D682 | Artificial Intelligence Optimization for Computer Scientists
WGU D682
Artificial Intelligence Optimization for Computer
Scientists
Task 1 - Initial AI Model Implementation
Complete narrative, Python implementation, testing evidence, and practice dataset
PRACTICE DATASET NOTICE
The dataset and metric results in this document are a complete synthetic practice
example created for this resource. They are not the WGU-supplied Excel workbook and
are not represented as official WGU data.
2026/2027 Updated
Prepared as a professional rubric-aligned reference for D682 Task 1
D682 Task 1 - Initial AI Model Implementation
Page 1
, D682 | Artificial Intelligence Optimization for Computer Scientists
Document Purpose and Alignment
D682 focuses on implementing, testing, optimizing, interpreting, and adapting artificial intelligence
solutions. Western Governors University describes the course as requiring research into AI
approaches, practical implementation and testing, optimization of AI models, predictive
interpretation, and adaptation to additional application scenarios (Western Governors University,
2026).
Public completion reports for the current performance assessment consistently describe Task 1 as the
initial-model stage: establish the GitLab work, research three candidate algorithms, implement an
uncomplicated Python model, evaluate it with two metrics, interpret the results, and support the
written analysis with sources. Recent 2026 student reports also continue to identify Tasks 1 and 2 as
the coding-intensive portions of D682.
Scope note
The authenticated WGU Task 1 page and official Excel workbook are not publicly available.
Accordingly, this document does not claim to reproduce WGU's private prompt verbatim. It follows
the strongest publicly corroborated Task 1 structure while keeping the practice dataset explicitly
separate from WGU-supplied materials.
Contents
1. Scenario and Problem Definition
2. Part A - Repository and Version Control
3. Part B - Algorithm Research and Selection
4. Part C - Python Implementation
5. Part D - Testing and Validation
6. Part E - Sources and Professional Documentation
7. Conclusions
Appendix A - Complete Python Script
Appendix B - requirements.txt and Run Instructions
Appendix C - Synthetic Practice Dataset
Appendix D - GitLab Submission Package
1. Scenario and Problem Definition
The working scenario for this Task 1 reference is an environmental-health prediction problem for an
EPA-style stakeholder. Public D682 completion reports describe the course case as involving
environmental conditions, weather-related information, and health-risk prediction. They also report
that the supplied case narrative and data workbook do not always align perfectly, which makes
documented assumptions important.
For this reference, the stakeholder needs an initial regression model that estimates a continuous
healthRiskScore from environmental and community-health indicators. The model is intentionally a
D682 Task 1 - Initial AI Model Implementation
Page 2
, D682 | Artificial Intelligence Optimization for Computer Scientists
baseline implementation. Feature engineering, extensive preprocessing, hyperparameter search, and
systematic model optimization are reserved for the next optimization stage rather than being folded
into Task 1.
The synthetic dataset contains 200 complete observations and ten predictor variables. Every row is
populated; there are no blank cells or missing values. The target variable is healthRiskScore. The
practice data were generated solely so that the implementation, metric calculations, plots, and written
interpretation in this PDF are internally consistent and fully reproducible.
Practice Data Dictionary
Field Role / Type Definition
Ambient temperature in degrees
temperatureF Numeric
Fahrenheit.
humidityPct Numeric Relative humidity percentage.
Observed wind speed in miles per
windSpeedMph Numeric
hour.
pressureHpa Numeric Atmospheric pressure in hectopascals.
precipitationIn Numeric Recorded precipitation in inches.
visibilityMiles Numeric Horizontal visibility in miles.
uvIndex Numeric Ultraviolet exposure index.
Practice pollen-burden index on a 0-12
pollenIndex Numeric
scale.
Practice community asthma
asthmaPrevalencePct Numeric
prevalence percentage.
populationDensity Numeric Practice population density measure.
Continuous practice health-risk score
healthRiskScore Target
used for regression.
Modeling Objective
The objective is to train an initial supervised regression model that predicts healthRiskScore from the
ten available predictors. The baseline model should be easy to reproduce, sufficiently flexible to
capture nonlinear relationships, and suitable for later optimization. Task 1 is therefore treated as a
controlled first implementation rather than an attempt to produce the best possible final model.
2. Part A - Repository and Version Control
The project is organized around a dedicated GitLab branch named Task1. A separate branch keeps
Task 1 work isolated from later optimization changes and gives the evaluator a clear, reproducible
version of the initial model. Public D682 completion reports specifically identify a Task1 branch, a
GitLab repository link, a GitLab graph image, a written report, and evaluator notes that identify the
requirements file and Python entry point.
