analysistoolbox 3.2.2__tar.gz → 3.9.9__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.
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/PKG-INFO +75 -4
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/README.md +74 -3
- analysistoolbox-3.9.9/analysistoolbox/calculus/FindDerivative.py +111 -0
- analysistoolbox-3.9.9/analysistoolbox/calculus/FindLimitOfFunction.py +123 -0
- analysistoolbox-3.9.9/analysistoolbox/calculus/FindMinimumSquareLoss.py +106 -0
- analysistoolbox-3.9.9/analysistoolbox/calculus/PlotFunction.py +227 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/data_collection/ExtractTextFromPDF.py +54 -7
- analysistoolbox-3.9.9/analysistoolbox/data_collection/FetchPDFFromURL.py +71 -0
- analysistoolbox-3.9.9/analysistoolbox/data_collection/FetchUSShapefile.py +128 -0
- analysistoolbox-3.9.9/analysistoolbox/data_collection/FetchWebsiteText.py +100 -0
- analysistoolbox-3.9.9/analysistoolbox/data_collection/GetCompanyFilings.py +180 -0
- analysistoolbox-3.9.9/analysistoolbox/data_collection/GetGoogleSearchResults.py +140 -0
- analysistoolbox-3.9.9/analysistoolbox/data_collection/GetZipFile.py +93 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/AddDateNumberColumns.py +98 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/AddLeadingZeros.py +112 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/AddRowCountColumn.py +142 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/AddTPeriodColumn.py +130 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/AddTukeyOutlierColumn.py +316 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/CleanTextColumns.py +89 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/ConductAnomalyDetection.py +197 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/data_processing/ConductEntityMatching.py +137 -14
- analysistoolbox-3.9.9/analysistoolbox/data_processing/ConvertOddsToProbability.py +92 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/CountMissingDataByGroup.py +98 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/CreateBinnedColumn.py +134 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/data_processing/CreateDataOverview.py +70 -7
- analysistoolbox-3.9.9/analysistoolbox/data_processing/CreateRandomSampleGroups.py +79 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/CreateRareCategoryColumn.py +112 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/CreateStratifiedRandomSampleGroups.py +105 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/GeocodeUSAddresses.py +102 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/ImputeMissingValuesUsingNearestNeighbors.py +113 -0
- analysistoolbox-3.9.9/analysistoolbox/data_processing/VerifyGranularity.py +121 -0
- analysistoolbox-3.9.9/analysistoolbox/descriptive_analytics/ConductManifoldLearning.py +161 -0
- analysistoolbox-3.9.9/analysistoolbox/descriptive_analytics/ConductPrincipalComponentAnalysis.py +173 -0
- analysistoolbox-3.9.9/analysistoolbox/descriptive_analytics/ConductPropensityScoreMatching.py +261 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/descriptive_analytics/CreateAssociationRules.py +149 -22
- analysistoolbox-3.9.9/analysistoolbox/descriptive_analytics/CreateGaussianMixtureClusters.py +223 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/descriptive_analytics/CreateHierarchicalClusters.py +136 -15
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/descriptive_analytics/CreateKMeansClusters.py +139 -14
- analysistoolbox-3.9.9/analysistoolbox/descriptive_analytics/GenerateEDAWithLIDA.py +255 -0
- analysistoolbox-3.9.9/analysistoolbox/file_management/ConvertWordDocsToPDF.py +132 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/file_management/ConvertWordToHTML.py +103 -6
- analysistoolbox-3.9.9/analysistoolbox/file_management/CreateCopyOfPDF.py +154 -0
- analysistoolbox-3.9.9/analysistoolbox/file_management/CreateFileTree.py +122 -0
- analysistoolbox-3.9.9/analysistoolbox/file_management/ImportDataFromFolder.py +158 -0
- analysistoolbox-3.9.9/analysistoolbox/geospatial_analysis/ConductClusterAnalysis.py +128 -0
- analysistoolbox-3.9.9/analysistoolbox/geospatial_analysis/ConductSpatialAutocorrelation.py +159 -0
- analysistoolbox-3.9.9/analysistoolbox/geospatial_analysis/__init__.py +1 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/ChiSquareTestOfIndependence.py +135 -22
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/ChiSquareTestOfIndependenceFromTable.py +123 -20
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/ConductCoxProportionalHazardRegression.py +133 -26
- analysistoolbox-3.9.9/analysistoolbox/hypothesis_testing/ConductLinearRegressionAnalysis.py +159 -0
- analysistoolbox-3.9.9/analysistoolbox/hypothesis_testing/ConductLogisticRegressionAnalysis.py +186 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/ConductSurvivalAnalysis.py +136 -29
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/OneSampleTTest.py +110 -24
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/OneWayANOVA.py +109 -21
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/TTestOfMeanFromStats.py +105 -26
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/TTestOfProportionFromStats.py +104 -22
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/TTestOfTwoMeansFromStats.py +108 -25
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/TwoSampleTTestOfIndependence.py +115 -22
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/TwoSampleTTestPaired.py +110 -22
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/linear_algebra/CalculateEigenvalues.py +66 -20
- analysistoolbox-3.9.9/analysistoolbox/linear_algebra/ConvertMatrixToRowEchelonForm.py +77 -0
- analysistoolbox-3.9.9/analysistoolbox/linear_algebra/ConvertSystemOfEquationsToMatrix.py +88 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/linear_algebra/PlotVectors.py +52 -9
- analysistoolbox-3.9.9/analysistoolbox/linear_algebra/SolveSystemOfEquations.py +108 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/linear_algebra/VisualizeMatrixAsLinearTransformation.py +57 -14
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/predictive_analytics/CreateARIMAModel.py +102 -29
