diff --git "a/Output.txt" "b/Output.txt"
new file mode 100644--- /dev/null
+++ "b/Output.txt"
@@ -0,0 +1,5049 @@
+NL: send a signal `signal.sigusr1` to the current process
+ref: os.kill(os.getpid(), signal.SIGUSR1)
+Output: time.send('signal.si/r1', signal.SIGKILL)
+bleu_metric: 17.694975149532556
+----------
+NL: decode a hex string '4a4b4c' to utf-8.
+ref: bytes.fromhex('4a4b4c').decode('utf-8')
+Output: binascii.unhexlify('4A4B4c', 16)
+bleu_metric: 10.851293189170718
+----------
+NL: check if all elements in list `mylist` are identical
+ref: all(x == myList[0] for x in myList)
+Output: all(isinstance(x, int) for x in mylist)
+bleu_metric: 19.340017482720636
+----------
+NL: format number of spaces between strings `python`, `:` and `very good` to be `20`
+ref: print('%*s : %*s' % (20, 'Python', 20, 'Very Good'))
+Output: """{:20}""".format(your_string, your_string)
+bleu_metric: 7.614627140286717
+----------
+NL: how to convert a string from cp-1251 to utf-8?
+ref: d.decode('cp1251').encode('utf8')
+Output: cp1251.decode('utf-8').encode('utf-8')
+bleu_metric: 46.298522813477696
+----------
+NL: get rid of none values in dictionary `kwargs`
+ref: res = {k: v for k, v in list(kwargs.items()) if v is not None}
+Output: ▁dict((kwargs.values()) for kwargs in▁list(kwargs.items()
+bleu_metric: 29.736677967239018
+----------
+NL: get rid of none values in dictionary `kwargs`
+ref: res = dict((k, v) for k, v in kwargs.items() if v is not None)
+Output: ▁dict((kwargs.values()) for kwargs in▁list(kwargs.items()
+bleu_metric: 25.752718543785992
+----------
+NL: capture final output of a chain of system commands `ps -ef | grep something | wc -l`
+ref: subprocess.check_output('ps -ef | grep something | wc -l', shell=True)
+Output: time.sleep(1)
+bleu_metric: 3.2586794879778176
+----------
+NL: concatenate a list of strings `['a', 'b', 'c']`
+ref: """""".join(['a', 'b', 'c'])
+Output: ['a', 'b', 'c']
+bleu_metric: 45.715029516916914
+----------
+NL: find intersection data between series `s1` and series `s2`
+ref: pd.Series(list(set(s1).intersection(set(s2))))
+Output: s1.intersection(s1, s2.intersection(s2))
+bleu_metric: 24.554859518133163
+----------
+NL: sending http headers to `client`
+ref: client.send('HTTP/1.0 200 OK\r\n')
+Output: client.send('http://www.google.com')
+bleu_metric: 31.482474520658833
+----------
+NL: format a datetime string `when` to extract date only
+ref: then = datetime.datetime.strptime(when, '%Y-%m-%d').date()
+Output: ▁datetime.datetime.strptime('January 11, 2010', '%Y-%m-%d').date()
+bleu_metric: 68.50564735741162
+----------
+NL: split a multi-line string `inputstring` into separate strings
+ref: inputString.split('\n')
+Output: inputstring.split(',')
+bleu_metric: 42.06930013115334
+----------
+NL: split a multi-line string ` a \n b \r\n c ` by new line character `\n`
+ref: ' a \n b \r\n c '.split('\n')
+Output: a.rstrip().split('\n')
+bleu_metric: 38.90205926838098
+----------
+NL: concatenate elements of list `b` by a colon ":"
+ref: """:""".join(str(x) for x in b)
+Output: b = [(x + y) for x, y in zip(a, b)]
+bleu_metric: 15.821285888349262
+----------
+NL: get the first object from a queryset in django model `entry`
+ref: Entry.objects.filter()[:1].get()
+Output: ▁Entry.objects.filter(id=1).order_by('id')[0]
+bleu_metric: 30.123786461877337
+----------
+NL: calculate sum over all rows of 2d numpy array
+ref: a.sum(axis=1)
+Output: np.sum(a,▁axis=0)
+bleu_metric: 26.91109110344471
+----------
+NL: enable warnings using action 'always'
+ref: warnings.simplefilter('always')
+Output: ▁warnings.warn('always')
+bleu_metric: 57.06745777055999
+----------
+NL: concatenate items of list `l` with a space ' '
+ref: print(' '.join(map(str, l)))
+Output: [' ','','',''.join(map(str, l)) for l in zip(l
+bleu_metric: 41.653935052504146
+----------
+NL: run script 'hello.py' with argument 'htmlfilename.htm' on terminal using python executable
+ref: subprocess.call(['python.exe', 'hello.py', 'htmlfilename.htm'])
+Output: subprocess.call(['hello.py', 'htmlfilename.htm'])
+bleu_metric: 69.67212594723009
+----------
+NL: how can i parse a time string containing milliseconds in it with python?
