How to use map(lambda x) with 3 conditions-CodePudding

How can I map the conditions in lambda like the code below:

map(lambda example: {'labels': 1 if example['sentiment'] == 'positive', 
                               2 if example['sentiment'] == 'negative'
                               else 0}

My dataset is:

sentiment
positive
negative
neutral

OBS: I know I can use get_dummies but I need to do this with map lambda, please.

CodePudding user response：

This is not a good approach, but if you want the code this should work


map(lambda example: {'labels': 1 if example['sentiment'] == 'positive' else 
                               (2 if example['sentiment'] == 'negative' else 0)}

CodePudding user response：

Whenever you have a bunch of simple if/else, it is almost always an hint that you should use a dict:

>>> labels_mapping = {"positive": 1, "negative": 2, "neutral": 0}

>>> data = ["positive", "negative", "neutral", "positive"]
>>> data_labels = list(map(lambda example: labels_mapping[example], data))
>>> data_labels
[1, 2, 0, 1]

Actually, you then didn't need a lambda anymore, because you can use the __getitem__ method of the constructed dict:

>>> list(map(labels_mapping.__getitem__, data))
[1, 2, 0, 1]

And to catch unexpected values (similar to the last else clause of the corresponding if/else statement), you can use a defaultdict to assign a default value:

>>> from collections import defaultdict

>>> labels_mapping = defaultdict(lambda: -1, {"positive": 1, "negative": 2, "neutral": 0})

>>> data = ["positive", "negative", "neutral", "positive", "noise"]
>>> data_labels = list(map(labels_mapping.__getitem__, data))
>>> data_labels
[1, 2, 0, 1, -1]