D682 Task 1 - Initial AI Model Implementation
Page 3
WGU D682
Artificial Intelligence Optimization for Computer
Scientists
Task 1 - Initial AI Model Implementation
Complete narrative, Python implementation, testing evidence, and practice dataset
PRACTICE DATASET NOTICE
The dataset and metric results in this document are a complete synthetic practice
example created for this resource. They are not the WGU-supplied Excel workbook and
are not represented as official WGU data.
2026/2027 Updated
Prepared as a professional rubric-aligned reference for D682 Task 1
D682 Task 1 - Initial AI Model Implementation
Page 1
, D682 | Artificial Intelligence Optimization for Computer Scientists
Document Purpose and Alignment
D682 focuses on implementing, testing, optimizing, interpreting, and adapting artificial intelligence
solutions. Western Governors University describes the course as requiring research into AI
approaches, practical implementation and testing, optimization of AI models, predictive
interpretation, and adaptation to additional application scenarios (Western Governors University,
2026).
Public completion reports for the current performance assessment consistently describe Task 1 as the
initial-model stage: establish the GitLab work, research three candidate algorithms, implement an
uncomplicated Python model, evaluate it with two metrics, interpret the results, and support the
written analysis with sources. Recent 2026 student reports also continue to identify Tasks 1 and 2 as
the coding-intensive portions of D682.
Scope note
The authenticated WGU Task 1 page and official Excel workbook are not publicly available.
Accordingly, this document does not claim to reproduce WGU's private prompt verbatim. It follows
the strongest publicly corroborated Task 1 structure while keeping the practice dataset explicitly
separate from WGU-supplied materials.
Contents
1. Scenario and Problem Definition
2. Part A - Repository and Version Control
3. Part B - Algorithm Research and Selection
4. Part C - Python Implementation
5. Part D - Testing and Validation
6. Part E - Sources and Professional Documentation
7. Conclusions
Appendix A - Complete Python Script
Appendix B - requirements.txt and Run Instructions
Appendix C - Synthetic Practice Dataset
Appendix D - GitLab Submission Package
1. Scenario and Problem Definition
The working scenario for this Task 1 reference is an environmental-health prediction problem for an
EPA-style stakeholder. Public D682 completion reports describe the course case as involving
environmental conditions, weather-related information, and health-risk prediction. They also report
that the supplied case narrative and data workbook do not always align perfectly, which makes
documented assumptions important.
For this reference, the stakeholder needs an initial regression model that estimates a continuous
healthRiskScore from environmental and community-health indicators. The model is intentionally a
D682 Task 1 - Initial AI Model Implementation
Page 2
, D682 | Artificial Intelligence Optimization for Computer Scientists
baseline implementation. Feature engineering, extensive preprocessing, hyperparameter search, and
systematic model optimization are reserved for the next optimization stage rather than being folded
into Task 1.
The synthetic dataset contains 200 complete observations and ten predictor variables. Every row is
populated; there are no blank cells or missing values. The target variable is healthRiskScore. The
practice data were generated solely so that the implementation, metric calculations, plots, and written
interpretation in this PDF are internally consistent and fully reproducible.
Practice Data Dictionary
Field Role / Type Definition
Ambient temperature in degrees
temperatureF Numeric
Fahrenheit.
humidityPct Numeric Relative humidity percentage.
Observed wind speed in miles per
windSpeedMph Numeric
hour.
pressureHpa Numeric Atmospheric pressure in hectopascals.
precipitationIn Numeric Recorded precipitation in inches.
visibilityMiles Numeric Horizontal visibility in miles.
uvIndex Numeric Ultraviolet exposure index.
Practice pollen-burden index on a 0-12
pollenIndex Numeric
scale.
Practice community asthma
asthmaPrevalencePct Numeric
prevalence percentage.
populationDensity Numeric Practice population density measure.
Continuous practice health-risk score
healthRiskScore Target
used for regression.
Modeling Objective
The objective is to train an initial supervised regression model that predicts healthRiskScore from the
ten available predictors. The baseline model should be easy to reproduce, sufficiently flexible to
capture nonlinear relationships, and suitable for later optimization. Task 1 is therefore treated as a
controlled first implementation rather than an attempt to produce the best possible final model.
2. Part A - Repository and Version Control
The project is organized around a dedicated GitLab branch named Task1. A separate branch keeps
Task 1 work isolated from later optimization changes and gives the evaluator a clear, reproducible
version of the initial model. Public D682 completion reports specifically identify a Task1 branch, a
GitLab repository link, a GitLab graph image, a written report, and evaluator notes that identify the
requirements file and Python entry point.
D682 Task 1 - Initial AI Model Implementation
Page 3