How mohxndaziz/predicting-ev-charging-loads-using-pytorch works
This repository is a notebook-based EV charging energy predictor. Its documented workflow aligns charging sessions with nearby traffic observations, cleans and encodes the merged data, trains a neural network, and saves its weights. The README describes these stages and the saved model, but the notebook’s source was not sampled, so the graph reflects documented behavior rather than verified notebook calls.
Main entry points: code.ipynb Align and Prepare · model_state_dict.pth Saved Weights · Dataset 1_EV charging reports.csv Charging Sessions · Dataset 6_Local traffic distribution.csv Traffic Counts
Components 6
Input Data
- Charging Sessions
Dataset 1_EV charging reports.csv - Traffic Counts
Dataset 6_Local traffic distribution.csv
Modeling
- Align and Prepare
code.ipynb - Train Network
code.ipynb - Saved Weights
model_state_dict.pth
Other components
- Analyst
Connections 6
- Analyst → Align and Prepare: runs notebook
README.md - Charging Sessions → Align and Prepare: session data
README.md - Traffic Counts → Align and Prepare: traffic data
README.md - Align and Prepare → Train Network: provides features
README.md - Train Network → Saved Weights: saves weights
README.md - Analyst → Saved Weights: loads weights
README.md