How gkrishna4346/battery-clinic-ai works
Battery Clinic AI is a Streamlit app for predicting laptop battery health, managing training data, exploring model behavior, and explaining results. The main workflow is user input or uploaded CSV → app orchestration → model prediction and interpretation → displayed or downloadable result. Separate branches support data replacement and restore, model comparison and analysis, and the Mr. GK assistant. Edges reflect the excerpts and README; unsampled app sections limit detail about some page-to-engine calls.
Main entry points: app.py Streamlit app · versions.py Version management · mrgk.py Mr. GK assistant · engine.py Modeling engine
Components 13
Application
- Streamlit app
app.py - Domain settings
config.py
Prediction and analysis
- Modeling engine
engine.py - Prediction reports
report.py
Data management
- Active dataset
data/battery_health.csv - Version management
versions.py - Version archive
data/versions
Assistant
- Mr. GK assistant
mrgk.py - Assistant knowledge
mrgk_kb.py - User guide
docs/user_guide.html
Other components
- Laptop user
- GitHub repository
- Anthropic API
Connections 13
- Laptop user → Streamlit app: uses
README.md - Streamlit app → Modeling engine: requests predictions
app.py - Modeling engine → Domain settings: reads domain rules
engine.py - Streamlit app → Domain settings: uses settings
app.py - Streamlit app → Version management: initializes versions
app.py - Version management → Active dataset: activates dataset
versions.py - Version management → Version archive: stores versions
versions.py - Streamlit app → Prediction reports: offers report downloads
README.md - Streamlit app → Mr. GK assistant: opens assistant
app.py - Mr. GK assistant → Assistant knowledge: selects answers
mrgk.py - Mr. GK assistant → User guide: searches guide
README.md - Mr. GK assistant → Anthropic API: requests AI answers
mrgk.py - Version management → GitHub repository: syncs optionally
versions.py