text.lower().split().count clear solution for Popular Words by flpo - python coding challenges

Enable Javascript in your browser and then refresh this page, for a much enhanced experience.

text.lower().split().count solution in Clear category for Popular Words by flpo

def popular_words(text, words):
    lower_count = text.lower().split().count
    return {word: lower_count(word) for word in words}

April 12, 2018

Comments:

alexander.gorelyshev on April 24, 2018, 11:42 a.m.
Creating a function object to do multiple transformations and then applying it to a dictionary comprehension... Very clever, I've learned from you :) It's probably worth pointing out, though, that this way we repeat not only (1) counting, but also (2) splitting and (3) bringing to lower case the whole input `text` each time we iterate over the list of `words`... If `text` is, in fact, a big corpus, this won't be a cheap solution.

flpo on April 26, 2018, 10 a.m.
Hi & thanks for the compliment :) BTW, what you are pointing out is not true: only (1) counting is repeated multiple times (inside the comprehension loop), but (2) splitting and (3) bringing to lower case have been applied only once (before the comprehension loop). Your comment would be true with the following definition of `lower_count`: <pre class='brush: python'>lower_count = lambda x: text.lower().split().count(x) return {word: lower_count(word) for word in words} </pre> That is equivalent to the one-line: <pre class='brush: python'>return {word: text.lower().split().count(word) for word in words} </pre> But my implementation is different: `text.lower().split().count` is not a composition of functions, instead it is the bound method `list.count` of the list object returned by `.split`, that is called only once, on the `str` object returned by `.lower`, that is called once too. You can unroll it as: <pre class='brush: python'>text_lower = text.lower() text_lower_words = text_lower.split() lower_count = text_lower_words.count </pre> Let's consider the following example: <pre class='brush: python'>text = 'a B A c' lower_count = text.lower().split().count </pre> As you said, `lower_count` is a function object (more precisely, a builtin method): <pre class='brush: python'>import types isinstance(lower_count, types.BuiltinFunctionType) # True </pre> But it is also bounded to the list instance `['a', 'b', 'a', 'c']`: <pre class='brush: python'>lower_count.__self__ # ['a', 'b', 'a', 'c'] </pre> You can also double-check that (2) and (3) were called multiple times with the one-line version but only once with my implemetation using `cProfile`: <pre class='brush: python'>def oneliner(text, words): return {word: text.lower().split().count(word) for word in words} def flpo(text, words): lower_count = text.lower().split().count return {word: lower_count(word) for word in words} import cProfile text = 'a B A c' words = ['a', 'b', 'c', 'd'] cProfile.run('oneliner(text, words)') cProfile.run('flpo(text, words)') # output filtered (only interesting lines) # 'oneliner(text, words)' ncalls tottime percall cumtime percall filename:lineno(function) 4 0.000 0.000 0.000 0.000 {method 'count' of 'list' objects} 4 0.000 0.000 0.000 0.000 {method 'lower' of 'str' objects} 4 0.000 0.000 0.000 0.000 {method 'split' of 'str' objects} # 'flpo(text, words)' ncalls tottime percall cumtime percall filename:lineno(function) 4 0.000 0.000 0.000 0.000 {method 'count' of 'list' objects} 1 0.000 0.000 0.000 0.000 {method 'lower' of 'str' objects} 1 0.000 0.000 0.000 0.000 {method 'split' of 'str' objects} </pre>

Nour on May 3, 2018, 6:08 p.m.
I realy appreciate your answer, I didn't know that Python have this realization However still not a fan of this approch for big strings since it is a lot better to have O(n) + O(m) then O(n*m) Thanks for your answer.

a1exstr on May 8, 2018, 8:10 p.m.
Preparing a method as an object and then using it is really awesome :) Thanks for the detailed explanation. It did help to understand your solution. I myself came up with the comprehension method, but I put count() into the comprehension itself

myname on July 22, 2018, 12:18 p.m.
Could you please provide an example of solution with complexity O(n) + O(m) ? In my understanding it is not as trivial because to get get of an element from dict we spent log(m) so we are talking about log(m) * n complexity

hugoboss on Aug. 5, 2018, 3:52 p.m.
Yeeeaah! I love this method! Clear AND clever. Something like compiling a 3-step function. But more subtle with 'count' method.

Sim0000 on Aug. 20, 2018, 7:17 p.m.
Nice one!

alekseykonotop on Sept. 19, 2018, 9:08 a.m.
I think this is the best solution. Brief and clear.

pawelek.miecz on Oct. 7, 2018, 5:13 p.m.
nice solution :)

Moudud_Tonmoy on Oct. 15, 2018, 5:27 p.m.
wow!clear and clever...

domgregori on Nov. 14, 2018, 8:50 p.m.
Like your answer but I think using re.split(r'(\s)', text.lower()).count would be better. If you look at the test cases, there are words split by newlines(\n). Checkio just happened to not use beginning/end words for the check.

koborC on Nov. 16, 2020, 12:29 p.m.
str.split() ignores newline symbols, <pre class='brush: python'>>>> a = "text \n with\n newline." >>> print(a) text with newline. >>> a.split() ['text', 'with', 'newline.'] </pre> because \n is a white space by definition. <pre class='brush: python'>>>> import string >>> string.whitespace ' \t\n\r\x0b\x0c' </pre>

caroft on Aug. 27, 2019, 7:51 p.m.
Concise and clean, I like it! I didn't know we could do that, thanks for the explanations.

accnext1997 on Oct. 13, 2019, 11:45 a.m.
Nice solution)

ProgrammerBruce on April 4, 2020, 6:35 p.m.
Prefer casefold() instead of lower(). Also, casefold the keys (in words) for best matching.

boormistr on July 5, 2020, 12:08 p.m.
Beautiful!

Muffinlavania404 on July 17, 2020, 2:47 p.m.
Geez.. thats good

lakshaykarani on March 8, 2021, 1:47 a.m.
How much time did you guys took to solve this? Just curious!

tkachuk.constantine on May 4, 2021, 12:37 a.m.
Thanks for your time and your willingness to share!!!

stroet on March 7, 2022, 9:36 p.m.
that's pretty awesome, thank's for sharing your knowledge. So it could be done in one line: return {: text.lower().split().count() for _ in words} I didn't know, that count could be used after split().

Look up in the Dict()

0 %

PyCharm The Python IDE for Professional Developers. Enjoy productive Python, web and scientific development with PyCharm. Download it now!