuracy_score, confusion_matrix\n\n# Add Constant\nX_train_sm = sm.add_constant(X_train)\nX_test_sm = sm.add_constant(X_test)\n\n# Create Model\nmodel_sm = sm.MNLogit(\n    y_train,\n    X_train_sm\n)\n\n# Train\nresult_sm = model_sm.fit()\n\n# Probabilities\ny_prob_sm = result_sm.predict(X_test_sm)\n\n# Class Prediction\ny_pred_sm = y_prob_sm.idxmax(axis=1)\n\n# Accuracy\naccuracy_sm = accuracy_score(\n    y_test,\n    y_pred_sm\n)\n\nprint("STATSMODELS.API")\n\nprint("Accuracy:", accuracy_sm)\n\nprint("\\nConfusion Matrix:")\nprint(confusion_matrix(y_test, y_pred_sm))\n\nprint(result_sm.summary())\n\nimport statsmodels.formula.api as smf\nfrom sklearn.metrics import accuracy_score, confusion_matrix\n\n# Create DataFrame\ntrain_df = X_train.copy()\ntrain_df["target"] = y_train\n\ntest_df = X_test.copy()\ntest_df["target"] = y_test\n\n# Rename columns\ntrain_df.columns = [\n    "sepal_length",\n    "sepal_width",\n    "petal_length",\n    "petal_width",\n    "target"\n]\n\ntest_df.columns = [\n    "sepal_length",\n    "sepal_width",\n    "petal_length",\n    "petal_width",\n    "target"\n]\n\n# Create Model\nmodel_formula = smf.mnlogit(\n    "target ~ sepal_length + sepal_width + petal_length + petal_width",\n    data=train_df\n)\n\n# Train\nresult_formula = model_formula.fit()\n\n# Probability Prediction\ny_prob_formula = result_formula.predict(test_df)\n\n# Class Prediction\ny_pred_formula = y_prob_formula.idxmax(axis=1)\n\n# Accuracy\naccuracy_formula = accuracy_score(\n    y_test,\n    y_pred_formula\n)\n\nprint("STATSMODELS.FORMULA.API")\n\nprint("Accuracy:", accuracy_formula)\n\nprint("\\nConfusion Matrix:")\nprint(confusion_matrix(y_test, y_pred_formula))\n\nprint(result_formula.summary())\n\nfrom sklearn.metrics import (\n    accuracy_score,\n    precision_score,\n    recall_score,\n    f1_score,\n    mean_squared_error,\n    mean_absolute_error,\n    log_loss\n)\n\ncomparison = pd.DataFrame({\n\n    "Package":[\n        "sklearn",\n        "statsmodels.api",\n        "statsmodels.formula.api"\n    ],\n\n    "Accuracy":[\n        accuracy_score(y_test,y_pred_sklearn),\n        accuracy_score(y_test,y_pred_sm),\n        accuracy_score(y_test,y_pred_formula)\n    ],\n\n    "Precision":[\n        precision_score(y_test,y_pred_sklearn,average=\'weighted\'),\n        precision_score(y_test,y_pred_sm,average=\'weighted\'),\n        precision_score(y_test,y_pred_formula,average=\'weighted\')\n    ],\n\n    "Recall":[\n        recall_score(y_test,y_pred_sklearn,average=\'weighted\'),\n        recall_score(y_test,y_pred_sm,average=\'weighted\'),\n        recall_score(y_test,y_pred_formula,average=\'weighted\')\n    ],\n\n    "F1 Score":[\n        f1_score(y_test,y_pred_sklearn,average=\'weighted\'),\n        f1_score(y_test,y_pred_sm,average=\'weighted\'),\n        f1_score(y_test,y_pred_formula,average=\'weighted\')\n    ],\n\n    "MSE":[\n        mean_squared_error(y_test,y_pred_sklearn),\n        mean_squared_error(y_test,y_pred_sm),\n        mean_squared_error(y_test,y_pred_formula)\n    ],\n\n    "MAE":[\n        mean_absolute_error(y_test,y_pred_sklearn),\n        mean_absolute_error(y_test,y_pred_sm),\n        mean_absolute_error(y_test,y_pred_formula)\n    ],\n\n    "Log Loss":[\n        log_loss(y_test,y_prob_sklearn),\n        log_loss(y_test,y_prob_sm),\n        log_loss(y_test,y_prob_formula)\n    ]\n})\n\nprint("\\nFINAL COMPARISON\\n")\n\nprint(comparison.round(4))\n\n'
+
+print(SOURCE)

CodeMint/code6.py

@@ -0,0 +1,4 @@
+# CodeMint source holder: importing this module displays the stored experiment.