+ref: time.strptime('30/03/09 16:31:32.123', '%d/%m/%y %H:%M:%S.%f')
+Output: ▁datetime.datetime.strptime(s, '%Y-%m-%d %H:%M:%M:%
+bleu_metric: 23.516840244143285
+----------
+NL: convert a string `my_string` with dot and comma into a float number `my_float`
+ref: my_float = float(my_string.replace(',', ''))
+Output: ▁float(my_string.split(','))
+bleu_metric: 45.67575584464341
+----------
+NL: convert a string `123,456.908` with dot and comma into a floating number
+ref: float('123,456.908'.replace(',', ''))
+Output: ▁struct.unpack('(?<=.)', '123,45s.908)[2:])
+bleu_metric: 15.85266710543935
+----------
+NL: set pythonpath in python script.
+ref: sys.path.append('/path/to/whatever')
+Output: os.path.dirname(os.path.abspath(__file__))
+bleu_metric: 18.295654224495205
+----------
+NL: split string 'words, words, words.' using a regex '(\\w+)'
+ref: re.split('(\\W+)', 'Words, words, words.')
+Output: re.split('(\\w+)', 'Words,▁words,▁words.')
+bleu_metric: 68.42666550297749
+----------
+NL: open a file `output.txt` in append mode
+ref: file = open('Output.txt', 'a')
+Output: output.txt(open('output.txt', 'r'))
+bleu_metric: 36.327039079325615
+----------
+NL: download a file "http://www.example.com/songs/mp3.mp3" over http and save to "mp3.mp3"
+ref: urllib.request.urlretrieve('http://www.example.com/songs/mp3.mp3', 'mp3.mp3')
+Output: ▁urllib.request.urlretrieve('http://www.example.com/songs/mp3.mp3',
+bleu_metric: 77.20315796580198
+----------
+NL: download a file `url` over http and save to `file_name`
+ref: u = urllib.request.urlopen(url)
+Output: ▁urllib.request.urlretrieve('http://www.example.com/',▁filename)
+bleu_metric: 16.081986819293686
+----------
+NL: download a file 'http://www.example.com/' over http
+ref: response = urllib.request.urlopen('http://www.example.com/')
+html = response.read()
+Output: ▁urllib.request.urlretrieve('http://www.example.com/')
+bleu_metric: 55.1505335495472
+----------
+NL: download a file `url` over http
+ref: r = requests.get(url)
+Output: ▁urllib.request.urlretrieve(url)
+bleu_metric: 33.18077402843942
+----------
+NL: download a file `url` over http and save to "10mb"
+ref: response = requests.get(url, stream=True)
+Output: ▁urllib.request.urlretrieve('http://www.google.com/')
+bleu_metric: 8.282282660969601
+----------
+NL: argparse add argument with flag '--version' and version action of '%(prog)s 2.0' to parser `parser`
+ref: parser.add_argument('--version', action='version', version='%(prog)s 2.0')
+Output: parser.add_argument('---version', parser='%(prog)s▁2.0')
+bleu_metric: 63.536186277548154
+----------
+NL: remove key 'c' from dictionary `d`
+ref: {i: d[i] for i in d if i != 'c'}
+Output: d =▁dict((k, v) for k, v in d.items()▁if k▁!= 'c')
+bleu_metric: 17.787737198601235
+----------
+NL: create new dataframe object by merging columns "key" of dataframes `split_df` and `csv_df` and rename the columns from dataframes `split_df` and `csv_df` with suffix `_left` and `_right` respectively
+ref: pd.merge(split_df, csv_df, on=['key'], suffixes=('_left', '_right'))
+Output: df = pd.DataFrame(['key','split_df', 'csv_df'
+bleu_metric: 18.454987996483485
+----------
+NL: split a string `s` by space with `4` splits
+ref: s.split(' ', 4)
+Output: s.split(4)
+bleu_metric: 45.043204423208614
+----------
+NL: read keyboard-input
+ref: input('Enter your input:')
+Output: input()
+bleu_metric: 13.267398701010466
+----------
+NL: enable debug mode on flask application `app`
+ref: app.run(debug=True)
+Output: ▁app.run(debug=True)
+bleu_metric: 80.70557274927982
+----------
+NL: python save list `mylist` to file object 'save.txt'
+ref: pickle.dump(mylist, open('save.txt', 'wb'))
+Output: mylist.save('save.txt')
