KeyError exception in List Comprehension

Question

I have the following function:

def calculate(blob, count_per_data):
   return geometric_mean( [score_per_count[ count_per_data[data] ] for data in combinations(blob)] )

The problem with my code is that if data is not found in count_per_data I get an exception. Instead, I wish count_per_data["unknown"] to evaluate to 0, i.e. "unknown"'s count is 0.

In turn, the value 0 exists in score_per_count and is not equal to 0. In other words, the score associated with a count of 0 is not itself 0.

How would you recommend I fix the above code to achieve my goal?

Answer 1

If you want to make sure that, the data exists in count_per_data and the value of count_per_data exists in score_per_count, you can use the list comprehension as a filter, like this

return geometric_mean([score_per_count[count_per_data[data]] for data in  combinations(blob) if data in count_per_data and count_per_data[data] in score_per_count])

More readable version,

return geometric_mean([score_per_count[count_per_data[data]]
     for data in combinations(blob)
     if data in count_per_data and count_per_data[data] in score_per_count])

But, if you want to use default values when key is not found in a dictionary, then you can use dict.get. Quoting from dict.get docs,

get(key[, default])

Return the value for key if key is in the dictionary, else default. If default is not given, it defaults to None, so that this method never raises a KeyError.

You can use it, like this

count_per_data.get(data, 0)

If data is not found in count_per_data, 0 will be used instead.