Create a list in which each element is a list with a name

Question

before you all tell me to search for stuff, let me tell you "I HAVE".

I have searched quite a bit. Forums, educational/training sites, stackexchange and all the regulars. I haven't been able to get what I need. I am not running to SE at the first sign of trouble either. I have registered on SE quite a while ago and this is my first question.

I have found some replies(suggestions, not solutions) and some ANSWERS(sort of) too. However, the most useful thing I have found is https://stackoverflow.com/a/6181978/15065496

I do know what dictionaries are, I know how to use them. I have tried the answer above too, but it isn't exactly what I want.

I am a mechanical engineer learning python to solve some engineering problems that I am having. Please read till the end to get a better idea of what I am trying to achieve.

I need to create a list of lists with names. I have been able to create a list of lists, which looks like this:

list_xyz = []
for i in range(1,5,1):
    temp_array_name = "list_" + str(i)
    list_xyz.append(temp_array_name)

>>> list_xyz
['list_1', 'list_2', 'list_3', 'list_4']

Now, I want to create a new list out of each of the elements.

Basically, what I want is to create lists called list_1, list_2, list_3, list_4. I want to automate the task of doing

list_1 = []
list_2 = []
list_3 = []
list_4 = []

what I have already tried are:

for i in list_xyz:
    i = list()

*this gives:*

>>> list_xyz
['list_1', 'list_2', 'list_3', 'list_4']

*I thought good*. Then I typed:

>>> list_1

*I was expecting the following *
[]

*Instead, I got:*
Traceback (most recent call last):
  File "<pyshell#93>", line 1, in <module>
    list_1
NameError: name 'list_1' is not defined


*Then I tried the following and understood what had happened.*
>>> i
[]

*So then, I tried:*

for i in range(0,len(list_xyz),1):
    list_xyz[i] = list()

*this gave:*

>>> list_xyz
[[], [], [], []]

Now, technically I could use the elements of list_xyz, but calling them would be a pain. If I need the third element inside the second element of list_xyz, then I would have to say list_xyz[1][2]

Instead, if they were named, I could just say list_2[2] and could get the element I was looking for.

I have also tried using dictionaries, as suggested by https://stackoverflow.com/users/455276/the-wolf in https://stackoverflow.com/a/6181978/15065496.

What I have tried is:

>>> dict1 = {}
>>> list_pqr = ['list_1', 'list_2', 'list_3', 'list_4']
>>> for i in range(0,len(list_pqr),1):
    dict1[list_pqr[i]] = list()

    
>>> dict1
{'list_3': [], 'list_2': [], 'list_1': [], 'list_4': []}

This is close to what I want but isn't exactly what I want. Please read the following to understand exactly what I want.

The reason I need this particular list this exact way is: I am trying to calculate some temperatures at different times at different positions. I want to create a dictionary with times as keys and a list of temperatures at various locations as the values.

and another dictionary with locations as keys and list temperatures at different times as the values.

What I want is:

dict1 = {time1: temps_at_time_1
time2: temps_at_time_2
time3: temps_at_time_3
time4: temps_at_time_4
time5: temps_at_time_5
}

dict2 = {loc1: temps_at_times_L1
loc2: temps_at_times_L2
loc3: temps_at_times_L3
loc4: temps_at_times_L4
loc5: temps_at_times_L5
}

where temps_at_times_L1 is the list of temperatures at location 1 at various times. where temp_at_time_1 is the list of temperatures at Time 1 at various locations.

let's say there are 25 locations and 100 timesteps, then:

len(temps_at_times_L1) = 100
len(temp_at_time_1) = 25

now, if temps_at_times_L1 and temp_at_time_1 are variables that I could call, It would make it very easy for me to plot the temperatures etc.

Basically, I want the same thing that I have achieved with dictionaries after using the wolf's suggestion, but the lists created there should have names that I could call so that when calculating temperatures, I could just update the lists by calling the append() method. If I am calculating timestep1, I wanna be able to just say:

while(timestep == 1):
   for i in temperature_location_array:
       temps_at_time_1.append(i)

Even better would be if i was able to do the following:

for a in range(startTime, endTime, stepSize):
while(timestep == a):
   for i in temperature_location_array:
       temps_at_time_a.append(i)       or      vars("temps_at_time_" + str(a)).append(i)

I'm pretty sure the last line will not work and I will have to rack my brains again for that. I have tried reading about how vars() works but it is giving me a headache.

If you guys can't help me, and/or tell me that the last line of my code is stooooopid, then I'll just have to make do with the basic dictionary and ridiculously complicated(for my level anyway) indexing and calling and assigning directly to the values dictionaries.

I have previously faced the same need whilst editing a spreadsheet using openpyxl, but was unable to create lists from variable names inside another list. I gave up and used another method that got the job done. However, this time I would like to know how to do it: if at all it can be done.

Thank you all for the trouble of reading through my long winded explanation. I did this so people who reply will understand exactly what I need and will not give suggestions that would not serve my purpose. I will read all suggestions and take all useful ones into account though. I am not trying to offend/insult anyone.

Answer 1

Sorry for the negativity but appending numbers to variables is the exact opposite what any programmer should be doing - nested_lists[1][2] is the correct way of handling nested lists, nested_list_1[2] is not. Do not write data into the names of variables, just don't, it won't end well and globals() is terrible misuse here.

What you need is some proper structure that will give meaning to your data. I would recommend pandas library. It's like Excel for Python.

Let's take the following example of 5 measurements in two locations foo and bar at various times.

# Assuming the raw data is in format:
#  (time, temperature, location)
import pandas as pd
data=pd.DataFrame({
    (1,12, 'foo'),
    (2,12,'bar'),
    (10,23,'foo'),
    (3,12,'foo'),
    (3,15,'bar')
    },columns=['time','temperature','location'])
print(data)

yields:

   time  temperature location
0     1           12      foo
1     2           12      bar
2     3           15      bar
3    10           23      foo
4     3           12      foo

I cannot enumerate all the things you can do with a dataframe but it can be sorted, filtered, mapped, indexed... see the documentation or search for answers here.

You might be interested in e.g. filtering data by location data[data['location']=='foo']

   time  temperature location
0     1           12      foo
3    10           23      foo
4     3           12      foo

or setting the dataframe's index to time column and indexing using it:

new_data = data.set_index('time')
print(new_data.loc[3])

      temperature location
time                      
3              15      bar
3              12      foo

Answer 2

As several commenters are mentioned before me you actually can create variables programatically, but is a very ugly way of solving a problem.

Anyway, you can add variables to the global namespace with globals() as follows:

>>> list_xyz
['list_1', 'list_2', 'list_3', 'list_4']
>>> for n in list_xyz:
...     globals()[n] = list()
... 
>>> list_1
[]
>>> list_2
[]
>>> list_3
[]
>>> list_4
[]
>>>