Skip to content

aleablu/CleanAirProject_analysis

Repository files navigation

Usage

Required dependencies: PyTorch, pandas, numpy, tqdm, skimage

Execute "Regression.py" to train the network and see loss function values.

Command line options:

  • -s, --save-model --> if present this flag forces saving of trained model
  • --learning-rate, --batch-size, --epochs --> to specify parameters to use, defaults are 0.001, 128, 20
  • -p --> plot predicted vs original labels
  • --time-frame --> daily, weekly, monthly, seasonally, deafults to daily

About

Analysis module for CleanAir Project, CNN & MLP to predict UFP (UltraFine Particles concentration) based on geospatial pollution measurements in a city

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors