HUFS-DAT/2024-2_Seminar
Nov 2024 – Dec 2024
2 Months active
Languages Used
Jupyter NotebookPython
Technical Skills
Data AnalysisData PreprocessingData VisualizationDimensionality ReductionMachine LearningMatplotlib
Jiyoon-B
PROFILE
Jiyoon-b
Over a two-month period, this developer contributed to the HUFS-DAT/2024-2_Seminar repository by building robust data analysis and preprocessing pipelines using Python and Jupyter Notebook. They developed a PCA-based image data notebook for Fashion MNIST, implementing multi-component dimensionality reduction, explained variance analysis, reconstruction error evaluation, and 2D visualizations with Matplotlib and Seaborn to enhance interpretability. Additionally, they improved reliability in baseball analytics notebooks by correcting file paths and aligning preprocessing steps. In December, they delivered an end-to-end data preprocessing notebook for the model1 dataset, incorporating missing value checks, correlation analysis, outlier handling, and scaling with Pandas and Scikit-learn.
Overall Statistics
Feature vs Bugs
67%Features
Repository Contributions
3Total
Bugs
1
Commits
3
Features
2
Lines of code
5,394
Activity Months2
Your Network
6 peopleWork History
December 2024
1 Commits • 1 Features
Dec 1, 2024Month: 2024-12 | Repository: HUFS-DAT/2024-2_Seminar | Focus: Feature delivery in data preprocessing notebook for model1 dataset with end-to-end data prep pipeline.
1 Commits • 1 Features
Dec 1, 2024Month: 2024-12 | Repository: HUFS-DAT/2024-2_Seminar | Focus: Feature delivery in data preprocessing notebook for model1 dataset with end-to-end data prep pipeline.
December 2024
November 2024
2 Commits • 1 Features
Nov 1, 2024November 2024 performance highlights focused on delivering a reproducible, analytics-ready environment in HUFS-DAT/2024-2_Seminar. Key features delivered a PCA-based image data notebook for Fashion MNIST with multi-component PCA, variance analysis, reconstruction error, and label-colored 2D visualization, enabling deeper dimensionality reduction experiments. Major fixes addressed notebook reliability in Baseball Analytics by correcting file paths, execution counts, and preprocessing/model parameters for metrics like woba and exit_velocity_avg. Collectively, these efforts improved data exploration capabilities, reliability, and onboarding for analysts, with tangible business value in faster insights and more robust analytics pipelines.
November 2024
2 Commits • 1 Features
Nov 1, 2024November 2024 performance highlights focused on delivering a reproducible, analytics-ready environment in HUFS-DAT/2024-2_Seminar. Key features delivered a PCA-based image data notebook for Fashion MNIST with multi-component PCA, variance analysis, reconstruction error, and label-colored 2D visualization, enabling deeper dimensionality reduction experiments. Major fixes addressed notebook reliability in Baseball Analytics by correcting file paths, execution counts, and preprocessing/model parameters for metrics like woba and exit_velocity_avg. Collectively, these efforts improved data exploration capabilities, reliability, and onboarding for analysts, with tangible business value in faster insights and more robust analytics pipelines.
Activity
Loading activity data...
Quality Metrics
Correctness66.6%
Maintainability66.6%
Architecture66.6%
Performance66.6%
AI Usage20.0%
Skills & Technologies
Programming Languages
Jupyter NotebookPython
Technical Skills
Data AnalysisData PreprocessingData VisualizationDimensionality ReductionMachine LearningMatplotlibPCAPandasPythonScikit-learnSeaborn
Repositories Contributed To
Overview of all repositories you've contributed to across your timeline