analysistoolbox 2.3.0__tar.gz → 2.4.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (137) hide show
  1. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/PKG-INFO +33 -1
  2. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/README.md +32 -0
  3. analysistoolbox-2.4.0/analysistoolbox/prescriptive_analytics/ConductLinearOptimization.py +291 -0
  4. analysistoolbox-2.4.0/analysistoolbox/prescriptive_analytics/__init__.py +2 -0
  5. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/CreateCorrelatedSIPs.py +7 -3
  6. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox.egg-info/PKG-INFO +33 -1
  7. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox.egg-info/SOURCES.txt +1 -0
  8. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/setup.py +1 -1
  9. analysistoolbox-2.3.0/analysistoolbox/prescriptive_analytics/__init__.py +0 -1
  10. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/LICENSE.md +0 -0
  11. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/__init__.py +0 -0
  12. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/calculus/FindDerivative.py +0 -0
  13. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/calculus/FindLimitOfFunction.py +0 -0
  14. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/calculus/FindMinimumSquareLoss.py +0 -0
  15. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/calculus/PlotFunction.py +0 -0
  16. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/calculus/__init__.py +0 -0
  17. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_collection/ExtractTextFromPDF.py +0 -0
  18. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_collection/FetchPDFFromURL.py +0 -0
  19. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_collection/FetchUSShapefile.py +0 -0
  20. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_collection/FetchWebsiteText.py +0 -0
  21. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_collection/GetCompanyFilings.py +0 -0
  22. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_collection/GetGoogleSearchResults.py +0 -0
  23. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_collection/GetZipFile.py +0 -0
  24. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_collection/__init__.py +0 -0
  25. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/AddDateNumberColumns.py +0 -0
  26. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/AddLeadingZeros.py +0 -0
  27. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/AddRowCountColumn.py +0 -0
  28. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/AddTPeriodColumn.py +0 -0
  29. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/AddTukeyOutlierColumn.py +0 -0
  30. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/CleanTextColumns.py +0 -0
  31. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/ConductAnomalyDetection.py +0 -0
  32. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/ConductEntityMatching.py +0 -0
  33. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/ConvertOddsToProbability.py +0 -0
  34. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/CountMissingDataByGroup.py +0 -0
  35. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/CreateBinnedColumn.py +0 -0
  36. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/CreateDataOverview.py +0 -0
  37. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/CreateRandomSampleGroups.py +0 -0
  38. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/CreateRareCategoryColumn.py +0 -0
  39. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/CreateStratifiedRandomSampleGroups.py +0 -0
  40. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/GeocodeUSAddresses.py +0 -0
  41. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/ImputeMissingValuesUsingNearestNeighbors.py +0 -0
  42. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/VerifyGranularity.py +0 -0
  43. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/data_processing/__init__.py +0 -0
  44. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/descriptive_analytics/ConductManifoldLearning.py +0 -0
  45. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/descriptive_analytics/ConductPrincipalComponentAnalysis.py +0 -0
  46. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/descriptive_analytics/ConductPropensityScoreMatching.py +0 -0
  47. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/descriptive_analytics/CreateAssociationRules.py +0 -0
  48. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/descriptive_analytics/CreateGaussianMixtureClusters.py +0 -0
  49. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/descriptive_analytics/CreateHierarchicalClusters.py +0 -0
  50. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/descriptive_analytics/CreateKMeansClusters.py +0 -0
  51. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/descriptive_analytics/GenerateEDAWithLIDA.py +0 -0
  52. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/descriptive_analytics/__init__.py +0 -0
  53. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/file_management/ConvertWordDocsToPDF.py +0 -0
  54. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/file_management/ConvertWordToHTML.py +0 -0
