How saiteja8326/ai-powered-smart-study-planner works
This is a Streamlit student-performance tracker: students register or log in, enter weekly topic marks, review their history and visualizations, and receive insights or a focused study plan. The main dashboard delegates persistence, analysis, deep-learning inference, and scheduling to utility modules. The README also describes model training and multiple architectures; training scripts and preprocessing are represented, but their runtime wiring is not established by the excerpts. The repository contains alternate Streamlit interfaces with different database APIs, so their links are limited to imports shown in those files.
Main entry points: main.py Main dashboard · train_models.py NN and autoencoder training · database.py SQLite records · engine_dl.py DL inference
Components 16
Student interfaces
- Main dashboard
app/main.py - AI Scholar portal
app/ui.py - Performance tracker
app/app_streamlit.py
Academic records
- SQLite records
utils/database.py
Analysis and guidance
- Topic analysis
utils/analysis.py - DL inference
utils/engine_dl.py - Smart schedule
utils/schedule.py
Model training
- Feature preprocessing
utils/preprocessing.py - NN and autoencoder training
models/train_models.py - Neural network trainer
models/train_nn.py - LSTM trainer
models/train_lstm.py - Autoencoder trainer
models/train_autoencoder.py - Student dataset
dataset/student_dataset_25000 (1).xlsx - Neural network model
models/saved_models/nn_model.h5 - Autoencoder model
models/saved_models/autoencoder.h5
Other components
- Student
Connections 20
- Student → Main dashboard: uses
README.md - Main dashboard → SQLite records: reads and writes
app/main.py - Main dashboard → Topic analysis: requests analysis
app/main.py - Main dashboard → DL inference: requests insights
app/main.py - Main dashboard → Smart schedule: requests plan
app/main.py - Main dashboard → Student dataset: visualizes progress
README.md - AI Scholar portal → SQLite records: imports storage API
app/ui.py - AI Scholar portal → Smart schedule: requests plan
app/ui.py - Performance tracker → SQLite records: reads and writes
app/app_streamlit.py - Performance tracker → Smart schedule: requests plan
app/app_streamlit.py - Main dashboard → SQLite records: stores weekly marks
README.md - NN and autoencoder training → Student dataset: loads
models/train_models.py - NN and autoencoder training → Neural network model: saves
models/train_models.py - NN and autoencoder training → Autoencoder model: saves
models/train_models.py - DL inference → Neural network model: loads
utils/engine_dl.py - DL inference → Autoencoder model: loads
utils/engine_dl.py - Student dataset → Feature preprocessing: supports feature preparation
- Neural network trainer → NN and autoencoder training: training variant
- LSTM trainer → NN and autoencoder training: sequential model variant
- Autoencoder trainer → NN and autoencoder training: training variant