+SOURCE = '# -*- coding: utf-8 -*-\n"""4_timeseries_dji.ipynb\n\nAutomatically generated by Colab.\n\nOriginal file is located at\n    https://colab.research.google.com/drive/1WSG9Qy3bRgnl6mPIeN6cqwviZD6KFKXf\n"""\n\nimport matplotlib.pyplot as plt\nimport numpy as np\nimport pandas as pd\n\ndataset = pd.read_csv(\'dow_jones_index.data\')\ndataset.head()\n\nobservations = {}\n\nfor el in dataset[["stock", "close"]].iterrows():\n    stock = el[1].stock\n    close = float(el[1].close.replace("$", ""))\n    try:\n        observations[stock].append(close)\n    except KeyError:\n        observations[stock] = [close]\nprint(observations)\n\nX = []\nstock_names = sorted(observations.keys())\n\nfor stock in stock_names:\n    X.append(observations[stock])\n\nX = np.array(X)\nprint(X)\n\nfrom sklearn.linear_model import LinearRegression\nfrom sklearn.metrics import r2_score, mean_absolute_error\n\nX_train = X[:, :12]\ny_train = X[:, 12]\nregr_1 = LinearRegression()\nregr_1.fit(X_train, y_train)\n\nplot_vals = []\n\nfor offset in range(0, X.shape[1] - X_train.shape[1]):\n    X_test = X[:, offset : 12 + offset]\n    y_test = X[:, 12 + offset]\n\n    r2 = r2_score(y_test, regr_1.predict(X_test))\n    mae = mean_absolute_error(y_test, regr_1.predict(X_test))\n\n    print("offset=", offset, "r2_score=", r2)\n    print("offset=", offset, "MAE =", mae)\n\n    plot_vals.append((offset, r2, mae))\nprint()\nprint(\n    "r2_score: mean=",\n    np.mean([x[1] for x in plot_vals]),\n    "variance=",\n    np.var([x[1] for x in plot_vals]),\n)\nprint(\n    "mae_score: mean=",\n    np.mean([x[2] for x in plot_vals]),\n    "variance=",\n    np.var([x[2] for x in plot_vals]),\n)\n\nfig, ax1 = plt.subplots()\nax1.plot([x[0] for x in plot_vals], [x[1] for x in plot_vals], "b-")\nax1.plot(plot_vals[0][0], plot_vals[0][1], "bo")\nax1.set_xlabel("test week")\n\n# Make the y-axis label and tick labels match the line color.\nax1.set_ylabel("r2_score", color="b")\n\nfor tl in ax1.get_yticklabels():\n    tl.set_color("b")\n\nax1.set_ylim([0.9, 1.1])\nax2 = ax1.twinx()\nax2.plot([x[0] for x in plot_vals], [x[2] for x in plot_vals], "r-")\nax2.plot(plot_vals[0][0], plot_vals[0][2], "ro")\nax2.set_ylabel("mae score", color="r")\n\nfor tl in ax2.get_yticklabels():\n    tl.set_color("r")\n\nax2.set_ylim([0, 3.3])\nplt.xlim([-0.1, 12.1])\nplt.show()\n\ntraining_len = 5\nX_train_short = X[:, :training_len]\ny_train_short = X[:, training_len]\n\nfor offset in range(1, 12 - training_len):\n    X_train_short = np.vstack((X_train_short, X[:, offset : training_len + offset]))\n    y_train_short = np.concatenate((y_train_short, X[:, training_len + offset]))\n\nregr_2 = LinearRegression()\nregr_2.fit(X_train_short, y_train_short)\n\nplot_vals = []\n\nfor offset in range(0, X.shape[1] - X_train.shape[1]):\n    X_test = X[:, 12 - training_len + offset : 12 + offset]\n    y_test = X[:, 12 + offset]\n\n    r2 = r2_score(y_test, regr_2.predict(X_test))\n    mae = mean_absolute_error(y_test, regr_2.predict(X_test))\n\n    print("offset=", offset, "r2_score=", r2)\n    print("offset=", offset, "MAE =", mae)\n\n    plot_vals.append((offset, r2, mae))\n\nprint(\n    "\\n r2_score: mean=",\n    np.mean([x[1] for x in plot_vals]),\n    "variance=",\n    np.var([x[1] for x in plot_vals]),\n)\nprint(\n    "mae_score: mean=",\n    np.mean([x[2] for x in plot_vals]),\n    "variance=",\n    np.var([x[2] for x in plot_vals]),\n)\n\nfig, ax1 = plt.subplots()\nax1.plot([x[0] for x in plot_vals], [x[1] for x in plot_vals], "b-")\nax1.plot(plot_vals[0][0], plot_vals[0][1], "bo")\nax1.set_xlabel("test week")\n\n# Make the y-axis label and tick labels match the line color.