  55. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/file_management/CreateCopyOfPDF.py +0 -0
  56. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/file_management/CreateFileTree.py +0 -0
  57. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/file_management/ImportDataFromFolder.py +0 -0
  58. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/file_management/__init__.py +0 -0
  59. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/ChiSquareTestOfIndependence.py +0 -0
  60. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/ChiSquareTestOfIndependenceFromTable.py +0 -0
  61. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/ConductCoxProportionalHazardRegression.py +0 -0
  62. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/ConductLinearRegressionAnalysis.py +0 -0
  63. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/ConductLogisticRegressionAnalysis.py +0 -0
  64. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/ConductSurvivalAnalysis.py +0 -0
  65. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/OneSampleTTest.py +0 -0
  66. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/OneWayANOVA.py +0 -0
  67. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/TTestOfMeanFromStats.py +0 -0
  68. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/TTestOfProportionFromStats.py +0 -0
  69. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/TTestOfTwoMeansFromStats.py +0 -0
  70. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/TwoSampleTTestOfIndependence.py +0 -0
  71. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/TwoSampleTTestPaired.py +0 -0
  72. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/hypothesis_testing/__init__.py +0 -0
  73. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/linear_algebra/CalculateEigenvalues.py +0 -0
  74. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/linear_algebra/ConvertMatrixToRowEchelonForm.py +0 -0
  75. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/linear_algebra/ConvertSystemOfEquationsToMatrix.py +0 -0
  76. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/linear_algebra/PlotVectors.py +0 -0
  77. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/linear_algebra/SolveSystemOfEquations.py +0 -0
  78. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/linear_algebra/VisualizeMatrixAsLinearTransformation.py +0 -0
  79. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/linear_algebra/__init__.py +0 -0
  80. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/llm/SendPromptToAnthropic.py +0 -0
  81. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/llm/SendPromptToChatGPT.py +0 -0
  82. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/llm/__init__.py +0 -0
  83. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/predictive_analytics/CreateARIMAModel.py +0 -0
  84. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/predictive_analytics/CreateBoostedTreeModel.py +0 -0
  85. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/predictive_analytics/CreateDecisionTreeModel.py +0 -0
  86. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/predictive_analytics/CreateLinearRegressionModel.py +0 -0
  87. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/predictive_analytics/CreateLogisticRegressionModel.py +0 -0
  88. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/predictive_analytics/CreateNeuralNetwork_SingleOutcome.py +0 -0
  89. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/predictive_analytics/__init__.py +0 -0
  90. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/prescriptive_analytics/CreateContentBasedRecommender.py +0 -0
  91. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/probability/ProbabilityOfAtLeastOne.py +0 -0
  92. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/probability/ProbabilityOfHypothesisGivenData.py +0 -0
  93. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/probability/__init__.py +0 -0
  94. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/CreateMetalogDistribution.py +0 -0
  95. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/CreateMetalogDistributionFromPercentiles.py +0 -0
  96. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/CreateSIPDataframe.py +0 -0
  97. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/CreateSLURPDistribution.py +0 -0
  98. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/SimulateCountOfSuccesses.py +0 -0
  99. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/SimulateCountOutcome.py +0 -0
  100. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/SimulateCountUntilFirstSuccess.py +0 -0
  101. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/SimulateNormallyDistributedOutcome.py +0 -0
  102. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/SimulateTDistributedOutcome.py +0 -0
  103. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/SimulateTimeBetweenEvents.py +0 -0
  104. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/SimulateTimeUntilNEvents.py +0 -0
  105. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/__init__.py +0 -0
