patvela/proyectos-universidad

How patvela/proyectos-universidad works

ECG classification with a residual deep-learning model is the product. The principal workflow is CSV upload in the Flask interface, signal preparation and model inference, then a summary with plots shown to the user. Separate capabilities include MIT-BIH data preparation and model training, plus CINC2017-based prediction. Edges reflect the README workflows and the available source references; several web and prediction internals are only partially excerpted, so their finer-grained wiring is not expanded.

Main entry points: app.py Flask application · train.py Model training · predict.py Prediction and summary · data.py MIT-BIH preparation

Components 17

Web application

Browser interface src/static/js/main.js
Page templates src/templates
Flask application src/app.py
Uploaded ECG src/uploads
Plots and results src/app.py

Other components

ECG user
PhysioNet datasets

Connections 22

ECG user → Browser interface: uses README.md
Browser interface → Flask application: submits ECG src/static/js/main.js
Flask application → Page templates: renders pages
Flask application → Uploaded ECG: saves uploads
Flask application → Prediction and summary: invokes prediction src/app.py
Flask application → Runtime configuration: configures sampling src/app.py
Flask application → Plots and results: serves plots src/app.py
Browser interface → Plots and results: displays results src/static/js/main.js
Prediction and summary → Signal preprocessing: preprocesses signal src/predict.py
Prediction and summary → ECG model: uses model src/predict.py
Prediction and summary → Runtime configuration: reads settings src/predict.py
Prediction and summary → CINC2017 records: loads records src/predict.py
ECG user → MIT-BIH preparation: requests preparation README.md
MIT-BIH preparation → PhysioNet datasets: downloads records src/data.py
MIT-BIH preparation → Prepared datasets: writes prepared data src/data.py
ECG user → Model training: starts training README.md
Model training → Prepared datasets: loads training data src/train.py
Model training → ECG model: builds model src/train.py
Model training → Trained model files: saves checkpoints src/train.py
Model training → Evaluation utilities: evaluates model src/train.py
Model training → Runtime configuration: reads settings src/train.py
MIT-BIH preparation → Runtime configuration: reads settings src/data.py

More Python diagrams: minitorch/module-0 · mrkeee/coherent · nasa-impact/marine_debris_ml · plushkinv/yadrenovpn · terrafloww/rasteret

Drawn by GitDiagram on September 29, 2026 from the repository's file tree, README and source files. Read as Markdown ·