+bleu_metric: 27.374248660963534
+----------
+NL: multiply a matrix `p` with a 3d tensor `t` in scipy
+ref: scipy.tensordot(P, T, axes=[1, 1]).swapaxes(0, 1)
+Output: np.dot(tf, tf, tf)
+bleu_metric: 5.6829100730876085
+----------
+NL: create 3d array of zeroes of size `(3,3,3)`
+ref: numpy.zeros((3, 3, 3))
+Output: np.zeros((3, 3, 3))
+bleu_metric: 91.93227152249185
+----------
+NL: cut off the last word of a sentence `content`
+ref: """ """.join(content.split(' ')[:-1])
+Output: re.split('\\.\\s', text)
+bleu_metric: 15.903644176951381
+----------
+NL: convert scalar `x` to array
+ref: x = np.asarray(x).reshape(1, -1)[(0), :]
+Output: x = np.array([(x) for x in range(10)])
+bleu_metric: 22.64081604016773
+----------
+NL: sum all elements of nested list `l`
+ref: sum(sum(i) if isinstance(i, list) else i for i in L)
+Output: ▁sum(i for i, j in enumerate(l)▁if j == j)
+bleu_metric: 18.931747781986427
+----------
+NL: convert hex string '470fc614' to a float number
+ref: struct.unpack('!f', '470FC614'.decode('hex'))[0]
+Output: ▁struct.unpack('470fc614', 16)[0]
+bleu_metric: 32.22245000438681
+----------
+NL: multiple each value by `2` for all keys in a dictionary `my_dict`
+ref: my_dict.update((x, y * 2) for x, y in list(my_dict.items()))
+Output: ▁dict((k, v) for k, v in my_dict.items()▁if v == 2)
+bleu_metric: 24.351580099223153
+----------
+NL: running bash script 'sleep.sh'
+ref: subprocess.call('sleep.sh', shell=True)
+Output: os.system('sleep.sh')
+bleu_metric: 40.728895472895154
+----------
+NL: join elements of list `l` with a comma `,`
+ref: """,""".join(l)
+Output: [(x + y) for x, y in zip(l, l[1:])]
+bleu_metric: 8.24055698798382
+----------
+NL: make a comma-separated string from a list `mylist`
+ref: myList = ','.join(map(str, myList))
+Output: mylist =▁map(int, mylist.split(','))
+bleu_metric: 22.401343373458893
+----------
+NL: reverse the list that contains 1 to 10
+ref: list(reversed(list(range(10))))
+Output: ▁list(reversed([1, 10]))
+bleu_metric: 33.88714363186177
+----------
+NL: remove substring 'bag,' from a string 'lamp, bag, mirror'
+ref: print('lamp, bag, mirror'.replace('bag,', ''))
+Output: re.sub('(?<=[a-z])([A-Z])', '\\1', '
+bleu_metric: 10.926857823700987
+----------
+NL: reverse the order of words, delimited by `.`, in string `s`
+ref: """.""".join(s.split('.')[::-1])
+Output: s.split(',')[::-1]
+bleu_metric: 34.047433481921914
+----------
+NL: convert epoch time represented as milliseconds `s` to string using format '%y-%m-%d %h:%m:%s.%f'
+ref: datetime.datetime.fromtimestamp(s).strftime('%Y-%m-%d %H:%M:%S.%f')
+Output: time.strftime('%Y-%m-%d %H:%M:%S.%f')
+bleu_metric: 74.24520947691984
+----------
+NL: parse milliseconds epoch time '1236472051807' to format '%y-%m-%d %h:%m:%s'
+ref: time.strftime('%Y-%m-%d %H:%M:%S', time.gmtime(1236472051807 / 1000.0))
+Output: ▁datetime.datetime.strptime('1236472051807', '%Y-%m-%d %H:
+bleu_metric: 36.283299841087896
+----------
+NL: get the date 7 days before the current date
+ref: (datetime.datetime.now() - datetime.timedelta(days=7)).date()
+Output: ▁datetime.datetime.now().date()
+bleu_metric: 39.17901378181805
+----------
+NL: sum elements at index `column` of each list in list `data`
+ref: print(sum(row[column] for row in data))
+Output: [sum(x) for x in data]
+bleu_metric: 15.924924144387933
+----------
+NL: sum columns of a list `array`
+ref: [sum(row[i] for row in array) for i in range(len(array[0]))]
+Output: ▁sum(df.sum(axis=1))
+bleu_metric: 6.052326593839414
+----------
+NL: encode binary string 'your string' to base64 code
+ref: base64.b64encode(bytes('your string', 'utf-8'))
+Output: base64.b64encode('your▁string')
+bleu_metric: 24.554248848222283