  106. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/pymetalog/__init__.py +0 -0
  107. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/pymetalog/a_vector.py +0 -0
  108. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/pymetalog/class_method.py +0 -0
  109. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/pymetalog/metalog.py +0 -0
  110. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/pymetalog/pdf_quantile_functions.py +0 -0
  111. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/simulations/pymetalog/support.py +0 -0
  112. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/statistics/CalculateConfidenceIntervalOfMean.py +0 -0
  113. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/statistics/CalculateConfidenceIntervalOfProportion.py +0 -0
  114. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/statistics/__init__.py +0 -0
  115. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/Plot100PercentStackedBarChart.py +0 -0
  116. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotBarChart.py +0 -0
  117. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotBoxWhiskerByGroup.py +0 -0
  118. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotBulletChart.py +0 -0
  119. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotCard.py +0 -0
  120. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotClusteredBarChart.py +0 -0
  121. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotContingencyHeatmap.py +0 -0
  122. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotCorrelationMatrix.py +0 -0
  123. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotDensityByGroup.py +0 -0
  124. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotDotPlot.py +0 -0
  125. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotHeatmap.py +0 -0
  126. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotOverlappingAreaChart.py +0 -0
  127. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotRiskTolerance.py +0 -0
  128. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotScatterplot.py +0 -0
  129. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotSingleVariableCountPlot.py +0 -0
  130. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotSingleVariableHistogram.py +0 -0
  131. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/PlotTimeSeries.py +0 -0
  132. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/RenderTableOne.py +0 -0
  133. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox/visualizations/__init__.py +0 -0
  134. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox.egg-info/dependency_links.txt +0 -0
  135. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox.egg-info/requires.txt +0 -0
  136. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/analysistoolbox.egg-info/top_level.txt +0 -0
  137. {analysistoolbox-2.3.0 → analysistoolbox-2.4.0}/setup.cfg +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.1
2
2
  Name: analysistoolbox
3
- Version: 2.3.0
3
+ Version: 2.4.0
4
4
  Summary: A collection tools in Python for data collection and processing, statistics, analytics, and intelligence analysis.
5
5
  Home-page: https://github.com/KyleProtho/AnalysisToolBox/tree/master/analysistoolbox
6
6
  Author: Kyle Protho
@@ -124,6 +124,7 @@ There are many modules in the analysistoolbox package, each with their own funct
124
124
  - [CreateLogisticRegressionModel](#createlogisticregressionmodel)
125
125
  - [CreateNeuralNetwork_SingleOutcome](#createneuralnetwork_singleoutcome)
126
126
  - [Prescriptive Analytics](#prescriptive-analytics)
127
+ - [ConductLinearOptimization](#conductlinearoptimization)
127
128
  - [CreateContentBasedRecommender](#createcontentbasedrecommender)
128
129
  - [Probability](#probability)
129
130
  - [ProbabilityOfAtLeastOne](#probabilityofatleastone)
@@ -1655,6 +1656,37 @@ model = CreateNeuralNetwork_SingleOutcome(
1655
1656
 
1656
1657
  The prescriptive analytics module provides tools for making data-driven recommendations and decisions:
1657
1658
 
1659
+ #### ConductLinearOptimization
1660
+
1661
+ Conducts linear optimization to find the optimal input values for a given output variable, with optional constraints.
1662
+
1663
+ ```python
1664
+ import pandas as pd
1665
+ from analysistoolbox.prescriptive_analytics.ConductLinearOptimization import ConductLinearOptimization
1666
+
1667
+ # Sample data
1668
+ data = pd.DataFrame({
1669
+ 'input1': [1, 2, 3, 4, 5],
1670
+ 'input2': [2, 4, 6, 8, 10],
1671
+ 'output': [10, 20, 30, 40, 50]
1672
+ })
1673
+
1674
+ # Define constraints (optional)
1675
+ constraints = {
1676
+ 'input1': (0, 10), # input1 between 0 and 10
1677
+ 'input2': (None, 15) # input2 maximum 15, no minimum
1678
+ }
1679
+
1680
+ # Run optimization
1681
+ results = ConductLinearOptimization(
1682
+ dataframe=data,
1683
+ output_variable='output',
1684
+ list_of_input_variables=['input1', 'input2'],
1685
+ optimization_type='maximize',
1686
+ input_constraints=constraints
1687
+ )
1688
+ ```
1689
+
1658
1690
  #### CreateContentBasedRecommender
1659
1691
 
1660
1692
  Builds a content-based recommendation system using neural networks to learn user and item embeddings.