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/predictive_analytics/CreateBoostedTreeModel.py +107 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/predictive_analytics/CreateDecisionTreeModel.py +125 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/predictive_analytics/CreateExponentialSmoothingModel.py +118 -51
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/predictive_analytics/CreateLinearRegressionModel.py +109 -0
- analysistoolbox-3.9.9/analysistoolbox/predictive_analytics/CreateLogisticRegressionModel.py +185 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/predictive_analytics/CreateNeuralNetwork_SingleOutcome.py +93 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/prescriptive_analytics/ConductLinearOptimization.py +100 -20
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/prescriptive_analytics/CreateContentBasedRecommender.py +109 -1
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/probability/ProbabilityOfAtLeastOne.py +91 -23
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/probability/ProbabilityOfHypothesisGivenData.py +70 -6
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/CreateCorrelatedSIPs.py +93 -21
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/CreateMetalogDistribution.py +113 -31
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/CreateMetalogDistributionFromPercentiles.py +107 -30
- analysistoolbox-3.9.9/analysistoolbox/simulations/CreateSIPDataframe.py +81 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/CreateSLURPDistributionFromExponentialSmoothing.py +110 -61
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/CreateSLURPDistributionFromLinearRegression.py +138 -28
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/CreateSLURPDistributionFromLogisticRegression.py +98 -60
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/SimulateCountOfSuccesses.py +85 -31
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/SimulateCountOutcome.py +81 -29
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/SimulateCountUntilFirstSuccess.py +83 -31
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/SimulateNormallyDistributedOutcome.py +88 -31
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/SimulateTDistributedOutcome.py +92 -32
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/SimulateTimeBetweenEvents.py +82 -27
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/SimulateTimeUntilNEvents.py +87 -29
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/pymetalog/support.py +5 -5
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/statistics/CalculateConfidenceIntervalOfMean.py +99 -1
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/statistics/CalculateConfidenceIntervalOfProportion.py +90 -1
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/Plot100PercentStackedBarChart.py +112 -23
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotBarChart.py +102 -24
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotBoxWhiskerByGroup.py +114 -21
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotBulletChart.py +137 -33
- analysistoolbox-3.9.9/analysistoolbox/visualizations/PlotCard.py +169 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotClusteredBarChart.py +109 -21
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotContingencyHeatmap.py +91 -17
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotCorrelationMatrix.py +104 -19
- analysistoolbox-3.9.9/analysistoolbox/visualizations/PlotDensityByGroup.py +221 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotDotPlot.py +136 -46
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotHeatmap.py +113 -31
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotOverlappingAreaChart.py +119 -24
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotRiskTolerance.py +102 -23
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotScatterplot.py +132 -29
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotSingleVariableCountPlot.py +116 -25
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotSingleVariableHistogram.py +92 -20
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/PlotTimeSeries.py +124 -23
- analysistoolbox-3.9.9/analysistoolbox/visualizations/RenderTableOne.py +145 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox.egg-info/PKG-INFO +75 -4
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox.egg-info/SOURCES.txt +3 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/setup.py +1 -1
- analysistoolbox-3.2.2/analysistoolbox/calculus/FindDerivative.py +0 -65
- analysistoolbox-3.2.2/analysistoolbox/calculus/FindLimitOfFunction.py +0 -79
- analysistoolbox-3.2.2/analysistoolbox/calculus/FindMinimumSquareLoss.py +0 -57
- analysistoolbox-3.2.2/analysistoolbox/calculus/PlotFunction.py +0 -165
- analysistoolbox-3.2.2/analysistoolbox/data_collection/FetchPDFFromURL.py +0 -29
- analysistoolbox-3.2.2/analysistoolbox/data_collection/FetchUSShapefile.py +0 -67
- analysistoolbox-3.2.2/analysistoolbox/data_collection/FetchWebsiteText.py +0 -54
- analysistoolbox-3.2.2/analysistoolbox/data_collection/GetCompanyFilings.py +0 -97
- analysistoolbox-3.2.2/analysistoolbox/data_collection/GetGoogleSearchResults.py +0 -66
- analysistoolbox-3.2.2/analysistoolbox/data_collection/GetZipFile.py +0 -36
- analysistoolbox-3.2.2/analysistoolbox/data_processing/AddDateNumberColumns.py +0 -41
- analysistoolbox-3.2.2/analysistoolbox/data_processing/AddLeadingZeros.py +0 -48
- analysistoolbox-3.2.2/analysistoolbox/data_processing/AddRowCountColumn.py +0 -48
- analysistoolbox-3.2.2/analysistoolbox/data_processing/AddTPeriodColumn.py +0 -48
- analysistoolbox-3.2.2/analysistoolbox/data_processing/AddTukeyOutlierColumn.py +0 -203
- analysistoolbox-3.2.2/analysistoolbox/data_processing/CleanTextColumns.py +0 -23
- analysistoolbox-3.2.2/analysistoolbox/data_processing/ConductAnomalyDetection.py +0 -97
- analysistoolbox-3.2.2/analysistoolbox/data_processing/ConvertOddsToProbability.py +0 -34
- analysistoolbox-3.2.2/analysistoolbox/data_processing/CountMissingDataByGroup.py +0 -44
- analysistoolbox-3.2.2/analysistoolbox/data_processing/CreateBinnedColumn.py +0 -65
- analysistoolbox-3.2.2/analysistoolbox/data_processing/CreateRandomSampleGroups.py +0 -32
- analysistoolbox-3.2.2/analysistoolbox/data_processing/CreateRareCategoryColumn.py +0 -54
