Less memory intensive way to parse large JSON file in Python - python

Here is my code
import json
data = []
with open("review.json") as f:
for line in f:
data.append(json.loads(line))
lst_string = []
lst_num = []
for i in range(len(data)):
if (data[i]["stars"] == 5.0):
x = data[i]["text"]
for word in x.split():
if word in lst_string:
lst_num[lst_string.index(word)] += 1
else:
lst_string.append(word)
lst_num.append(1)
result = set(zip(lst_string, lst_num))
print(result)
with open("set.txt", "w") as g:
g.write(str(result))
I'm trying to write a set of all words in reviews that were given 5 stars from a pulled in json file formatted like
{"review_id":"Q1sbwvVQXV2734tPgoKj4Q","user_id":"hG7b0MtEbXx5QzbzE6C_VA","business_id":"ujmEBvifdJM6h6RLv4wQIg","stars":1.0,"useful":6,"funny":1,"cool":0,"text":"Total bill for this horrible service? Over $8Gs. These crooks actually had the nerve to charge us $69 for 3 pills. I checked online the pills can be had for 19 cents EACH! Avoid Hospital ERs at all costs.","date":"2013-05-07 04:34:36"}
{"review_id":"GJXCdrto3ASJOqKeVWPi6Q","user_id":"yXQM5uF2jS6es16SJzNHfg","business_id":"NZnhc2sEQy3RmzKTZnqtwQ","stars":1.0,"useful":0,"funny":0,"cool":0,"text":"I *adore* Travis at the Hard Rock's new Kelly Cardenas Salon! I'm always a fan of a great blowout and no stranger to the chains that offer this service; however, Travis has taken the flawless blowout to a whole new level! \n\nTravis's greets you with his perfectly green swoosh in his otherwise perfectly styled black hair and a Vegas-worthy rockstar outfit. Next comes the most relaxing and incredible shampoo -- where you get a full head message that could cure even the very worst migraine in minutes --- and the scented shampoo room. Travis has freakishly strong fingers (in a good way) and use the perfect amount of pressure. That was superb! Then starts the glorious blowout... where not one, not two, but THREE people were involved in doing the best round-brush action my hair has ever seen. The team of stylists clearly gets along extremely well, as it's evident from the way they talk to and help one another that it's really genuine and not some corporate requirement. It was so much fun to be there! \n\nNext Travis started with the flat iron. The way he flipped his wrist to get volume all around without over-doing it and making me look like a Texas pagent girl was admirable. It's also worth noting that he didn't fry my hair -- something that I've had happen before with less skilled stylists. At the end of the blowout & style my hair was perfectly bouncey and looked terrific. The only thing better? That this awesome blowout lasted for days! \n\nTravis, I will see you every single time I'm out in Vegas. You make me feel beauuuutiful!","date":"2017-01-14 21:30:33"}
{"review_id":"2TzJjDVDEuAW6MR5Vuc1ug","user_id":"n6-Gk65cPZL6Uz8qRm3NYw","business_id":"WTqjgwHlXbSFevF32_DJVw","stars":1.0,"useful":3,"funny":0,"cool":0,"text":"I have to say that this office really has it together, they are so organized and friendly! Dr. J. Phillipp is a great dentist, very friendly and professional. The dental assistants that helped in my procedure were amazing, Jewel and Bailey helped me to feel comfortable! I don't have dental insurance, but they have this insurance through their office you can purchase for $80 something a year and this gave me 25% off all of my dental work, plus they helped me get signed up for care credit which I knew nothing about before this visit! I highly recommend this office for the nice synergy the whole office has!","date":"2016-11-09 20:09:03"}
{"review_id":"yi0R0Ugj_xUx_Nek0-_Qig","user_id":"dacAIZ6fTM6mqwW5uxkskg","business_id":"ikCg8xy5JIg_NGPx-MSIDA","stars":1.0,"useful":0,"funny":0,"cool":0,"text":"Went in for a lunch. Steak sandwich was delicious, and the Caesar salad had an absolutely delicious dressing, with a perfect amount of dressing, and distributed perfectly across each leaf. I know I'm going on about the salad ... But it was perfect.\n\nDrink prices were pretty good.\n\nThe Server, Dawn, was friendly and accommodating. Very happy with her.\n\nIn summation, a great pub experience. Would go again!","date":"2018-01-09 20:56:38"}
{"review_id":"yi0R0Ugj_xUx_Nek0-_Qig","user_id":"dacAIZ6fTM6mqwW5uxkskg","business_id":"ikCg8xy5JIg_NGPx-MSIDA","stars":5.0,"useful":0,"funny":0,"cool":0,"text":"a b aa bb a b","date":"2018-01-09 20:56:38"}
but it is using all the memory on my computer before it can output into a text file. How can I use a less memory intensive way?