@@ -106,6 +106,7 @@ There are many modules in the analysistoolbox package, each with their own funct
106
106
  - [CreateLogisticRegressionModel](#createlogisticregressionmodel)
107
107
  - [CreateNeuralNetwork_SingleOutcome](#createneuralnetwork_singleoutcome)
108
108
  - [Prescriptive Analytics](#prescriptive-analytics)
109
+ - [ConductLinearOptimization](#conductlinearoptimization)
109
110
  - [CreateContentBasedRecommender](#createcontentbasedrecommender)
110
111
  - [Probability](#probability)
111
112
  - [ProbabilityOfAtLeastOne](#probabilityofatleastone)
@@ -1637,6 +1638,37 @@ model = CreateNeuralNetwork_SingleOutcome(
1637
1638
 
1638
1639
  The prescriptive analytics module provides tools for making data-driven recommendations and decisions:
1639
1640
 
1641
+ #### ConductLinearOptimization
1642
+
1643
+ Conducts linear optimization to find the optimal input values for a given output variable, with optional constraints.
1644
+
1645
+ ```python
1646
+ import pandas as pd
1647
+ from analysistoolbox.prescriptive_analytics.ConductLinearOptimization import ConductLinearOptimization
1648
+
1649
+ # Sample data
1650
+ data = pd.DataFrame({
1651
+ 'input1': [1, 2, 3, 4, 5],
1652
+ 'input2': [2, 4, 6, 8, 10],
1653
+ 'output': [10, 20, 30, 40, 50]
1654
+ })
1655
+
1656
+ # Define constraints (optional)
1657
+ constraints = {
1658
+ 'input1': (0, 10), # input1 between 0 and 10
1659
+ 'input2': (None, 15) # input2 maximum 15, no minimum
1660
+ }
1661
+
1662
+ # Run optimization
1663
+ results = ConductLinearOptimization(
1664
+ dataframe=data,
1665
+ output_variable='output',
1666
+ list_of_input_variables=['input1', 'input2'],
1667
+ optimization_type='maximize',
1668
+ input_constraints=constraints
1669
+ )
1670
+ ```
1671
+
1640
1672
  #### CreateContentBasedRecommender
1641
1673
 
1642
1674
  Builds a content-based recommendation system using neural networks to learn user and item embeddings.
@@ -0,0 +1,291 @@
1
+ # Load packages
2
+ from matplotlib import pyplot as plt
3
+ import numpy as np
4
+ import pandas as pd
5
+ import seaborn as sns
6
+ from scipy.optimize import linprog
7
+ import textwrap
8
+ import warnings
9
+
10
+ # Declare function
11
+ def ConductLinearOptimization(dataframe,
12
+ output_variable,
13
+ list_of_input_variables,
14
+ optimization_type="maximize", # "maximize" or "minimize"
15
+ input_constraints=None, # Dictionary with variable names as keys and (min, max) tuples as values
16
+ # Output arguments
17
+ print_optimization_summary=True,
18
+ print_constraint_summary=True,
19
+ # All plot arguments
20
+ data_source_for_plot=None,
21
+ # Optimization results plot arguments
22
+ plot_optimization_results=True,
23
+ dot_fill_color="#999999",
24
+ line_color="#b0170c",
25
+ figure_size_for_optimization_plot=(10, 6),
26
+ title_for_optimization_plot="Linear Optimization Results",
27
+ subtitle_for_optimization_plot="Optimal input values and resulting output value.",
28
+ caption_for_optimization_plot=None):
29
+ """
30
+ Conduct linear optimization to find optimal input values that maximize or minimize an output variable.
31
+
32
+ This function uses linear programming to find the optimal combination of input variables
33
+ that will maximize or minimize the output variable, subject to optional constraints on input values.
