如何在Dict {key：List}中找到最小值？

Question

I am trying to find the minimum flight price in a dictionary that looks like:

{datetime.date(2018, 12, 16): ['KL', 'AMS', '59.84'],
  datetime.date(2018, 12, 24): ['KL', 'AMS', '59.84'],
  datetime.date(2018, 12, 25): ['KL', 'AMS', '59.84'],
  datetime.date(2018, 12, 27): ['KL', 'AMS', '59.84'],
  datetime.date(2018, 12, 18): ['KL', 'AMS', '59.84'],
  datetime.date(2018, 12, 19): ['KL', 'AMS', '59.84'],
  datetime.date(2018, 12, 20): ['KL', 'AMS', '59.84'],
  datetime.date(2018, 12, 23): ['KL', 'AMS', '59.84'],
  datetime.date(2018, 12, 17): ['AF', 'AMS', '70.24'],
  datetime.date(2018, 12, 21): ['SK', 'AMS', '97.93'],
  datetime.date(2018, 12, 26): ['SK', 'AMS', '97.93'],
  datetime.date(2018, 12, 28): ['SK', 'AMS', '97.93'],
  datetime.date(2019, 1, 2): ['SK', 'AMS', '97.93'],
  datetime.date(2019, 1, 3): ['SK', 'AMS', '97.93'],
  datetime.date(2018, 12, 29): ['AF', 'AMS', '111.64'],
  datetime.date(2018, 12, 31): ['EW', 'AMS', '127.51'],
  datetime.date(2019, 1, 1): ['EW', 'AMS', '127.51'],
  datetime.date(2018, 12, 30): ['EW', 'AMS', '147.51'],
  datetime.date(2018, 12, 22): ['KL', 'AMS', '148.84']}

The first key is a datetime, and value = list of info. I'm trying to simply get the max and min price over that time series. I've tried the following:

_tempPairAndPrices = {}
for item in permutations(destinations[1:],2):    
    #try:
    key_max = max(_fltDictTemp_CachedData[item].keys(),key=(lambda k: _fltDictTemp_CachedData[item][k]))
    key_min = min(_fltDictTemp_CachedData[item].keys(),key=(lambda k: _fltDictTemp_CachedData[item][k]))
    maxPx = _fltDictTemp_CachedData[item][key_max]
    minPx = _fltDictTemp_CachedData[item][key_min]

    _tempPairAndPrices[item] = {'maxPx':maxPx,'minPx':minPx}

I don't think it's properly accessing the list of information though, as it's not properly assigning max and min values... (often my min >> max)

Thanks!

Answer 1

IIUC, use min() with a key

[min(v.values(), key=lambda x: float(x[-1])) for k,v in d.items()]

Outputs

[['KL', 'AMS', '59.84']]

---

For your edited data, just do

min(d.values(), key=lambda x: float(x[-1]))

Answer 2

If you are just interested in getting the minimum flight price, this is one possible solution:

prices = [float(i) for i in np.array(dict.values()[0].values())[:,-1]]
min_price = min(prices)

where dict is your dictionary. The reason I used np.array() in the list comprehension was in order to use the indexing [:,-1] since the prices are the last element of the list. But I see there are several prices which are equal and the minimum. I don't know how you plan to tackle them. You can of course combine the above two lines into one and write:

min_price = min([float(i) for i in np.array(dict.values()[0].values())[:,-1]])