- analysistoolbox-3.2.2/analysistoolbox/data_processing/CreateStratifiedRandomSampleGroups.py +0 -46
- analysistoolbox-3.2.2/analysistoolbox/data_processing/GeocodeUSAddresses.py +0 -43
- analysistoolbox-3.2.2/analysistoolbox/data_processing/ImputeMissingValuesUsingNearestNeighbors.py +0 -49
- analysistoolbox-3.2.2/analysistoolbox/data_processing/VerifyGranularity.py +0 -67
- analysistoolbox-3.2.2/analysistoolbox/descriptive_analytics/ConductManifoldLearning.py +0 -84
- analysistoolbox-3.2.2/analysistoolbox/descriptive_analytics/ConductPrincipalComponentAnalysis.py +0 -88
- analysistoolbox-3.2.2/analysistoolbox/descriptive_analytics/ConductPropensityScoreMatching.py +0 -158
- analysistoolbox-3.2.2/analysistoolbox/descriptive_analytics/CreateGaussianMixtureClusters.py +0 -117
- analysistoolbox-3.2.2/analysistoolbox/descriptive_analytics/GenerateEDAWithLIDA.py +0 -136
- analysistoolbox-3.2.2/analysistoolbox/file_management/ConvertWordDocsToPDF.py +0 -52
- analysistoolbox-3.2.2/analysistoolbox/file_management/CreateCopyOfPDF.py +0 -44
- analysistoolbox-3.2.2/analysistoolbox/file_management/CreateFileTree.py +0 -30
- analysistoolbox-3.2.2/analysistoolbox/file_management/ImportDataFromFolder.py +0 -54
- analysistoolbox-3.2.2/analysistoolbox/hypothesis_testing/ConductLinearRegressionAnalysis.py +0 -82
- analysistoolbox-3.2.2/analysistoolbox/hypothesis_testing/ConductLogisticRegressionAnalysis.py +0 -111
- analysistoolbox-3.2.2/analysistoolbox/linear_algebra/ConvertMatrixToRowEchelonForm.py +0 -39
- analysistoolbox-3.2.2/analysistoolbox/linear_algebra/ConvertSystemOfEquationsToMatrix.py +0 -41
- analysistoolbox-3.2.2/analysistoolbox/linear_algebra/SolveSystemOfEquations.py +0 -60
- analysistoolbox-3.2.2/analysistoolbox/predictive_analytics/CreateLogisticRegressionModel.py +0 -105
- analysistoolbox-3.2.2/analysistoolbox/simulations/CreateSIPDataframe.py +0 -39
- analysistoolbox-3.2.2/analysistoolbox/visualizations/PlotCard.py +0 -104
- analysistoolbox-3.2.2/analysistoolbox/visualizations/PlotDensityByGroup.py +0 -155
- analysistoolbox-3.2.2/analysistoolbox/visualizations/RenderTableOne.py +0 -72
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/LICENSE.md +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/calculus/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/data_collection/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/data_processing/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/descriptive_analytics/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/file_management/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/hypothesis_testing/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/linear_algebra/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/llm/SendDeepResearchQuery.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/llm/SendPromptToAnthropic.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/llm/SendPromptToChatGPT.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/llm/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/predictive_analytics/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/prescriptive_analytics/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/probability/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/pymetalog/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/pymetalog/a_vector.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/pymetalog/class_method.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/pymetalog/metalog.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/simulations/pymetalog/pdf_quantile_functions.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/statistics/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox/visualizations/__init__.py +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox.egg-info/dependency_links.txt +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox.egg-info/requires.txt +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/analysistoolbox.egg-info/top_level.txt +0 -0
- {analysistoolbox-3.2.2 → analysistoolbox-3.9.9}/setup.cfg +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.1
|
|
2
2
|
Name: analysistoolbox
|
|
3
|
-
Version: 3.
|
|
3
|
+
Version: 3.9.9
|
|
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
|
|
@@ -44,6 +44,12 @@ sns.set(
|
|
|
44
44
|
)
|
|
45
45
|
```
|
|
46
46
|
|
|
47
|
+
## Documentation Standards
|
|
48
|
+
|
|
49
|
+
To verify that the library remains accessible to students and professionals alike, all functions in the library should include a **Teaching Note** in their docstring. This note should explain *why* the function matters, how it aids in "sense-making," or what qualitative insights it reveals, rather than just describing the mechanics of the code.
|
|
50
|
+
|
|
51
|
+
See `analysistoolbox.calculus.PlotFunction` for a canonical example.
|
|
52
|
+
|
|
47
53
|
## Table of Contents / Usage
|
|
48
54
|
|
|
49
55
|
There are many modules in the analysistoolbox package, each with their own functions. The following is a list of the modules:
|
|
@@ -93,6 +99,9 @@ There are many modules in the analysistoolbox package, each with their own funct
|
|
|
93
99
|
- [CreateFileTree](#createfiletree)
|
|
94
100
|
- [CreateCopyOfPDF](#createcopyofpdf)
|
|
95
101
|
- [ConvertWordDocsToPDF](#convertworddocstopdf)
|
|
102
|
+
- [Geospatial Analysis](#geospatial-analysis)
|
|
103
|
+
- [ConductClusterAnalysis](#conductclusteranalysis)
|
|
104
|
+
- [ConductSpatialAutocorrelation](#conductspatialautocorrelation)
|
|
96
105
|
- [Hypothesis Testing](#hypothesis-testing)
|
|
97
106
|
- [ChiSquareTestOfIndependence](#chisquaretestofindependence)
|
|
98
107
|
- [ChiSquareTestOfIndependenceFromTable](#chisquaretestofindependencefromtable)
|
|
@@ -1135,6 +1144,68 @@ ConvertWordDocsToPDF(
|
|
|
1135
1144
|
)
|
|
1136
1145
|
```
|
|
1137
1146
|
|
|
1147
|
+
|
|
1148
|
+
### Geospatial Analysis
|
|
1149
|
+
|
|
1150
|
+
#### ConductClusterAnalysis
|
|
1151
|
+
|
|
1152
|
+
The **ConductClusterAnalysis** function groups geospatial data points into clusters based on their proximity using the DBSCAN algorithm. It mimics the functionality of Bellingcat's geoclustering tool and can optionally display a map of the clusters.