34
+
35
+ Parameters:
36
+ -----------
37
+ dataframe : pandas.DataFrame
38
+ DataFrame containing input and output variables
39
+ output_variable : str
40
+ Name of the output variable to optimize
41
+ list_of_input_variables : list
42
+ List of input variable names to optimize
43
+ optimization_type : str, optional
44
+ Type of optimization: "maximize" or "minimize" (default: "maximize")
45
+ input_constraints : dict, optional
46
+ Dictionary with input variable names as keys and (min, max) tuples as values
47
+ Example: {"variable1": (0, 10), "variable2": (None, 5)} where None means no constraint
48
+ print_optimization_summary : bool, optional
49
+ Whether to print optimization results summary (default: True)
50
+ print_constraint_summary : bool, optional
51
+ Whether to print constraint information (default: True)
52
+ plot_optimization_results : bool, optional
53
+ Whether to plot optimization results (default: True)
54
+
55
+ Returns:
56
+ --------
57
+ dict
58
+ Dictionary containing optimization results including optimal values, objective value, and status
59
+ """
60
+
61
+ # Keep only the inputs and output variable
62
+ dataframe = dataframe[list_of_input_variables + [output_variable]].copy()
63
+
64
+ # Replace inf with nan, and drop rows with nan
65
+ dataframe.replace([np.inf, -np.inf], np.nan, inplace=True)
66
+ dataframe.dropna(inplace=True)
67
+
68
+ if len(dataframe) < len(list_of_input_variables) + 1:
69
+ raise ValueError(f"Insufficient data points ({len(dataframe)}) for {len(list_of_input_variables)} variables")
70
+
71
+ # Fit linear regression to get coefficients for the objective function
72
+ from sklearn.linear_model import LinearRegression
73
+
74
+ X = dataframe[list_of_input_variables]
75
+ y = dataframe[output_variable]
76
+
77
+ model = LinearRegression()
78
+ model.fit(X, y)
79
+
80
+ # Get coefficients and intercept
81
+ coefficients = model.coef_
82
+ intercept = model.intercept_
83
+
84
+ # Prepare objective function coefficients
85
+ if optimization_type.lower() == "maximize":
86
+ # For maximization, we minimize the negative of the objective
87
+ objective_coefficients = -coefficients
88
+ else:
89
+ # For minimization, we minimize the objective directly
90
+ objective_coefficients = coefficients
91
+
92
+ # Prepare bounds for input variables
93
+ bounds = []
94
+ for var in list_of_input_variables:
95
+ if input_constraints and var in input_constraints:
96
+ min_val, max_val = input_constraints[var]
97
+ bounds.append((
98
+ min_val if min_val is not None else -np.inf,
99
+ max_val if max_val is not None else np.inf
100
+ ))
101
+ else:
102
+ # Use data range as bounds if no constraints specified
103
+ var_min = dataframe[var].min()
104
+ var_max = dataframe[var].max()
105
+ bounds.append((var_min, var_max))
106
+
107
+ # Solve the linear programming problem
108
+ try:
109
+ result = linprog(
110
+ c=objective_coefficients,
111
+ bounds=bounds,
112
+ method='highs' # Use the Highs solver for better performance
113
+ )
114
+
115
+ if result.success:
116
+ optimal_inputs = result.x
117
+ if optimization_type.lower() == "maximize":
118
+ optimal_output = -result.fun + intercept
119
+ else:
120
+ optimal_output = result.fun + intercept
121
+ else:
122
+ warnings.warn(f"Optimization failed: {result.message}")
123
+ optimal_inputs = None
124
+ optimal_output = None
125
+
126
+ except Exception as e:
127
+ warnings.warn(f"Optimization error: {str(e)}")
128
+ optimal_inputs = None
129
+ optimal_output = None
130
+
131
+ # Print optimization summary if requested
132
+ if print_optimization_summary:
133
+ print(f"\n{'='*60}")
134
+ print(f"LINEAR OPTIMIZATION RESULTS")
135
+ print(f"{'='*60}")
136
+ print(f"Optimization Type: {optimization_type.upper()}")
137
+ print(f"Output Variable: {output_variable}")
138
+ print(f"Input Variables: {', '.join(list_of_input_variables)}")
139
+ print(f"\nModel Coefficients:")
140
+ for i, var in enumerate(list_of_input_variables):
141
+ print(f" {var}: {coefficients[i]:.6f}")
142
+ print(f"Intercept: {intercept:.6f}")
143
+
144
+ if optimal_inputs is not None:
145
+ print(f"\nOptimal Solution:")
146
+ for i, var in enumerate(list_of_input_variables):
147
+ print(f" {var}: {optimal_inputs[i]:.6f}")
148
+ print(f"Optimal {output_variable}: {optimal_output:.6f}")
149
+ else:
150
+ print(f"\nNo optimal solution found.")
