CODEWITHSUNDEEP
python
AMayer
AMayer

Reputation: 425

How to get only the first elements from each items of a list

I have to following list:

[[(640, 92, 0.80969274, 0)], [(524, 180, 0.58553714, 1)], [(494, 181, 0.64085102, 2)], [(477, 338, 0.67508131, 3)], [(554, 480, 0.64356369, 4)]]

I'm interested only in the first 2 elements in the items, for example:

640, 92
524, 180
494, 181

I can get them using

print list[1][0][0:2]

But what I want to do is to take max and mean by column. First I want to take the max and min for the first column, and then max and min from the second column.

in matlab we can do something like: 

max(list(:,1));
min(list(:,1));
max(list(:,2));
min(list(:,2)));

Is there a way to do this in python without any headache?

Thank you!

Upvotes: 2

Views: 284

Answers (3)

juanpa.arrivillaga
juanpa.arrivillaga

Reputation: 96350

As the other answers have pointed out, fundamentally, you are not using a good data structure for your task. If you want to similar functionality to Matlab in Python, you need to use numpy and should take a look at pandas. Python is a general purpose programming language, so its sequence types like list are not really meant to efficiently do the same things as a Matlab mutli-dimensional array (which is what numpy gives you). But that being said, there is a relatively painless way of doing what you want. It would be made even more painless if instead of using a list of length 1 lists, each containing a tuple that actually holds your data, you just use a list of tuples. But, python makes this sort of munging pretty painless:

>>> x = [[(640, 92, 0.80969274, 0)], [(524, 180, 0.58553714, 1)], [(494, 181, 0.64085102, 2)], [(477, 338, 0.67508131, 3)], [(554, 480, 0.64356369, 4)]]
>>> extracted = [sub[0:2] for l in x for sub in l]

In the list comprehension above, I just iterate over the list, then iterate over the (length one) list wrapping the tuple, then slice the columns I want from the tuple (sub).

Now, to do a "transpose" in vanilla python, unpack into zip:

>>> for col in zip(*extracted):
...     print("Max: {}".format(max(col)))
...     print("Min: {}".format(min(col)))
...
Max: 640
Min: 477
Max: 480
Min: 92
>>>

By the way, using Python 3.6 makes things even more painless by giving us f-strings!:

>>> for col in zip(*extracted):
...     print(f"Max: {max(col)}")
...     print(f"Min: {min(col)}")
...
Max: 640
Min: 477
Max: 480
Min: 92
>>>

Upvotes: 3

Arya McCarthy
Arya McCarthy

Reputation: 8814

It's a list of lists (of length 1) of tuples, so we have to do just a bit of legwork.

import pandas as pd

df = pd.DataFrame([d[0] for d in data])
df[0].max()
df[1].max()
df[0].mean()
df[0].min()

Tons of operations are supported.

Upvotes: 3

mohammad
mohammad

Reputation: 2208

max([x[0][0] for x in list])
min([x[0][0] for x in list])

This takes max and min for the first column. similarly for the second column you can do this :

max([x[0][1] for x in list])
min([x[0][1] for x in list])

Upvotes: -1

Related Questions