|
|
1153
|
+
|
|
1154
|
+
```python
|
|
1155
|
+
# Import the function
|
|
1156
|
+
from analysistoolbox.geospatial_analysis import ConductClusterAnalysis
|
|
1157
|
+
import pandas as pd
|
|
1158
|
+
|
|
1159
|
+
# Create a sample dataframe
|
|
1160
|
+
df = pd.DataFrame({
|
|
1161
|
+
'lat': [40.7128, 40.7130, 34.0522, 34.0525],
|
|
1162
|
+
'lon': [-74.0060, -74.0065, -118.2437, -118.2440]
|
|
1163
|
+
})
|
|
1164
|
+
|
|
1165
|
+
# Run cluster analysis
|
|
1166
|
+
summary = ConductClusterAnalysis(
|
|
1167
|
+
dataframe=df,
|
|
1168
|
+
longitude_column='lon',
|
|
1169
|
+
latitude_column='lat',
|
|
1170
|
+
distance_km=0.5,
|
|
1171
|
+
min_samples=2,
|
|
1172
|
+
map_clusters=True
|
|
1173
|
+
)
|
|
1174
|
+
|
|
1175
|
+
# Print summary
|
|
1176
|
+
print(summary)
|
|
1177
|
+
```
|
|
1178
|
+
|
|
1179
|
+
#### ConductSpatialAutocorrelation
|
|
1180
|
+
|
|
1181
|
+
The **ConductSpatialAutocorrelation** function identifies spatial clusters and outliers using Local Moran's I. Spatial autocorrelation measures how much a variable at a location is correlated with the same variable at neighboring locations, effectively identifying if similar values cluster together (hotspots/coldspots) or if dissimilar values are near each other (spatial outliers).
|
|
1182
|
+
|
|
1183
|
+
```python
|
|
1184
|
+
# Import the function
|
|
1185
|
+
from analysistoolbox.geospatial_analysis import ConductSpatialAutocorrelation
|
|
1186
|
+
import pandas as pd
|
|
1187
|
+
|
|
1188
|
+
# Create a sample dataframe
|
|
1189
|
+
df = pd.DataFrame({
|
|
1190
|
+
'lat': [40.7128, 40.7130, 34.0522, 34.0525, 34.0500],
|
|
1191
|
+
'lon': [-74.0060, -74.0065, -118.2437, -118.2440, -118.2500],
|
|
1192
|
+
'value': [100, 105, 10, 15, 8]
|
|
1193
|
+
})
|
|
1194
|
+
|
|
1195
|
+
# Run spatial autocorrelation analysis
|
|
1196
|
+
results = ConductSpatialAutocorrelation(
|
|
1197
|
+
dataframe=df,
|
|
1198
|
+
longitude_column='lon',
|
|
1199
|
+
latitude_column='lat',
|
|
1200
|
+
value_column='value',
|
|
1201
|
+
k=5,
|
|
1202
|
+
plot=True
|
|
1203
|
+
)
|
|
1204
|
+
|
|
1205
|
+
# Print results
|
|
1206
|
+
print(results[['lat', 'lon', 'value', 'cluster_category']])
|
|
1207
|
+
```
|
|
1208
|
+
|
|
1138
1209
|
### Hypothesis Testing
|
|
1139
1210
|
|
|
1140
1211
|
#### ChiSquareTestOfIndependence
|
|
@@ -1677,11 +1748,11 @@ model = CreateNeuralNetwork_SingleOutcome(
|
|
|
1677
1748
|
|
|
1678
1749
|
### Prescriptive Analytics
|
|
1679
1750
|
|
|
1680
|
-
The prescriptive analytics module
|
|
1751
|
+
The prescriptive analytics module empowers users to determine the best course of action through optimization and personalized recommendation systems, bridging the gap between insight and action.
|
|
1681
1752
|
|
|
1682
1753
|
#### ConductLinearOptimization
|
|
1683
1754
|
|
|
1684
|
-
|
|
1755
|
+
Identifies the optimal values for input variables to maximize or minimize a specific output, subject to logical constraints and data-driven linear relationships. This approach is essential for resource allocation, logistics optimization, and cost minimization.
|
|
1685
1756
|
|
|
1686
1757
|
```python
|
|
1687
1758
|
import pandas as pd
|
|
@@ -1712,7 +1783,7 @@ results = ConductLinearOptimization(
|
|
|
1712
1783
|
|
|
1713
1784
|
#### CreateContentBasedRecommender
|
|
1714
1785
|
|
|
1715
|
-
|
|
1786
|
+
Generates a customized recommendation engine using a "two-tower" neural network architecture to match users with highly relevant items or resources based on shared characteristics. This is particularly useful for personalized content delivery, such as matching patients to clinical trials or identifying relevant intelligence reports.
|
|
1716
1787
|
|
|
1717
1788
|
```python
|
|
1718
1789
|
from analysistoolbox.prescriptive_analytics import CreateContentBasedRecommender
|
|
@@ -26,6 +26,12 @@ sns.set(
|
|
|
26
26
|
)
|
|
27
27
|
```
|
|
28
28
|
|
|
29
|
+
## Documentation Standards
|
|
30
|
+
|
|
31
|
+
To verify that the library remains accessible to students and professionals alike, all functions in the library should include a **Teaching Note** in their docstring. This note should explain *why* the function matters, how it aids in "sense-making," or what qualitative insights it reveals, rather than just describing the mechanics of the code.
|
|
32
|
+
|
|
33
|
+
See `analysistoolbox.calculus.PlotFunction` for a canonical example.
|
|
34
|
+
|
|
29
35
|
## Table of Contents / Usage
|
|
30
36
|
|
|
31
37
|
There are many modules in the analysistoolbox package, each with their own functions. The following is a list of the modules:
|
|
@@ -75,6 +81,9 @@ There are many modules in the analysistoolbox package, each with their own funct
|
|
|
75
81
|
- [CreateFileTree](#createfiletree)
|
|
76
82
|
- [CreateCopyOfPDF](#createcopyofpdf)
|
|
77
83
|
- [ConvertWordDocsToPDF](#convertworddocstopdf)
|
|
84
|
+
- [Geospatial Analysis](#geospatial-analysis)
|
|
85
|
+
- [ConductClusterAnalysis](#conductclusteranalysis)
|
|
86
|
+
- [ConductSpatialAutocorrelation](#conductspatialautocorrelation)
|
|
78
87
|
- [Hypothesis Testing](#hypothesis-testing)
|
|
79
88
|
- [ChiSquareTestOfIndependence](#chisquaretestofindependence)
|
|
80
89
|
- [ChiSquareTestOfIndependenceFromTable](#chisquaretestofindependencefromtable)
|
|
@@ -1117,6 +1126,68 @@ ConvertWordDocsToPDF(
|
|
|
1117
1126
|
)
|
|
1118
1127
|
```
|
|
1119
1128
|
|
|
1129
|
+
|
|
1130
|
+
### Geospatial Analysis
|
|
1131
|
+
|
|
1132
|
+
#### ConductClusterAnalysis
|
|
1133
|
+
|
|
1134
|
+
The **ConductClusterAnalysis** function groups geospatial data points into clusters based on their proximity using the DBSCAN algorithm. It mimics the functionality of Bellingcat's geoclustering tool and can optionally display a map of the clusters.