151
+
152
+ # Print constraint summary if requested
153
+ if print_constraint_summary and input_constraints:
154
+ print(f"\n{'='*40}")
155
+ print(f"CONSTRAINT SUMMARY")
156
+ print(f"{'='*40}")
157
+ for var, (min_val, max_val) in input_constraints.items():
158
+ min_str = f"{min_val:.6f}" if min_val is not None else "-∞"
159
+ max_str = f"{max_val:.6f}" if max_val is not None else "∞"
160
+ print(f"{var}: [{min_str}, {max_str}]")
161
+
162
+ # Plot optimization results if requested
163
+ if plot_optimization_results and optimal_inputs is not None:
164
+ # Set the size of the plot
165
+ plt.figure(figsize=figure_size_for_optimization_plot)
166
+
167
+ # Create subplot for input variables
168
+ plt.subplot(1, 2, 1)
169
+
170
+ # Create bar plot of optimal input values
171
+ ax1 = sns.barplot(
172
+ x=list_of_input_variables,
173
+ y=optimal_inputs,
174
+ color=dot_fill_color,
175
+ alpha=0.7
176
+ )
177
+
178
+ # Add value labels on bars
179
+ for i, v in enumerate(optimal_inputs):
180
+ ax1.text(i, v, f'{v:.3f}', ha='center', va='bottom')
181
+
182
+ plt.title('Optimal Input Values', fontsize=12, color="#262626")
183
+ plt.ylabel('Value', fontsize=10, color="#666666")
184
+ plt.xticks(rotation=45, ha='right')
185
+
186
+ # Remove spines
187
+ ax1.spines['top'].set_visible(False)
188
+ ax1.spines['right'].set_visible(False)
189
+ ax1.spines['bottom'].set_color('#666666')
190
+ ax1.spines['left'].set_color('#666666')
191
+
192
+ # Create subplot for output variable
193
+ plt.subplot(1, 2, 2)
194
+
195
+ # Create bar plot of optimal output value
196
+ ax2 = sns.barplot(
197
+ x=[output_variable],
198
+ y=[optimal_output],
199
+ color=line_color,
200
+ alpha=0.7
201
+ )
202
+
203
+ # Add value label on bar
204
+ ax2.text(0, optimal_output, f'{optimal_output:.3f}', ha='center', va='bottom')
205
+
206
+ plt.title(f'Optimal {output_variable}', fontsize=12, color="#262626")
207
+ plt.ylabel('Value', fontsize=10, color="#666666")
208
+
209
+ # Remove spines
210
+ ax2.spines['top'].set_visible(False)
211
+ ax2.spines['right'].set_visible(False)
212
+ ax2.spines['bottom'].set_color('#666666')
213
+ ax2.spines['left'].set_color('#666666')
214
+
215
+ # Add main title
216
+ plt.suptitle(
217
+ title_for_optimization_plot,
218
+ fontsize=14,
219
+ color="#262626",
220
+ y=0.95
221
+ )
222
+
223
+ # Add subtitle
224
+ plt.figtext(
225
+ 0.5, 0.90,
226
+ subtitle_for_optimization_plot,
227
+ fontsize=11,
228
+ color="#666666",
229
+ ha='center'
230
+ )
231
+
232
+ # Add caption if provided
233
+ if caption_for_optimization_plot or data_source_for_plot:
234
+ caption_text = ""
235
+ if caption_for_optimization_plot:
236
+ caption_text = caption_for_optimization_plot
237
+ if data_source_for_plot:
238
+ caption_text += f"\n\nSource: {data_source_for_plot}"
239
+
240
+ plt.figtext(
241
+ 0.5, 0.02,
242
+ caption_text,
243
+ fontsize=8,
244