Only get text where stars == 5:
Data:
Based on the question, the data is a file containing rows of dicts.
Get the text into a list:
Given the data from Yelp Challenge, getting the 5 stars text into a list, doesn't take that much memory.
The Windows resource manager showed an increase of about 1.3GB, but the object size of text_list was about 25MB.
import json
text_list = list()
with open("review.json", encoding="utf8") as f:
for line in f:
line = json.loads(line)
if line['stars'] == 5:
text_list.append(line['text'])
print(text_list)
>>> ['Test text, example 1!', 'Test text, example 2!']
Extra:
Everything after loading the data, seems to require a lot of memory that isn't being released.
When cleaning the text, Windows resource manager went up by 16GB, though the final size of clean_text was also only about 25MB.
Interestingly, deleting clean_text does not release the 16GB of memory.
In Jupyter Lab, restarting the Kernel will release the memory
In PyCharm, stopping the process also releases the memory
I tried manually running the garbage collector, but that didn't release the memory
Clean text_list:
import string
def clean_string(value: str) -> list:
value = value.lower()
value = value.translate(str.maketrans('', '', string.punctuation))
value = value.split()
return value
clean_text = [clean_string(item) for item in text_list]
print(clean_text)
>>> [['test', 'text', 'example', '1'], ['test', 'text', 'example', '2']]
Count words in clean_text:
from collection import Counter
words = Counter()
for item in clean_text:
words.update(item)
print(words)
>>> Counter({'test': 2, 'text': 2, 'example': 2, '1': 1, '2': 1})

Related

Improve speed/performance of web-scraping with lots of exceptions

I've written some web-scraping code that is currently working, however quite slow. Some background: I am using Selenium as it requires several stages of clicks and entry, along with BeautifulSoup. My code is looking at a list of materials within subcategories on a website(image below) and scraping them. If the material scraped from the website is one of the 30 I am interested in (lst below), then it writes the number 1 to a dataframe which I later convert to an Excel sheet.
The reason it is so slow, I believe anyway, is due to the fact that there are a lot of exceptions. However, I am not sure how to handle these besides try/except. The main bits of code can be seen below, as the entire piece of code is quite lengthy. I have also attached an image of the website in question for reference.
lst = ["Household cleaner and detergent bottles", "Plastic milk bottles", "Toiletries and shampoo bottles", "Plastic drinks bottles",
"Drinks cans", "Food tins", "Metal lids from glass jars", "Aerosols",
"Food pots and tubs", "Margarine tubs", "Plastic trays","Yoghurt pots", "Carrier bags",
"Aluminium foil", "Foil trays",
"Cardboard sleeves", "Cardboard egg boxes", "Cardboard fruit and veg punnets", "Cereal boxes", "Corrugated cardboard", "Toilet roll tubes", "Food and drink cartons",
"Newspapers", "Window envelopes", "Magazines", "Junk mail", "Brown envelopes", "Shredded paper", "Yellow Pages" , "Telephone directories",
"Glass bottles and jars"]
def site_scraper(site):
page_loc = ('//*[#id="wrap-rlw"]/div/div[2]/div/div/div/div[2]/div/ol/li[{}]/div').format(site)
page = driver.find_element_by_xpath(page_loc)
page.click()
driver.execute_script("arguments[0].scrollIntoView(true);", page)
soup=BeautifulSoup(driver.page_source, 'lxml')
for i in x:
for j in y:
try:
material = soup.find_all("div", class_ = "rlw-accordion-content")[i].find_all('li')[j].get_text(strip=True).encode('utf-8')
if material in lst:
df.at[code_no, material] = 1
else:
continue
continue
except IndexError:
continue
x = xrange(0,8)
y = xrange(0,9)
p = xrange(1,31)
for site in p:
site_scraper(site)
Specifically, the i's and j's rarely go to 6,7 or 8 but when they do, it is important that I capture that information too. For context, the i's correspond to the number of different categories in the image below (Automative, Building materials etc.) whilst the j's represent the sub-list (car batteries and engine oil etc.). Because these two loops are repeated for all 30 sites for each code, and I have 1500 codes, this is extremely slow. Currently it is taking 6.5 minutes for 10 codes.
Is there a way I could improve this process? I tried list comprehension however it was difficult to handle errors like this and my results were no longer accurate. Could an "if" function be a better choice for this and if so, how would I incorporate it? I also would be happy to attach the full code if required. Thank you!
EDIT:
by changing
except IndexError:
continue
to
except IndexError:
break
it is now running almost twice as fast! Obviously it is best to exit to loop after it fails once, as the later iterations will also fail. However any other pythonic tips are still welcome :)
It sounds like you just need the text of those lis:
lis = driver.execute_script("[...document.querySelectorAll('.rlw-accordion-content li')].map(li => li.innerText.trim())")
Now you can use those for your logic:
for material in lis:
if material in lst:
df.at[code_no, material] = 1