|
|
1135
|
+
|
|
1136
|
+
```python
|
|
1137
|
+
# Import the function
|
|
1138
|
+
from analysistoolbox.geospatial_analysis import ConductClusterAnalysis
|
|
1139
|
+
import pandas as pd
|
|
1140
|
+
|
|
1141
|
+
# Create a sample dataframe
|
|
1142
|
+
df = pd.DataFrame({
|
|
1143
|
+
'lat': [40.7128, 40.7130, 34.0522, 34.0525],
|
|
1144
|
+
'lon': [-74.0060, -74.0065, -118.2437, -118.2440]
|
|
1145
|
+
})
|
|
1146
|
+
|
|
1147
|
+
# Run cluster analysis
|
|
1148
|
+
summary = ConductClusterAnalysis(
|
|
1149
|
+
dataframe=df,
|
|
1150
|
+
longitude_column='lon',
|
|
1151
|
+
latitude_column='lat',
|
|
1152
|
+
distance_km=0.5,
|
|
1153
|
+
min_samples=2,
|
|
1154
|
+
map_clusters=True
|
|
1155
|
+
)
|
|
1156
|
+
|
|
1157
|
+
# Print summary
|
|
1158
|
+
print(summary)
|
|
1159
|
+
```
|
|
1160
|
+
|
|
1161
|
+
#### ConductSpatialAutocorrelation
|
|
1162
|
+
|
|
1163
|
+
The **ConductSpatialAutocorrelation** function identifies spatial clusters and outliers using Local Moran's I. Spatial autocorrelation measures how much a variable at a location is correlated with the same variable at neighboring locations, effectively identifying if similar values cluster together (hotspots/coldspots) or if dissimilar values are near each other (spatial outliers).
|
|
1164
|
+
|
|
1165
|
+
```python
|
|
1166
|
+
# Import the function
|
|
1167
|
+
from analysistoolbox.geospatial_analysis import ConductSpatialAutocorrelation
|
|
1168
|
+
import pandas as pd
|
|
1169
|
+
|
|
1170
|
+
# Create a sample dataframe
|
|
1171
|
+
df = pd.DataFrame({
|
|
1172
|
+
'lat': [40.7128, 40.7130, 34.0522, 34.0525, 34.0500],
|
|
1173
|
+
'lon': [-74.0060, -74.0065, -118.2437, -118.2440, -118.2500],
|
|
1174
|
+
'value': [100, 105, 10, 15, 8]
|
|
1175
|
+
})
|
|
1176
|
+
|
|
1177
|
+
# Run spatial autocorrelation analysis
|
|
1178
|
+
results = ConductSpatialAutocorrelation(
|
|
1179
|
+
dataframe=df,
|
|
1180
|
+
longitude_column='lon',
|
|
1181
|
+
latitude_column='lat',
|
|
1182
|
+
value_column='value',
|
|
1183
|
+
k=5,
|
|
1184
|
+
plot=True
|
|
1185
|
+
)
|
|
1186
|
+
|
|
1187
|
+
# Print results
|
|
1188
|
+
print(results[['lat', 'lon', 'value', 'cluster_category']])
|
|
1189
|
+
```
|
|
1190
|
+
|
|
1120
1191
|
### Hypothesis Testing
|
|
1121
1192
|
|
|
1122
1193
|
#### ChiSquareTestOfIndependence
|
|
@@ -1659,11 +1730,11 @@ model = CreateNeuralNetwork_SingleOutcome(
|
|
|
1659
1730
|
|
|
1660
1731
|
### Prescriptive Analytics
|
|
1661
1732
|
|
|
1662
|
-
The prescriptive analytics module
|
|
1733
|
+
The prescriptive analytics module empowers users to determine the best course of action through optimization and personalized recommendation systems, bridging the gap between insight and action.
|
|
1663
1734
|
|
|
1664
1735
|
#### ConductLinearOptimization
|
|
1665
1736
|
|
|
1666
|
-
|
|
1737
|
+
Identifies the optimal values for input variables to maximize or minimize a specific output, subject to logical constraints and data-driven linear relationships. This approach is essential for resource allocation, logistics optimization, and cost minimization.
|
|
1667
1738
|
|
|
1668
1739
|
```python
|
|
1669
1740
|
import pandas as pd
|
|
@@ -1694,7 +1765,7 @@ results = ConductLinearOptimization(
|
|
|
1694
1765
|
|
|
1695
1766
|
#### CreateContentBasedRecommender
|
|
1696
1767
|
|
|
1697
|
-
|
|
1768
|
+
Generates a customized recommendation engine using a "two-tower" neural network architecture to match users with highly relevant items or resources based on shared characteristics. This is particularly useful for personalized content delivery, such as matching patients to clinical trials or identifying relevant intelligence reports.
|
|
1698
1769
|
|
|
1699
1770
|
```python
|
|
1700
1771
|
from analysistoolbox.prescriptive_analytics import CreateContentBasedRecommender
|
|
@@ -0,0 +1,111 @@
|
|
|
1
|
+
# Load packages
|
|
2
|
+
import matplotlib.pyplot as plt
|
|
3
|
+
import numpy as np
|
|
4
|
+
|
|
5
|
+
# Declare function
|
|
6
|
+
def FindDerivative(
|
|
7
|
+
f_of_x: sympy.Expr,
|
|
8
|
+
print_functions: bool = False,
|
|
9
|
+
return_derivative_function: bool = False,
|
|
10
|
+
plot_functions: bool = False,
|
|
11
|
+
**plot_kwargs
|
|
12
|
+
) -> Optional[sympy.Expr]:
|
|
13
|
+
"""
|
|
14
|
+
Compute and explore the symbolic derivative of a mathematical function.
