KU-BIG/KUBIG_2025_FALL
Jul 2025 – Dec 2025
3 Months active
Languages Used
Jupyter NotebookPythonShellBase64Markdown
Technical Skills
Attention MechanismsBag of WordsBeautifulSoupCBOW ModelData AnalysisData Preprocessing
Seohyeon
PROFILE
Seohyeon
Beth contributed to the KU-BIG/KUBIG_2025_FALL repository by developing a suite of educational resources and hands-on notebooks spanning natural language processing, time series forecasting, and computer vision. She implemented end-to-end workflows for tasks such as sentiment analysis, neural machine translation with attention, stock price prediction using RNN and LSTM models, and FashionMNIST neural network training, leveraging Python, PyTorch, and Jupyter Notebooks. Beth also expanded the curriculum with BERT-based classification and PEFT-enabled text generation experiments, and improved onboarding through comprehensive documentation updates. Her work emphasized reproducibility, modular design, and clear instructional materials, supporting both curriculum adoption and practical experimentation.
Overall Statistics
Feature vs Bugs
100%Features
Repository Contributions
20Total
Bugs
0
Commits
20
Features
7
Lines of code
43,394
Activity Months3
Your Network
1028 people
Work History
December 2025
4 Commits • 1 Features
Dec 1, 2025December 2025 in KU-BIG/KUBIG_2025_FALL delivered a comprehensive PATHFINDER README update to clarify purpose, architecture, features, installation, usage, and contributor information. No major bugs were fixed this month. Overall impact: improved onboarding, faster feature adoption, and reduced support queries due to a single, maintained source of truth. Demonstrated skills in markdown documentation, onboarding design, and version-control discipline through iterative README refinements.
4 Commits • 1 Features
Dec 1, 2025December 2025 in KU-BIG/KUBIG_2025_FALL delivered a comprehensive PATHFINDER README update to clarify purpose, architecture, features, installation, usage, and contributor information. No major bugs were fixed this month. Overall impact: improved onboarding, faster feature adoption, and reduced support queries due to a single, maintained source of truth. Demonstrated skills in markdown documentation, onboarding design, and version-control discipline through iterative README refinements.
December 2025
August 2025
6 Commits • 3 Features
Aug 1, 2025August 2025 performance for KU-BIG/KUBIG_2025_FALL focused on expanding NLP curriculum resources and enabling hands-on experimentation with modern NLP models. Delivered study materials updates (including a Week 4 diagram image as base64 and a Week 6 NLP PDF) and three notebooks covering BERT-based classification and PEFT-enabled generation workflows, enhancing reproducibility, evaluation, and learning impact for the fall cohort.
August 2025
6 Commits • 3 Features
Aug 1, 2025August 2025 performance for KU-BIG/KUBIG_2025_FALL focused on expanding NLP curriculum resources and enabling hands-on experimentation with modern NLP models. Delivered study materials updates (including a Week 4 diagram image as base64 and a Week 6 NLP PDF) and three notebooks covering BERT-based classification and PEFT-enabled generation workflows, enhancing reproducibility, evaluation, and learning impact for the fall cohort.
July 2025
10 Commits • 3 Features
Jul 1, 2025July 2025 Monthly Summary — KU-BIG/KUBIG_2025_FALL: Deliverables overview: - NLP educational notebooks and tutorials: cross-lingual (English & Korean) coverage of Word2Vec, FastText, CBOW, sentiment analysis with multiple models, and neural machine translation with attention; includes hands-on tutorials and model comparisons. - Stock price prediction notebook: end-to-end workflow for data fetching, preprocessing, and training vanilla RNN/LSTM models with evaluation against true values. - FashionMNIST neural network training notebook: end-to-end pipeline for data loading, model definition, training, evaluation, and model persistence. Impact: - Provided scalable, reproducible learning resources that accelerate curriculum adoption and practitioners' ability to run end-to-end ML/NLP experiments. - Demonstrated practical data pipelines and model development across NLP, time-series forecasting, and computer vision domains. Technologies/skills demonstrated: - Python, Jupyter notebooks, data preprocessing, and model training/evaluation pipelines - NLP: Word2Vec, FastText, CBOW, sentiment analysis, attention-based neural machine translation - Time-series: RNN/LSTM for stock forecasting - Computer vision: FashionMNIST training workflow
10 Commits • 3 Features
Jul 1, 2025July 2025 Monthly Summary — KU-BIG/KUBIG_2025_FALL: Deliverables overview: - NLP educational notebooks and tutorials: cross-lingual (English & Korean) coverage of Word2Vec, FastText, CBOW, sentiment analysis with multiple models, and neural machine translation with attention; includes hands-on tutorials and model comparisons. - Stock price prediction notebook: end-to-end workflow for data fetching, preprocessing, and training vanilla RNN/LSTM models with evaluation against true values. - FashionMNIST neural network training notebook: end-to-end pipeline for data loading, model definition, training, evaluation, and model persistence. Impact: - Provided scalable, reproducible learning resources that accelerate curriculum adoption and practitioners' ability to run end-to-end ML/NLP experiments. - Demonstrated practical data pipelines and model development across NLP, time-series forecasting, and computer vision domains. Technologies/skills demonstrated: - Python, Jupyter notebooks, data preprocessing, and model training/evaluation pipelines - NLP: Word2Vec, FastText, CBOW, sentiment analysis, attention-based neural machine translation - Time-series: RNN/LSTM for stock forecasting - Computer vision: FashionMNIST training workflow
July 2025
Activity
Loading activity data...
Quality Metrics
Correctness82.6%
Maintainability80.0%
Architecture80.0%
Performance73.6%
AI Usage32.0%
Skills & Technologies
Programming Languages
Base64Jupyter NotebookMarkdownPythonShell
Technical Skills
API integrationAttention MechanismsBERTBag of WordsBeautifulSoupCBOW ModelData AnalysisData PreprocessingData ScienceData VisualizationDeep LearningFastTextGated Recurrent Unit (GRU)GensimGoogle Colab
Repositories Contributed To
Overview of all repositories you've contributed to across your timeline