\nax1.set_ylabel("r2_score", color="b")\n\nfor tl in ax1.get_yticklabels():\n    tl.set_color("b")\n\nax1.set_ylim([0.95, 1.05])\nax2 = ax1.twinx()\nax2.plot([x[0] for x in plot_vals], [x[2] for x in plot_vals], "r-")\nax2.plot(plot_vals[0][0], plot_vals[0][2], "ro")\nax2.set_ylabel("mae score", color="r")\n\nfor tl in ax2.get_yticklabels():\n    tl.set_color("r")\n\nax2.set_ylim([0, 2.2])\nplt.xlim([-0.1, 12.1])\nplt.show()\n\ntraining_lens = range(1, 13)\nmodels = {}\n\nfor training_len in training_lens:\n    X_train_short = X[:, :training_len]\n    y_train_short = X[:, training_len]\n\n    for offset in range(1, 12 - training_len):\n        X_train_short = np.vstack((X_train_short, X[:, offset : training_len + offset]))\n        y_train_short = np.concatenate((y_train_short, X[:, training_len + offset]))\n        regr_x = LinearRegression()\n        regr_x.fit(X_train_short, y_train_short)\n        models[training_len] = regr_x\n        plot_vals = []\n\n        for offset in range(0, X.shape[1] - X_train.shape[1]):\n            X_test = X[:, 12 - training_len + offset : 12 + offset]\n            y_test = X[:, 12 + offset]\n            r2 = r2_score(y_test, regr_x.predict(X_test))\n            mae = mean_absolute_error(y_test, regr_x.predict(X_test))\n            plot_vals.append((offset, r2, mae))\n\n        fig, ax1 = plt.subplots()\n        ax1.plot([x[0] for x in plot_vals], [x[1] for x in plot_vals], "b-")\n        ax1.plot(plot_vals[0][0], plot_vals[0][1], "bo")\n        ax1.set_xlabel("test week")\n        # Make the y-axis label and tick labels match the line color.\n        ax1.set_ylabel("r2_score", color="b")\n\n        for tl in ax1.get_yticklabels():\n            tl.set_color("b")\n\n        ax1.set_ylim([0.95, 1.05])\n        ax2 = ax1.twinx()\n        ax2.plot([x[0] for x in plot_vals], [x[2] for x in plot_vals], "r-")\n        ax2.plot(plot_vals[0][0], plot_vals[0][2], "ro")\n        ax2.set_ylabel("mae score", color="r")\n\n        for tl in ax2.get_yticklabels():\n            tl.set_color("r")\n\n        ax2.set_ylim([0, max([2.2, 1.1 * np.max([x[2] for x in plot_vals])])])\n        plt.xlim([-0.1, 12.1])\n\n        plt.title("results with training_len={}".format(training_len))\n        plt.show()\n\n        print(\n            "r2_score: mean=",\n            np.mean([x[1] for x in plot_vals]),\n            "variance=",\n            np.var([x[1] for x in plot_vals]),\n        )\n        print(\n            "mae_score: mean=",\n            np.mean([x[2] for x in plot_vals]),\n            "variance=",\n            np.var([x[2] for x in plot_vals]),\n        )\n'
+
+print(SOURCE)

CodeMint/code7.py

@@ -0,0 +1,4 @@
+# CodeMint source holder: importing this module displays the stored experiment.