The usage of radix sort

Let's say i have a file containing data on users and their favourite movies.
Ace: FANTASTIC FOUR, IRONMAN
Jane: EXOTIC WILDLIFE, TRANSFORMERS, NARNIA
Jack: IRONMAN, FANTASTIC FOUR
and based of that, the program I'm about to write returns me the name of the users that likes the same movies.
Since Ace and Jack likes the same movie, they will be partners hence the program would output:
Movies: FANTASTIC FOUR, IRONMAN
Partners: Ace, Jack
Jane would be exempted since she doesn't have anyone who shares the same interest in movies as her.
The problem I'm having now is figuring out on how Radix Sort would help me achieve this as I've been thinking whole day long. I don't have much knowledge on radix sort but i know that it compares elements one by one but I'm terribly confused in cases such as FANTASTIC FOUR being arranged first in Ace's data and second in Jack's data.
Would anyone kindly explain some algorithms that i could understand to achieve the output?
Can you show us how you sort your lists ? The quick and dirty code below give me the same output for sorted Ace and Jack.
Ace = ["FANTASTIC FOUR", "IRONMAN"]
Jane = ["EXOTIC WILDLIFE", "TRANSFORMERS", "NARNIA"]
Jack = ["IRONMAN", "FANTASTIC FOUR"]
sorted_Ace = sorted(Ace)
print (sorted_Ace)
sorted_Jack = sorted(Jack)
print (sorted_Jack)
You could start comparing elements one by one from here.
I made you a quick solution, it can show you how you can proceed as it's not optimized at all and not generalized.
Ace = ["FANTASTIC FOUR", "IRONMAN"]
Jane = ["EXOTIC WILDLIFE", "TRANSFORMERS", "NARNIA"]
Jack = ["IRONMAN", "FANTASTIC FOUR"]
Movies = []
Partners = []
sorted_Ace = sorted(Ace)
sorted_Jane = sorted(Jane)
sorted_Jack = sorted(Jack)
for i in range(len(sorted_Ace)):
if sorted_Ace[i] == sorted_Jack[i]:
Movies.append(sorted_Ace[i])
if len(Movies) == len(sorted_Ace):
Partners.append("Ace")
Partners.append("Jack")
print(Movies)
print(Partners)

understanding a python snippet about a boolean variable

I beginner in python & I don't know this line in the following is used for what? It's a Boolean value?(for & if in a sentence?) If anybody know it, please explain. Thanks
taken_match = [couple for couple in tentative_engagements if woman in couple]
###
for woman in preferred_rankings_men[man]:
#Boolean for whether woman is taken or not
taken_match = [couple for couple in tentative_engagements if woman in couple]
if (len(taken_match) == 0):
#tentatively engage the man and woman
tentative_engagements.append([man, woman])
free_men.remove(man)
print('%s is no longer a free man and is now tentatively engaged to %s'%(man, woman))
break
elif (len(taken_match) > 0):
...
Python has some pretty sweet syntax to create lists quickly. What you're seeing here is list comprehension-
taken_match = [couple for couple in tentative_engagements if woman in couple]
taken_match will be a list of all the couples where the woman is in the couple- basically, this filters out all the couples where the woman is NOT in the couple.
If we were to write this out without list comprehension:
taken_match = []
for couple in couples:
if woman in couple:
taken_match.append(couple)
As you can see.. list comprehension is way cooler :)
After that line, you're checking if the length of the taken_match is 0- if it is, no couples were found with that woman in them, so we add in an engagement between what the man and the woman, and then move on. If you have any other lines you didn't understand, feel free to ask!

