This is a simple utility for regression of SATVAM gas sensor data. Data is ingested by specifying paths to the files containing the reference monitor data and the data from the SATVAM gas sensors. Time alignment is taken care of by the script.
matplotlib >= 3.0.2 <br> sklearn <br> numpy
Run the following command on a BASH terminal:
python autoreg.py ref-monitor.xlsx ebam.csv sens1.csv sens2.csv sens3.csv output-prefix
# replace ref-monitor.xlsx with Excel file directly downloaded
# from the reference monitor and ebam.csv with the EBAM data file
# replace sens*.csv with CSV files of the sensors
# directly downloaded from Graphana
# replace outfile with the output file prefix
To change the configuration switches to pertain to your desired deployment, do the following in the file autoreg.py
DEPLOY_SITE='mpcb'
DEPLOYMENT=2
# the DEPLOYMENT switch indicates if the deployment is 1st (Aug/Sept '18)
# or 2nd (Dec '18/current)
# the DEPLOY_SITE indicates if the site is MPCB or MRU
The output PDFs will be generated starting with the prefix specified in the command above, one each for NO2, O3 and PM2.5 inference plots