I have a nested dictionary of the form:

 {'2015-01-01': {'time': '8', 'capacity': '5'}, '2015-01-02': {'time': '8', 'capacity': '7'},'2015-01-03': {'time': '8', 'capacity': '8'} etc}

The dictionary is created from a csv file using dictreader. What I would like to be able to do is return the maximum value of capacity. So in this case 8.

I can use:

for k,v in input_dict.items():if temp_max < int(v['capacity']):temp_max = int(v['capacity'])

which works but I wondered if there was a neater method? I've searched and found methods to extract the top level key associated with the maximum value which is of course not what I need. See below:

max(input_dict, key=lambda v: input_dict[v]['capacity'])

which would return '2015-01-03', So I imagine there is a simple mod to the above one liner that will give me what I need but is has me stumped!

Any ideas?

You want

max(int(d['capacity']) for d in input_dict.values())

Explanation:

If you don't care about the keys, just iterate over the nested dicts (IOW the outer dict's values)

Also your inner dicts "capacity" values are stored as strings, I assume you want to test the integer value. To find out the difference check this:

>>> max(["1", "5", "18", "01"])
'5'
>>> max([1, 5, 18, 01])
18
>>>