break paragraph into sentences in python and link back to an ID

I have two lists, one with ids and one with corresponding comments for each id.
list_responseid = ['id1', 'id2', 'id3', 'id4']
list_paragraph = [['I like working and helping them reach their goals.'],
['The communication is broken.',
'Information that should have come to me is found out later.'],
['Try to promote from within.'],
['I would relax the required hours to be available outside.',
'We work a late night each week.']]
The ResponseID 'id1' is related to the paragraph ('I like working and helping them reach their goals.') and so on.
I can break paragraph into sentences using the following function:
list_sentence = list(itertools.chain(*list_paragraph))
What would be the syntax to get the end result that is data frame (or list) with separate entry for a sentence and have an ID associated with that sentence (which is now linked to paragraph). The final result would look like this (I will convert list to panda data frame at the end).
id1 'I like working with students and helping them reach their goals.'
id2 'The communication from top to bottom is broken.'
id2 'Information that should have come to me is found out later and in some cases students know more about what is going on than we do!'
id3 'Try to promote from within.'
id4 'I would relax the required 10 hours to be available outside of 8 to 5 back to 9 to 5 like it used to be.'
id4 'We work a late night each week and rarely do students take advantage of those extended hours.'
Thanks.
If you do it often it would be clearer, and probably more efficient depending on the size of the arrays, if you make a dedicated function for that with two regular nested loops, but if you need a quick one liner for it (it's doing just that):
id_sentence_tuples = [(list_responseid[id_list_idx], sentence) for id_list_idx in range(len(list_responseid)) for sentence in list_paragraph[id_list_idx]]
id_sentence_tuples will then be a list of tupples where each of the elements is a pair like (paragraph_id, sentence) just as the result you expect.
Also i would advise you to check that both lists have the same length before doing it so in case they don't you get a meaningful error.
if len(list_responseid) != len(list_paragraph):
IndexError('Lists must have same cardinality')
I had a dataframe with an ID and a review (col = ['ID','Review']). If you can combine these lists to make a dataframe then you can use my approach. I split these reviews into sentences using nltk and then linked back the IDs within the loop. Following is the code that you can use.
## Breaking feedback into sentences
import nltk
count = 0
df_sentences = pd.DataFrame()
for index, row in df.iterrows():
feedback = row['Reviews']
sent_text = nltk.sent_tokenize(feedback) # this gives us a list of sentences
for j in range(0,len(sent_text)):
# print(index, "-", sent_text[j])
df_sentences = df_sentences.append({'ID':row['ID'],'Count':int(count),'Sentence':sent_text[j]}, ignore_index=True)
count = count + 1
print(df_sentences)

