Linear Relation Visualization feature request

[apoorvagnihotri]

linearRelation

I think adding the following features would help better understand the plot.

• What are the blue dots representing? A legend would help with this.
• Xlabel should instead be the actual years.
• The left ylabel could instead be Pollutant ratio (SO2/NOX).
  If that is what is meant.

@prerna-khanna Hi Prerna, can you please take this up?

[apoorvagnihotri]

Currently, I am not sure about the correctness of the code here. The new linearRelation generates the following plot.

If the line is just the average of the then why should we have 2 y-ticks?

In addition to that, the lines that I have highlighted below also need to be checked from openAir and addressed accordingly.

vayu/vayu/linearRelation.py

Lines 58 to 65 in e006917

	pol_1 = df_new[pol1].mean()
	# this might not be entirely correct.
	# TODO: we want to take the average of the slope.
	# [Not take the average and find the slope]
	# Additionally, we would like to have the option of
	# changing the granularity.
	pol_2 = df_new[pol2].mean()
	values.append(pol_2 / pol_1)

Additionally, do we fit a linear relation of just take the ratio of the two pollutants? I think we need to fit a linear relation and then find the slope.

Original LinearRelation - https://github.com/davidcarslaw/openair/blob/master/R/linearRelation.R