+SOURCE = '# -*- coding: utf-8 -*-\n"""EX7.ipynb\n\nAutomatically generated by Colab.\n\nOriginal file is located at\n    https://colab.research.google.com/drive/1ANPGc0Ul6_QHBQLcliKU4YOa-2hdjgnV\n"""\n\nfrom sklearn.datasets import load_iris\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.metrics import accuracy_score, classification_report, confusion_matrix\n\n# Load dataset\niris = load_iris()\n\nX = iris.data\ny = iris.target\n\n# Split data\nX_train, X_test, y_train, y_test = train_test_split(\n    X, y, test_size=0.2, random_state=42, stratify=y\n)\n\n# Multiclass Logistic Regression\nmodel = LogisticRegression(\n    multi_class="multinomial",\n    max_iter=200\n)\n\n# Train\nmodel.fit(X_train, y_train)\n\n# Prediction\ny_pred = model.predict(X_test)\n\n# Evaluation\nprint("Accuracy:", accuracy_score(y_test, y_pred))\n\nprint("\\nClassification Report:")\nprint(classification_report(y_test, y_pred))\n\nprint("\\nConfusion Matrix:")\nprint(confusion_matrix(y_test, y_pred))\n\n# !pip install statsmodels\n\nimport statsmodels.api as sm\nfrom sklearn.datasets import load_iris\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import accuracy_score, classification_report\nimport pandas as pd\n\n# Load Iris dataset\niris = load_iris()\n\nX = pd.DataFrame(iris.data, columns=iris.feature_names)\ny = pd.Series(iris.target, name="target")\n\n# Add constant\nX = sm.add_constant(X)\n\n# Split data\nX_train, X_test, y_train, y_test = train_test_split(\n    X, y, test_size=0.2, random_state=42, stratify=y\n)\n\n# Multinomial Logistic Regression\nmodel = sm.MNLogit(y_train, X_train)\n\n# Fit model\nresult = model.fit()\n\n# Model summary\nprint(result.summary())\n\n# Prediction\npred_prob = result.predict(X_test)\n\n# Convert probabilities to class labels\ny_pred = pred_prob.idxmax(axis=1)\n\n# Evaluation\nprint("\\nAccuracy:", accuracy_score(y_test, y_pred))\n\nprint("\\nClassification Report:")\nprint(classification_report(y_test, y_pred))\n\nimport statsmodels.formula.api as smf\nfrom sklearn.datasets import load_iris\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import accuracy_score, classification_report\nimport pandas as pd\n\n# Load dataset\niris = load_iris()\n\ndf = pd.DataFrame(\n    iris.data,\n    columns=["sepal_length", "sepal_width",\n             "petal_length", "petal_width"]\n)\n\ndf["target"] = iris.target\n\n# Split dataset\ntrain, test = train_test_split(\n    df,\n    test_size=0.2,\n    random_state=42,\n    stratify=df["target"]\n)\n\n# Multinomial Logistic Regression\nmodel = smf.mnlogit(\n    "target ~ sepal_length + sepal_width + petal_length + petal_width",\n    data=train\n)\n\n# Fit model\nresult = model.fit()\n\n# Display summary\nprint(result.summary())\n\n# Prediction\npred_prob = result.predict(test)\n\n# Convert probabilities to class\ny_pred = pred_prob.idxmax(axis=1)\n\n# Actual values\ny_test = test["target"]\n\n# Evaluation\nprint("\\nAccuracy:", accuracy_score(y_test, y_pred))\n\nprint("\\nClassification Report:")\nprint(classification_report(y_test, y_pred))\n'
+
+print(SOURCE)

CodeMint/code8.py

@@ -0,0 +1,4 @@
+# CodeMint source holder: importing this module displays the stored experiment.