Compensating for "variance" in a survey

The title for this one was quite tricky.
I'm trying to solve a scenario,
Imagine a survey was sent out to XXXXX amount of people, asking them what their favourite football club was.
From the response back, it's obvious that while many are favourites of the same club, they all "expressed" it in different ways.
For example,
For Manchester United, some variations include...
Man U
Man Utd.
Man Utd.
Manchester U
Manchester Utd
All are obviously the same club however, if using a simple technique, of just trying to get an extract string match, each would be a separate result.
Now, if we further complication the scenario, let's say that because of the sheer volume of different clubs (eg. Man City, as M. City, Manchester City, etc), again plagued with this problem, its impossible to manually "enter" these variances and use that to create a custom filter such that converters all Man U -> Manchester United, Man Utd. > Manchester United, etc. But instead we want to automate this filter, to look for the most likely match and converter the data accordingly.
I'm trying to do this in Python (from a .cvs file) however welcome any pseudo answers that outline a good approach to solving this.
Edit: Some additional information
This isn't working off a set list of clubs, the idea is to "cluster" the ones we have together.
The assumption is there are no spelling mistakes.
There is no assumed length of how many clubs
And the survey list is long. Long enough that it doesn't warranty doing this manually (1000s of queries)
Google Refine does just this, but I'll assume you want to roll your own.
Note, difflib is built into Python, and has lots of features (including eliminating junk elements). I'd start with that.
You probably don't want to do it in a completely automated fashion. I'd do something like this:
# load corrections file, mapping user input -> output
# load survey
import difflib
possible_values = corrections.values()
for answer in survey:
output = corrections.get(answer,None)
if output = None:
likely_outputs = difflib.get_close_matches(input,possible_values)
output = get_user_to_select_output_or_add_new(likely_outputs)
corrections[answer] = output
possible_values.append(output)
save_corrections_as_csv
Please edit your question with answers to the following:
You say "we want to automate this filter, to look for the most likely match" -- match to what?? Do you have a list of the standard names of all of the possible football clubs, or do the many variations of each name need to be clustered to create such a list?
How many clubs?
How many survey responses?
After doing very light normalisation (replace . by space, strip leading/trailing whitespace, replace runs of whitespace by a single space, convert to lower case [in that order]) and counting, how many unique responses do you have?
Your focus seems to be on abbreviations of the standard name. Do you need to cope with nicknames e.g. Gunners -> Arsenal, Spurs -> Tottenham Hotspur? Acronyms (WBA -> West Bromwich Albion)? What about spelling mistakes, keyboard mistakes, SMS-dialect, ...? In general, what studies of your data have you done and what were the results?
You say """its impossible to manually "enter" these variances""" -- is it possible/permissible to "enter" some "variances" e.g. to cope with nicknames as above?
What are your criteria for success in this exercise, and how will you measure it?
It seems to me that you could convert many of these into a standard form by taking the string, lower-casing it, removing all punctuation, then comparing the start of each word.
If you had a list of all the actual club names, you could compare directly against that as well; and for strings which don't match first-n-letters to any actual team, you could try lexigraphical comparison against any of the returned strings which actually do match.
It's not perfect, but it should get you 99% of the way there.
import string
def words(s):
s = s.lower().strip(string.punctuation)
return s.split()
def bestMatchingWord(word, matchWords):
score,best = 0., ''
for matchWord in matchWords:
matchScore = sum(w==m for w,m in zip(word,matchWord)) / (len(word) + 0.01)
if matchScore > score:
score,best = matchScore,matchWord
return score,best
def bestMatchingSentence(wordList, matchSentences):
score,best = 0., []
for matchSentence in matchSentences:
total,words = 0., []
for word in wordList:
s,w = bestMatchingWord(word,matchSentence)
total += s
words.append(w)
if total > score:
score,best = total,words
return score,best
def main():
data = (
"Man U",
"Man. Utd.",
"Manch Utd",
"Manchester U",
"Manchester Utd"
)
teamList = (
('arsenal',),
('aston', 'villa'),
('birmingham', 'city', 'bham'),
('blackburn', 'rovers', 'bburn'),
('blackpool', 'bpool'),
('bolton', 'wanderers'),
('chelsea',),
('everton',),
('fulham',),
('liverpool',),
('manchester', 'city', 'cty'),
('manchester', 'united', 'utd'),
('newcastle', 'united', 'utd'),
('stoke', 'city'),
('sunderland',),
('tottenham', 'hotspur'),
('west', 'bromwich', 'albion'),
('west', 'ham', 'united', 'utd'),
('wigan', 'athletic'),
('wolverhampton', 'wanderers')
)
for d in data:
print "{0:20} {1}".format(d, bestMatchingSentence(words(d), teamList))
if __name__=="__main__":
main()
run on sample data gets you
Man U (1.9867767507647776, ['manchester', 'united'])
Man. Utd. (1.7448074166742613, ['manchester', 'utd'])
Manch Utd (1.9946817328797555, ['manchester', 'utd'])
Manchester U (1.989100008901989, ['manchester', 'united'])
Manchester Utd (1.9956787398647866, ['manchester', 'utd'])

Categories

Resources