+ color="#666666",
245
+ ha='center'
246
+ )
247
+
248
+ plt.tight_layout()
249
+ plt.show()
250
+
251
+
252
+
253
+ # Return results
254
+ results = {
255
+ 'success': optimal_inputs is not None,
256
+ 'optimization_type': optimization_type,
257
+ 'output_variable': output_variable,
258
+ 'input_variables': list_of_input_variables,
259
+ 'coefficients': coefficients,
260
+ 'intercept': intercept,
261
+ 'optimal_inputs': optimal_inputs,
262
+ 'optimal_output': optimal_output,
263
+ 'constraints': input_constraints,
264
+ 'model': model
265
+ }
266
+
267
+ return results
268
+
269
+
270
+ # # Test the function
271
+ # # Sample data
272
+ # data = pd.DataFrame({
273
+ # 'input1': [1, 2, 3, 4, 5],
274
+ # 'input2': [2, 4, 6, 8, 10],
275
+ # 'output': [10, 20, 30, 40, 50]
276
+ # })
277
+
278
+ # # Define constraints (optional)
279
+ # constraints = {
280
+ # 'input1': (0, 10), # input1 between 0 and 10
281
+ # 'input2': (None, 15) # input2 maximum 15, no minimum
282
+ # }
283
+
284
+ # # Run optimization
285
+ # results = ConductLinearOptimization(
286
+ # dataframe=data,
287
+ # output_variable='output',
288
+ # list_of_input_variables=['input1', 'input2'],
289
+ # optimization_type='maximize',
290
+ # input_constraints=constraints
291
+ # )
@@ -0,0 +1,2 @@
1
+ from .CreateContentBasedRecommender import CreateContentBasedRecommender
2
+ from .ConductLinearOptimization import ConductLinearOptimization
@@ -14,6 +14,8 @@ def CreateCorrelatedSIPs(mean_of_variable_1,
14
14
  number_of_samples=10000,
15
15
  variable_1_name='Variable 1',
16
16
  variable_2_name='Variable 2',
17
+ # Printing arguments
18
+ print_simulation_result_summary=False,
17
19
  # Scatterplot arguments
18
20
  show_scatterplot=True,
19
21
  dot_fill_color="#999999",
@@ -79,12 +81,14 @@ def CreateCorrelatedSIPs(mean_of_variable_1,
79
81
 
80
82
  # Calculate the correlation between the two variables
81
83
  simulated_correlation = dataframe[variable_1_name].corr(dataframe[variable_2_name])
82
- print("Correlation between the two variables in the simulated data: ", simulated_correlation)
84
+ if print_simulation_result_summary:
85
+ print("Correlation between the two variables in the simulated data: ", simulated_correlation)
83
86
 
84
87
  # Show the summary statistics of the dataframe
85
88
  simulation_summary = dataframe.describe()
86
- print("Summary statistics of the simulated data:")
87
- print(simulation_summary)
89
+ if print_simulation_result_summary:
90
+ print("Summary statistics of the simulated data:")
91
+ print(simulation_summary)
88
92
 
89
93
  # If scatterplot requested, generate plot
90
94
  if show_scatterplot:
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.1
2
2
  Name: analysistoolbox
3
- Version: 2.3.0
3
+ Version: 2.4.0
4
4
  Summary: A collection tools in Python for data collection and processing, statistics, analytics, and intelligence analysis.