+SOURCE = '# -*- coding: utf-8 -*-\n"""RA_EX_8.ipynb\n\nAutomatically generated by Colab.\n\nOriginal file is located at\n    https://colab.research.google.com/drive/1VHtmPdUQ7appWAqWnBqykFPYEZgV9O2T\n"""\n\nimport pandas as pd\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\nfrom sklearn.preprocessing import LabelEncoder\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.metrics import (\n    accuracy_score,\n    classification_report,\n    confusion_matrix\n)\nfrom sklearn.utils import resample\n\n\n# ==========================================================\n# 1. LOAD KDD CUP 1999 DATASET\n# ==========================================================\n\ncolumns = [\n    "duration", "protocol_type", "service", "flag",\n    "src_bytes", "dst_bytes", "land", "wrong_fragment",\n    "urgent", "hot", "num_failed_logins", "logged_in",\n    "num_compromised", "root_shell", "su_attempted",\n    "num_root", "num_file_creations", "num_shells",\n    "num_access_files", "num_outbound_cmds",\n    "is_host_login", "is_guest_login", "count",\n    "srv_count", "serror_rate", "srv_serror_rate",\n    "rerror_rate", "srv_rerror_rate", "same_srv_rate",\n    "diff_srv_rate", "srv_diff_host_rate",\n    "dst_host_count", "dst_host_srv_count",\n    "dst_host_same_srv_rate", "dst_host_diff_srv_rate",\n    "dst_host_same_src_port_rate",\n    "dst_host_srv_diff_host_rate",\n    "dst_host_serror_rate", "dst_host_srv_serror_rate",\n    "dst_host_rerror_rate", "dst_host_srv_rerror_rate",\n    "outcome"\n]\n\ndf = pd.read_csv(\n    "/content/kddcup.data_10_percent.gz",\n    names=columns\n)\n\n\n# ==========================================================\n# 2. USE A SMALLER SAMPLE\n# ==========================================================\n\ndf = df.sample(\n    n=50000,\n    random_state=42\n)\n\nprint("Dataset Shape:")\nprint(df.shape)\n\nprint("\\nOriginal Class Distribution:")\nprint(df["outcome"].value_counts())\n\n\n# ==========================================================\n# 3. REMOVE CLASSES WITH ONLY ONE SAMPLE\n# ==========================================================\n\nclass_counts = df["outcome"].value_counts()\n\nrare_classes = class_counts[\n    class_counts < 2\n].index\n\ndf = df[\n    ~df["outcome"].isin(rare_classes)\n]\n\nprint("\\nClass Distribution After Removing Rare Classes:")\nprint(df["outcome"].value_counts())\n\n\n# ==========================================================\n# 4. SEPARATE FEATURES AND TARGET\n# ==========================================================\n\nX = df.drop(\n    "outcome",\n    axis=1\n)\n\ny = df["outcome"]\n\n\n# ==========================================================\n# 5. CONVERT CATEGORICAL FEATURES TO NUMERICAL\n# ==========================================================\n\nX = pd.get_dummies(X)\n\n\n# ==========================================================\n# 6. ENCODE TARGET CLASSES\n# ==========================================================\n\nle = LabelEncoder()\n\ny = le.fit_transform(y)\n\nprint("\\nClass Labels:")\nprint(le.classes_)\n\n\n# ==========================================================\n# 7. TRAIN-TEST SPLIT\n# ==========================================================\n\nX_train, X_test, y_train, y_test = train_test_split(\n    X,\n    y,\n    test_size=0.2,\n    random_state=42,\n    stratify=y\n)\n\nprint("\\nTraining Samples:", X_train.shape[0])\nprint("Testing Samples:", X_test.shape[0])\n\n\n# ==========================================================\n# 8. LOGISTIC REGRESSION BEFORE BALANCING\n# ==========================================================\n\nmodel_before = LogisticRegression(\n    max_iter=1000,\n    solver="liblinear"\n)\n\n# Train on original imbalanced data\nmodel_before.fit(\n    X_train,\n    y_train\n)\n\n# Prediction\ny_pred_before = model_before.predict(\n    X_test\n)\n\n\n# ==========================================================\n# 9. ACCURACY BEFORE BALANCING\n# ==========================================================\n\naccuracy_before = accuracy_score(\n    y_test,\n    y_pred_before\n)\n\nprint("\\n==========================================")\nprint("BEFORE BALANCING")\nprint("==========================================")\n\nprint("\\nAccuracy:")\nprint(accuracy_before)\n\n\n# ==========================================================\n# 10. CLASSIFICATION REPORT BEFORE BALANCING\n# ==========================================================\n\nprint("\\nClassification Report:")\n\nprint(\n    classification_report(\n        y_test,\n        y_pred_before,\n        labels=np.unique(y_test),\n        target_names=le.inverse_transform(\n            np.unique(y_test)\n        ),\n        zero_division=0\n    )\n)\n\n\n# ==========================================================\n# 11. CONFUSION MATRIX BEFORE BALANCING\n# ==========================================================\n\ncm_before = confusion_matrix(\n    y_test,\n    y_pred_before\n)\n\nplt.figure(\n    figsize=(12, 10)\n)\n\nsns.heatmap(\n    cm_before,\n    annot=True,\n    fmt="d",\n    cmap="Blues",\n    xticklabels=le.classes_,\n    yticklabels=le.classes_\n)\n\nplt.xlabel(\n    "Predicted Label"\n)\n\nplt.ylabel(\n    "True Label"\n)\n\nplt.title(\n    "Confusion Matrix - BEFORE Balancing"\n)\n\nplt.xticks(rotation=90)\nplt.yticks(rotation=0)\n\nplt.tight_layout()\n\nplt.show()\n\n\n# ==========================================================\n# 12. CREATE TRAINING DATAFRAME\n# ==========================================================\n\ntrain_data = pd.DataFrame(\n    X_train\n)\n\ntrain_data["target"] = y_train\n\n\n# ==========================================================\n# 13. BALANCE THE TRAINING DATA\n# ==========================================================\n\nclasses = train_data[\n    "target"\n].unique()\n\nbalanced_data = []\n\n# Maximum number of samples per class\nmax_samples = 5000\n\nfor c in classes:\n\n    class_data = train_data[\n        train_data["target"] == c\n    ]\n\n    # ----------------------------------------------\n    # Oversampling minority classes\n    # ----------------------------------------------\n\n    if len(class_data) < max_samples:\n\n        class_data = resample(\n            class_data,\n            replace=True,\n            n_samples=max_samples,\n            random_state=42\n        )\n\n    # ----------------------------------------------\n    # Undersampling majority classes\n    # ----------------------------------------------\n\n    else:\n\n        class_data = resample(\n            class_data,\n            replace=False,\n            n_samples=max_samples,\n            random_state=42\n        )\n\n    balanced_data.append(\n        class_data\n    )\n\n\n# ==========================================================\n# 14. COMBINE BALANCED CLASSES\n# ==========================================================\n\nbalanced_data = pd.concat(\n    balanced_data\n)\n\nprint("\\n==========================================")\nprint("BALANCED CLASS DISTRIBUTION")\nprint("==========================================")\n\nprint(\n    balanced_data["target"].value_counts()\n)\n\n\n# ==========================================================\n# 15. SEPARATE BALANCED FEATURES AND TARGET\n# ==========================================================\n\nX_train_balanced = balanced_data.drop(\n    "target",\n    axis=1\n)\n\ny_train_balanced = balanced_data[\n    "target"\n]\n\n\n# ==========================================================\n# 16. LOGISTIC REGRESSION AFTER BALANCING\n# ==========================================================\n\nmodel_after = LogisticRegression(\n    max_iter=1000,\n    solver="liblinear"\n)\n\n# Train using balanced data\nmodel_after.fit(\n    X_train_balanced,\n    y_train_balanced\n)\n\n\n# ==========================================================\n# 17. PREDICTION AFTER BALANCING\n# ==========================================================\n\ny_pred_after = model_after.predict(\n    X_test\n)\n\n\n# ==========================================================\n# 18. ACCURACY AFTER BALANCING\n# ==========================================================\n\naccuracy_after = accuracy_score(\n    y_test,\n    y_pred_after\n)\n\nprint("\\n==========================================")\nprint("AFTER BALANCING")\nprint("==========================================")\n\nprint("\\nAccuracy:")\nprint(accuracy_after)\n\n\n# ==========================================================\n# 19. CLASSIFICATION REPORT AFTER BALANCING\n# ==========================================================\n\nprint("\\nClassification Report:")\n\nprint(\n    classification_report(\n        y_test,\n        y_pred_after,\n        labels=np.unique(y_test),\n        target_names=le.inverse_transform(\n            np.unique(y_test)\n        ),\n        zero_division=0\n    )\n)\n\n\n# ==========================================================\n# 20. CONFUSION MATRIX AFTER BALANCING\n# ==========================================================\n\ncm_after = confusion_matrix(\n    y_test,\n    y_pred_after\n)\n\nplt.figure(\n    figsize=(12, 10)\n)\n\nsns.heatmap(\n    cm_after,\n    annot=True,\n    fmt="d",\n    cmap="Blues",\n    xticklabels=le.classes_,\n    yticklabels=le.classes_\n)\n\nplt.xlabel(\n    "Predicted Label"\n)\n\nplt.ylabel(\n    "True Label"\n)\n\nplt.title(\n    "Confusion Matrix - AFTER Balancing"\n)\n\nplt.xticks(rotation=90)\nplt.yticks(rotation=0)\n\nplt.tight_layout()\n\nplt.show()\n\n\n# ==========================================================\n# 21. COMPARE ACCURACY\n# ==========================================================\n\nprint("\\n==========================================")\nprint("ACCURACY COMPARISON")\nprint("==========================================")\n\nprint(\n    "Accuracy Before Balancing:",\n    accuracy_before\n)\n\nprint(\n    "Accuracy After Balancing :",\n    accuracy_after\n)'
+
+print(SOURCE)

codemint-0.1.0.dist-info/METADATA

@@ -0,0 +1,6 @@
+Metadata-Version: 2.4
+Name: CodeMint
+Version: 0.1.0
+Summary: Eight experiment source-code modules for offline display
+Requires-Python: >=3.8
+Dynamic: requires-python

codemint-0.1.0.dist-info/RECORD

@@ -0,0 +1,13 @@
+CodeMint/__init__.py,sha256=TMsiW2ynKD_MCMvWN8U8fsEjc5KMvB7Fox-y5eOHQK8,90
+CodeMint/code1.py,sha256=LiXy3VUrvMz6PuXVUk8KlShloblkpxMOnPLTaTxkjQc,4495
+CodeMint/code2.py,sha256=OlB4XTLgaenoNnFX2cz-mYxkx8m1187RZsgf8ivme_k,10426
+CodeMint/code3.py,sha256=xSQG2aVwCLfqGr_62QB_tNgClPQ0oFrEijr4a7NXZd0,11862
+CodeMint/code4.py,sha256=t3zwnWEJC5PVlI6Sj1oZHw3-3544XiNUcq6f0KszC0Q,3862
+CodeMint/code5.py,sha256=aG1uEAu8OTTdzlV83g9ZJuGWIC4ijfpV93X0apIKv00,11299
+CodeMint/code6.py,sha256=dJPEmRQmlgBfS5SQxESINqMHTQ4habhotESG_7MMiUM,6388
+CodeMint/code7.py,sha256=TVrY87K1XpCbFFP5zGjwbJA6ScH0eAY9UInLxFEQ928,3271
+CodeMint/code8.py,sha256=5i6Pi6iQ9z8iqVn-azeXNGQnc6qwXZNOYjfiyi1e6fs,10124
+codemint-0.1.0.dist-info/METADATA,sha256=cMV4n16vtwRas75lEVINnvZszl3XqMWlJ0xD3C6KdhQ,166
+codemint-0.1.0.dist-info/WHEEL,sha256=aeYiig01lYGDzBgS8HxWXOg3uV61G9ijOsup-k9o1sk,91
+codemint-0.1.0.dist-info/top_level.txt,sha256=uOkiou1HFOfwHARXGTnxMdhP_OdOStQYaxkUtUihZ7Q,9
+codemint-0.1.0.dist-info/RECORD,,

codemint-0.1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (82.0.1)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

codemint-0.1.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+CodeMint