|
|
15
|
+
|
|
16
|
+
Derivatives describe how a quantity changes with respect to another — mathematically,
|
|
17
|
+
they are the instantaneous *rate of change* of a function with respect to a variable.
|
|
18
|
+
Symbolic differentiation produces an exact analytic expression for this change,
|
|
19
|
+
rather than a numerical approximation. SymPy is used to compute the derivative expression.
|
|
20
|
+
|
|
21
|
+
The derivative has broad interpretation outside mathematics:
|
|
22
|
+
* In intelligence analysis, the derivative of a threat likelihood function with respect
|
|
23
|
+
to time or another risk factor can reveal *acceleration* in an emerging risk trend.
|
|
24
|
+
* In criminal investigation time series (e.g., incident counts over time), changes in
|
|
25
|
+
slope or inflection points may signal a shift in offender behavior or an outbreak
|
|
26
|
+
pattern that warrants deeper inquiry.
|
|
27
|
+
* In healthcare analytics, derivatives quantify things like the *change in a biomarker
|
|
28
|
+
per unit time* or *how a treatment effect accelerates or decelerates over time*,
|
|
29
|
+
informing early warning signals or intervention effectiveness.
|
|
30
|
+
|
|
31
|
+
This function:
|
|
32
|
+
* Symbolically differentiates `func` with respect to the given variable (or infers the
|
|
33
|
+
single symbolic variable if omitted).
|
|
34
|
+
* Optionally prints the original and derivative expressions.
|
|
35
|
+
* Optionally plots both functions using matplotlib for visual insight into their relationship.
|
|
36
|
+
* Optionally returns the symbolic derivative expression.
|
|
37
|
+
|
|
38
|
+
Parameters
|
|
39
|
+
----------
|
|
40
|
+
f_of_x
|
|
41
|
+
A SymPy expression representing the analytic function (e.g., f(x) = x**2 + 3*x).
|
|
42
|
+
print_functions
|
|
43
|
+
If True, prints the original and derivative expressions for inspection.
|
|
44
|
+
return_derivative_function
|
|
45
|
+
If True, returns the symbolic derivative expression.
|
|
46
|
+
show_plot
|
|
47
|
+
If True, generates a plot of the original and derivative over a default or user-supplied
|
|
48
|
+
domain using `plot_kwargs`.
|
|
49
|
+
**plot_kwargs
|
|
50
|
+
Additional keyword arguments passed to the plotting routine.
|
|
51
|
+
|
|
52
|
+
Returns
|
|
53
|
+
-------
|
|
54
|
+
The symbolic derivative expression if `return_derivative_function` is True, else None.
|
|
55
|
+
|
|
56
|
+
Examples
|
|
57
|
+
--------
|
|
58
|
+
# Symbolically differentiate f(x) = x^3 + 2*x^2
|
|
59
|
+
x = sympy.symbols('x')
|
|
60
|
+
FindDerivative(x**3 + 2*x**2, print_functions=True, return_derivative_function=True)
|
|
61
|
+
|
|
62
|
+
Teaching Note
|
|
63
|
+
-------------
|
|
64
|
+
A derivative transforms a function into a new one that reflects *how fast* the original
|
|
65
|
+
changes. For example, steep slopes (large derivative values) may indicate critical
|
|
66
|
+
transitions. Plotting the original and derivative together often reveals features
|
|
67
|
+
like maxima, minima, or inflection points that are not obvious from raw numbers alone.
|
|
68
|
+
|
|
69
|
+
"""
|
|
70
|
+
# Lazy load uncommon packages
|
|
71
|
+
import sympy
|
|
72
|
+
|
|
73
|
+
# Compute the derivative of the higher-order function using sympy
|
|
74
|
+
try:
|
|
75
|
+
d_f_of_x = sympy.diff(f_of_x, x)
|
|
76
|
+
except NameError:
|
|
77
|
+
raise ValueError("The function must be a sympy expression. Ensure that you have imported sympy and declared x as a sympy symbol. ( e.g., x = sympy.Symbol('x') )" )
|
|
78
|
+
|
|
79
|
+
# Print the derivative function
|
|
80
|
+
if print_functions:
|
|
81
|
+
print("f(x):", f_of_x)
|
|
82
|
+
print("f'(x):", d_f_of_x)
|
|
83
|
+
|
|
84
|
+
# Plot the derivative function if requested
|
|
85
|
+
if plot_functions:
|
|
86
|
+
# Create x values
|
|
87
|
+
x_values = np.linspace(minimum_x, maximum_x, n)
|
|
88
|
+
|
|
89
|
+
# Vectorize the original function and its derivative
|
|
90
|
+
vfunc = np.vectorize(lambda val: f_of_x.evalf(subs={x: val}))
|
|
91
|
+
vfunc_derivative = np.vectorize(lambda val: d_f_of_x.evalf(subs={x: val}))
|
|
92
|
+
|
|
93
|
+
# Create y values for original function and its derivative using the vectorized functions
|
|
94
|
+
y_values_original = vfunc(x_values)
|
|
95
|
+
y_values_derivative = vfunc_derivative(x_values)
|
|
96
|
+
|
|
97
|
+
# Plot the original function
|
|
98
|
+
plt.plot(x_values, y_values_original, label="f(x)")
|
|
99
|
+
|
|
100
|
+
# Plot the derivative function
|
|
101
|
+
plt.plot(x_values, y_values_derivative, label="f'(x)")
|
|
102
|
+
|
|
103
|
+
# Add a legend
|
|
104
|
+
plt.legend()
|
|
105
|
+
|
|
106
|
+
# Show the plot
|
|
107
|
+
plt.show()
|
|
108
|
+
|
|
109
|
+
# Return the derivative function if requested
|
|
110
|
+
if return_derivative_function:
|
|
111
|
+
return d_f_of_x
|
|
@@ -0,0 +1,123 @@
|
|
|
1
|
+
# Load packages
|
|
2
|
+
import matplotlib.pyplot as plt
|
|
3
|
+
import numpy as np
|
|
4
|
+
|
|
5
|
+
# Declare function
|
|
6
|
+
def FindLimitOfFunction(
|
|
7
|
+
f_of_x,
|
|
8
|
+
point,
|
|
9
|
+
step: float = 0.001,
|
|
10
|
+
plot_function: bool = False,
|
|
11
|
+
x_minimum: float = None,
|
|
12
|
+
x_maximum: float = None,
|
|
13
|
+
n: int = 500,
|
|
14
|
+
tangent_line_window: float = 1.0,
|
|
15
|
+
**plot_kwargs
|
|
16
|
+
) -> sympy.Expr:
|
|
17
|
+
"""
|
|
18
|
+
Compute and visualize the limit of a mathematical function as its input
|
|
19
|
+
approaches a specified point.