5
5
  Home-page: https://github.com/KyleProtho/AnalysisToolBox/tree/master/analysistoolbox
6
6
  Author: Kyle Protho
@@ -124,6 +124,7 @@ There are many modules in the analysistoolbox package, each with their own funct
124
124
  - [CreateLogisticRegressionModel](#createlogisticregressionmodel)
125
125
  - [CreateNeuralNetwork_SingleOutcome](#createneuralnetwork_singleoutcome)
126
126
  - [Prescriptive Analytics](#prescriptive-analytics)
127
+ - [ConductLinearOptimization](#conductlinearoptimization)
127
128
  - [CreateContentBasedRecommender](#createcontentbasedrecommender)
128
129
  - [Probability](#probability)
129
130
  - [ProbabilityOfAtLeastOne](#probabilityofatleastone)
@@ -1655,6 +1656,37 @@ model = CreateNeuralNetwork_SingleOutcome(
1655
1656
 
1656
1657
  The prescriptive analytics module provides tools for making data-driven recommendations and decisions:
1657
1658
 
1659
+ #### ConductLinearOptimization
1660
+
1661
+ Conducts linear optimization to find the optimal input values for a given output variable, with optional constraints.
1662
+
1663
+ ```python
1664
+ import pandas as pd
1665
+ from analysistoolbox.prescriptive_analytics.ConductLinearOptimization import ConductLinearOptimization
1666
+
1667
+ # Sample data
1668
+ data = pd.DataFrame({
1669
+ 'input1': [1, 2, 3, 4, 5],
1670
+ 'input2': [2, 4, 6, 8, 10],
1671
+ 'output': [10, 20, 30, 40, 50]
1672
+ })
1673
+
1674
+ # Define constraints (optional)
1675
+ constraints = {
1676
+ 'input1': (0, 10), # input1 between 0 and 10
1677
+ 'input2': (None, 15) # input2 maximum 15, no minimum
1678
+ }
1679
+
1680
+ # Run optimization
1681
+ results = ConductLinearOptimization(
1682
+ dataframe=data,
1683
+ output_variable='output',
1684
+ list_of_input_variables=['input1', 'input2'],
1685
+ optimization_type='maximize',
1686
+ input_constraints=constraints
1687
+ )
1688
+ ```
1689
+
1658
1690
  #### CreateContentBasedRecommender
1659
1691
 
1660
1692
  Builds a content-based recommendation system using neural networks to learn user and item embeddings.
@@ -85,6 +85,7 @@ analysistoolbox/predictive_analytics/CreateLinearRegressionModel.py
85
85
  analysistoolbox/predictive_analytics/CreateLogisticRegressionModel.py
86
86
  analysistoolbox/predictive_analytics/CreateNeuralNetwork_SingleOutcome.py
87
87
  analysistoolbox/predictive_analytics/__init__.py
88
+ analysistoolbox/prescriptive_analytics/ConductLinearOptimization.py
88
89
  analysistoolbox/prescriptive_analytics/CreateContentBasedRecommender.py
89
90
  analysistoolbox/prescriptive_analytics/__init__.py
90
91
  analysistoolbox/probability/ProbabilityOfAtLeastOne.py
@@ -12,7 +12,7 @@ setup(
12
12
  long_description=long_description,
13
13
  long_description_content_type='text/markdown',
14
14
  url="https://github.com/KyleProtho/AnalysisToolBox/tree/master/analysistoolbox",
15
- version='2.3.0',
15
+ version='2.4.0',
16
16
  author='Kyle Protho',
17
17
  author_email='kyletprotho@gmail.com',
18
18
  license='MIT',
@@ -1 +0,0 @@
1
- from .CreateContentBasedRecommender import CreateContentBasedRecommender