Finding the difference in values of matching keys in two dictonaries in python3-CodePudding

So I have two dictionaries that store specific "channels" and their values, these channels update every second. My end goal is to create a tool/GUI that will help us determine how off or misaligned our channels are, they correspond to optical mount DOFs.

I currently have a function that grabs the current time and determines the channels value at that time and puts it in a dictionary and another function that lets you select a time and then does the same.

The part of this project I'm a bit lost on is how to compare these dictionaries.

I want to match the keys (which are my channels) and then find the difference between their two values and store that in a 3rd dictionary with the same channels.

So here's the basic idea in code for what I'm looking to do:

diff_channels_dict = {}
### After my functions that create these dictionaries run, I'll have something like the following, but a lot longer

ref_channels_dict = {

    'IM1_M1_P_OFFSET': 8,
    'IM1_M1_Y_OFFSET': 8,
    'IM2_M1_P_OFFSET': 8}
now_channels_dict = {

    'IM1_M1_P_OFFSET': 2,
    'IM1_M1_Y_OFFSET': 2,
    'IM2_M1_P_OFFSET': 2}


def some_func():

    for key in now_channels_dict and ref_channels_dict:
        diff = now_channels_dict[value] - ref_channels_dict[value]
        diff_channels_dict[key].append(diff)
    return

some_func()

### So then
diff_channels_dict = {

    'IM1_M1_P_OFFSET': -6,
    'IM1_M1_Y_OFFSET': -6,
    'IM2_M1_P_OFFSET': -6}

Can anyone give me some guidance on how to go about doing this? I looked into a few libraries, deepdiff, and Dictdiffer but I do not have access to them with the env that is used on all the machines that will use this tool.

CodePudding user response：

I'm going to assume that the starting dictionaries are like this (written in python format):

ref_channels_dict = {
            'IM1_M1_P_OFFSET': 8,
            'IM1_M1_Y_OFFSET': 8,
            'IM2_M1_P_OFFSET': 8
            }
            
now_channels_dict = {
            'IM1_M1_P_OFFSET': 2,
            'IM1_M1_Y_OFFSET': 2,
            'IM2_M1_P_OFFSET': 2
            }

From this, we can write the function that compares and put the differences into a new dictionary:

def compare_dictionaries(ref_dict, now_dict):
    assert ref_dict.keys() == now_dict.keys() # check if the keys are the same

    diff_dict = {}

    for key in ref_dict.keys(): #iterate over the keys and fill the diff_dict with the differences
        # this can be done because we know that the keys in both dictionaries are the same
        diff_dict[key] = now_dict[key] - ref_dict[key]

    return diff_dict

After that, we can define the whole code as:


def compare_dictionaries(ref_dict, now_dict):
    assert ref_dict.keys() == now_dict.keys() # check if the keys are the same

    diff_dict = {}

    for key in ref_dict.keys(): #iterate over the keys and fill the diff_dict with the differences
        # this can be done because we know that the keys in both dictionaries are the same
        diff_dict[key] = now_dict[key] - ref_dict[key]

    return diff_dict





if __name__ == "__main__":
    ref_channels_dict = {
            'IM1_M1_P_OFFSET': 8,
            'IM1_M1_Y_OFFSET': 8,
            'IM2_M1_P_OFFSET': 8
            }

    now_channels_dict = {
            'IM1_M1_P_OFFSET': 2,
            'IM1_M1_Y_OFFSET': 2,
            'IM2_M1_P_OFFSET': 2
            }

    print(compare_dictionaries(ref_channels_dict, now_channels_dict))

Which outputs:

>>> {'IM1_M1_P_OFFSET': -6, 'IM1_M1_Y_OFFSET': -6, 'IM2_M1_P_OFFSET': -6}

CodePudding user response：

with pandas you can do it without looping, what could be quite useful for really large dataset:

import pandas as pd

diff_dict = (pd.Series(now_channels_dict) - pd.Series(ref_channels_dict)).to_dict()

print(diff_dict)  
'''
{'IM1_M1_P_OFFSET': -6, 'IM1_M1_Y_OFFSET': -6, 'IM2_M1_P_OFFSET': -6}