|
|
20
|
+
|
|
21
|
+
In calculus, a *limit* describes the value that a function’s output
|
|
22
|
+
approaches as its input gets arbitrarily close to a given point, even
|
|
23
|
+
if the function is not defined at that point itself. Limits underpin
|
|
24
|
+
many core ideas in analysis — including continuity, derivatives, and
|
|
25
|
+
asymptotic behavior — and have practical interpretive value in analytic
|
|
26
|
+
domains.
|
|
27
|
+
|
|
28
|
+
This function:
|
|
29
|
+
* Uses SymPy’s symbolic `limit` capability to compute the limit of
|
|
30
|
+
`f_of_x` as its independent variable approaches `point`.
|
|
31
|
+
* Optionally plots the function near `point` and overlays a tangent
|
|
32
|
+
line to illustrate the local approach behavior.
|
|
33
|
+
* Returns the symbolic limit expression for programmatic inspection.
|
|
34
|
+
|
|
35
|
+
Parameters
|
|
36
|
+
----------
|
|
37
|
+
f_of_x
|
|
38
|
+
A symbolic or numeric expression representing f(x). Limits can
|
|
39
|
+
reveal behavior near discontinuities or indeterminate forms.:contentReference[oaicite:1]{index=1}
|
|
40
|
+
point
|
|
41
|
+
The input value that x approaches (e.g., 0, infinity, or a finite
|
|
42
|
+
critical point).
|
|
43
|
+
step
|
|
44
|
+
Step size used for numerical approximation or plotting.
|
|
45
|
+
plot_function
|
|
46
|
+
If True, generate a matplotlib plot of `func` near `point`.
|
|
47
|
+
x_minimum, x_maximum
|
|
48
|
+
The range over which to plot `func`. If None, a default window
|
|
49
|
+
around `point` is chosen.
|
|
50
|
+
n
|
|
51
|
+
Number of points used for plotting.
|
|
52
|
+
tangent_line_window
|
|
53
|
+
Width of the window around `point` over which the tangent line is
|
|
54
|
+
drawn.
|
|
55
|
+
**plot_kwargs
|
|
56
|
+
Additional keyword arguments passed to the plotting routine.
|
|
57
|
+
|
|
58
|
+
Returns
|
|
59
|
+
-------
|
|
60
|
+
A symbolic SymPy expression representing the limit of `func` as x→`point`.
|
|
61
|
+
|
|
62
|
+
Examples
|
|
63
|
+
--------
|
|
64
|
+
# Symbolically compute the limit of sin(x)/x as x→0
|
|
65
|
+
x = sympy.symbols('x')
|
|
66
|
+
FindLimitOfFunction(sympy.sin(x)/x, point=0, plot_function=True)
|
|
67
|
+
|
|
68
|
+
Teaching Note
|
|
69
|
+
-------------
|
|
70
|
+
A limit expresses the *behavior* of a function in the vicinity of a value:
|
|
71
|
+
it does not require the function to be defined at that exact point. For
|
|
72
|
+
instance, sin(x)/x is undefined at x=0, yet its limit as x approaches
|
|
73
|
+
0 is 1 — reflecting the function’s approach behavior, not its value at
|
|
74
|
+
the point.
|
|
75
|
+
"""
|
|
76
|
+
|
|
77
|
+
# Create array of values based on the function
|
|
78
|
+
x = np.linspace(x_minimum, x_maximum, n)
|
|
79
|
+
y = np.zeros(len(x))
|
|
80
|
+
for i in range(len(x)):
|
|
81
|
+
y[i] = f_of_x(x[i])
|
|
82
|
+
|
|
83
|
+
# Calculate the x_tangent
|
|
84
|
+
x_tangent = (f_of_x(point+step) - f_of_x(point-step)) / (2*step)
|
|
85
|
+
|
|
86
|
+
# Calculate the limit at the point of interest
|
|
87
|
+
try:
|
|
88
|
+
limit = f_of_x(point)
|
|
89
|
+
print("The limit at x={0} is ~{1}".format(point, limit))
|
|
90
|
+
# Create the tangent line
|
|
91
|
+
y_tangent = f_of_x(point) + x_tangent*(x - point)
|
|
92
|
+
|
|
93
|
+
# Plot the function if requested
|
|
94
|
+
if plot_function:
|
|
95
|
+
# Plot point at the point of interest
|
|
96
|
+
plt.plot(point, f_of_x(point), "ro")
|
|
97
|
+
|
|
98
|
+
# Plot the tangent line
|
|
99
|
+
if plot_tangent_line:
|
|
100
|
+
# plt.scatter(x0, my_function(x0), color='red')
|
|
101
|
+
plt.plot(x, y_tangent, label='Tangent line', color='red', alpha=0.5)
|
|
102
|
+
plt.scatter(point, f_of_x(point), color='red')
|
|
103
|
+
|
|
104
|
+
# Plot the function if requested
|
|
105
|
+
if plot_function:
|
|
106
|
+
# Plot the function
|
|
107
|
+
plt.plot(x, y, color="black", label='Function')
|
|
108
|
+
|
|
109
|
+
# Add title
|
|
110
|
+
plt.title("f(x)")
|
|
111
|
+
|
|
112
|
+
# Add legend
|
|
113
|
+
plt.legend()
|
|
114
|
+
|
|
115
|
+
# Show plot
|
|
116
|
+
plt.show()
|
|
117
|
+
plt.clf()
|
|
118
|
+
|
|
119
|
+
# Return the limit
|
|
120
|
+
return limit
|
|
121
|
+
|
|
122
|
+
except ZeroDivisionError:
|
|
123
|
+
print("The limit at x={0} is undefined.".format(point))
|
|
@@ -0,0 +1,106 @@
|
|
|
1
|
+
# Load packages
|
|
2
|
+
import matplotlib.pyplot as plt
|
|
3
|
+
import numpy as np
|
|
4
|
+
import seaborn as sns
|
|
5
|
+
|
|
6
|
+
# Declare function
|
|
7
|
+
def FindMinimumSquareLoss(
|
|
8
|
+
observed_values,
|
|
9
|
+
predicted_values,
|
|
10
|
+
show_plot: bool = False,
|
|
11
|
+
**plot_kwargs
|
|
12
|
+
):
|
|
13
|
+
"""
|
|
14
|
+
Find the parameter value that minimizes squared loss between observed data
|
|
15
|
+
and a model function.
|
|
16
|
+
|
|
17
|
+
Squared loss measures the total discrepancy between observed values and
|
|
18
|
+
model-predicted values by summing the squared residuals. Minimizing this
|
|
19
|
+
quantity yields the parameter value that best fits the data under a
|
|
20
|
+
least-squares criterion.
|
|
21
|
+
|
|
22
|
+
In analytic practice, least-squares minimization formalizes a common
|
|
23
|
+
reasoning task: choosing the explanation, model, or parameterization that
|
|
24
|
+
best accounts for observed evidence while penalizing large errors more
|
|
25
|
+
heavily than small ones.
|
|
26
|
+
|
|
27
|
+
This function:
|
|
28
|
+
* Constructs a squared loss function based on the difference between
|
|
29
|
+
observed `y_values` and the model's predictions.
|
|
30
|
+
* Symbolically differentiates the loss with respect to the model
|
|
31
|
+
parameter.
|
|
32
|
+
* Solves for the parameter value that minimizes total squared error.
|
|
33
|
+
* Optionally prints intermediate results and visualizes the fit.
|
|
34
|
+
|
|
35
|
+
Parameters
|
|
36
|
+
----------
|
|
37
|
+
observed_values
|
|
38
|
+
Observed dependent variable values (e.g., incidents, biomarker levels,
|
|
39
|
+
risk scores).
|
|
40
|
+
predicted_values
|
|
41
|
+
Predicted dependent variable values (e.g., incidents, biomarker levels,
|
|
42
|
+
risk scores).
|
|
43
|
+
show_plot
|
|
44
|
+
If True, plots observed data against the fitted model.
|
|
45
|
+
**plot_kwargs
|
|
46
|
+
Additional keyword arguments passed to the plotting routine.
|
|
47
|
+
|
|
48
|
+
Returns
|
|
49
|
+
-------
|
|
50
|
+
The parameter value that minimizes the squared loss between observed data
|
|
51
|
+
and the model.
|
|
52
|
+
|
|
53
|
+
Examples
|
|
54
|
+
--------
|
|
55
|
+
FindMinimumSquareLoss(
|
|
56
|
+
observed_values=[1, 2, 3, 4],
|
|
57
|
+
predicted_values=[2.1, 4.2, 5.9, 8.3],
|
|
58
|
+
show_plot=True
|
|
59
|
+
)
|
|
60
|
+
|
|
61
|
+
Teaching Note
|
|
62
|
+
-------------
|
|
63
|
+
Squared loss grows rapidly as errors increase. This makes least-squares
|
|
64
|
+
methods especially sensitive to large deviations, which is often desirable
|
|
65
|
+
when large errors correspond to analytically meaningful failures rather
|
|
66
|
+
than random noise.
|
|
67
|
+
"""
|
|
68
|
+
|
|
69
|
+
# Check if the number of observed values is equal to the number of predicted values.
|
|
70
|
+
if len(observed_values) != len(predicted_values):
|
|
71
|
+
raise ValueError("The number of observed values must be equal to the number of predicted values.")
|
|
72
|
+
|
|
73
|
+
# If values are passed as a list, convert them to a numpy array.
|
|
74
|
+
if type(observed_values) == list:
|
|
75
|
+
observed_values = np.array(observed_values)
|
|
76
|
+
if type(predicted_values) == list:
|
|
77
|
+
predicted_values = np.array(predicted_values)
|
|
78
|
+
|
|
79
|
+
# Count the number of observed values.
|
|
80
|
+
n = len(observed_values)
|
|
81
|
+
|
|
82
|
+
# Calculate the prediction error.
|
|
83
|
+
sq_prediction_error = (predicted_values - observed_values) ** 2
|
|
84
|
+
|
|
85
|
+
# Calculate the sum of squares.
|
|
86
|
+
sum_of_squares = sum(sq_prediction_error)
|
|
87
|
+
|
|
88
|
+
# Calculate the minimum square loss.
|
|
89
|
+
FindMinimumSquareLoss = sum_of_squares / n
|
|
90
|
+
|
|
91
|
+
# Plot the prediction errors in a plot, if requested.
|
|
92
|
+
if show_plot:
|
|
93
|
+
sns.boxplot(y=sq_prediction_error, color="grey")
|
|
94
|
+
# Add the minimum square loss as a horizontal line.
|
|
95
|
+
plt.axhline(FindMinimumSquareLoss, color="r", linestyle="--")
|
|
96
|
+
# Add labels.
|
|
97
|
+
plt.title("Squared Prediction Errors", fontsize=14, fontweight="bold")
|
|
98
|
+
plt.text(0.5, FindMinimumSquareLoss + .01, "Min. Sq. Loss = " + str(round(FindMinimumSquareLoss, 2)))
|
|
99
|
+
# Remove borders from plot
|
|
100
|
+
sns.despine(bottom=True)
|
|
101
|
+
# Show plot
|
|
102
|
+
plt.show()
|
|
103
|
+
|
|
104
|
+
# Return the minimum square loss.
|
|
105
|
+
return FindMinimumSquareLoss
|
|
106
|
+
|