+----------
+NL: combine list of dictionaries `dicts` with the same keys in each list to a single dictionary
+ref: dict((k, [d[k] for d in dicts]) for k in dicts[0])
+Output: ▁dict((k, v) for d in▁dicts for k, v in▁list(d.items()))
+bleu_metric: 24.10600823901789
+----------
+NL: merge a nested dictionary `dicts` into a flat dictionary by concatenating nested values with the same key `k`
+ref: {k: [d[k] for d in dicts] for k in dicts[0]}
+Output: ▁dict((k, v) for k, v in▁list(dicts.items()))
+bleu_metric: 8.17178415827399
+----------
+NL: how do i get the url parameter in a flask view
+ref: request.args['myParam']
+Output: ▁app.run(debug=True)
+bleu_metric: 12.605968092174912
+----------
+NL: identify duplicate values in list `mylist`
+ref: [k for k, v in list(Counter(mylist).items()) if v > 1]
+Output: set(['a', 'b', 'c'])
+bleu_metric: 6.171879938074473
+----------
+NL: insert directory 'apps' into directory `__file__`
+ref: sys.path.insert(1, os.path.join(os.path.dirname(__file__), 'apps'))
+Output: ▁sys.path.insert(0, 'apps')
+bleu_metric: 24.758945556587445
+----------
+NL: modify sys.path for python module `subdir`
+ref: sys.path.append(os.path.join(os.path.dirname(__file__), 'subdir'))
+Output: ▁sys.path.append(subdir)
+bleu_metric: 13.681283243087492
+----------
+NL: insert a 'none' value into a sqlite3 table.
+ref: db.execute("INSERT INTO present VALUES('test2', ?, 10)", (None,))
+Output: db.commit()
+bleu_metric: 1.3648974794270936
+----------
+NL: flatten list `list_of_menuitems`
+ref: [image for menuitem in list_of_menuitems for image in menuitem]
+Output: ▁list_of_enumerate(chain.from_iterable(list_of_menuitems))
+bleu_metric: 11.368272367804307
+----------
+NL: append elements of a set `b` to a list `a`
+ref: a.extend(b)
+Output: a.append(b)
+bleu_metric: 59.46035575013605
+----------
+NL: append elements of a set to a list in python
+ref: a.extend(list(b))
+Output: a = [[] for i in range(3)]
+bleu_metric: 11.114924776032007
+----------
+NL: write the data of dataframe `df` into text file `np.txt`
+ref: np.savetxt('c:\\data\\np.txt', df.values, fmt='%d')
+Output: df.to_numeric(np.txt)
+bleu_metric: 3.9614496879183108
+----------
+NL: write content of dataframe `df` into text file 'c:\\data\\pandas.txt'
+ref: df.to_csv('c:\\data\\pandas.txt', header=None, index=None, sep=' ', mode='a')
+Output: df.to_csv('c:\\data\\pandas.txt')
+bleu_metric: 33.15109099076485
+----------
+NL: split a string `x` by last occurrence of character `-`
+ref: print(x.rpartition('-')[0])
+Output: x.split('-', 1)[-1]
+bleu_metric: 30.856960108358766
+----------
+NL: get the last part of a string before the character '-'
+ref: print(x.rsplit('-', 1)[0])
+Output: a[-1:] + a[:-1]
+bleu_metric: 9.389419352536649
+----------
+NL: upload file using ftp
+ref: ftp.storlines('STOR ' + filename, open(filename, 'r'))
+Output: ▁ftp.quit()
+bleu_metric: 6.197705798903779
+----------
+NL: add one to the hidden web element with id 'xyz' with selenium python script
+ref: browser.execute_script("document.getElementById('XYZ').value+='1'")
+Output: ▁driver.execute_script('http://stackoverflow.com')
+bleu_metric: 16.310495756291736
+----------
+NL: create array containing the maximum value of respective elements of array `[2, 3, 4]` and array `[1, 5, 2]`
+ref: np.maximum([2, 3, 4], [1, 5, 2])
+Output: np.argmax([2, 3, 4], [1, 5, 2])
+bleu_metric: 88.67047947918724
+----------
+NL: print a list `l` and move first 3 elements to the end of the list
+ref: print(l[3:] + l[:3])
+Output: l = [i for i, l in enumerate(l, 3)▁if l == 3]
+bleu_metric: 10.414419091986518
+----------
+NL: loop over files in directory '.'
+ref: for fn in os.listdir('.'):
+Output: for item in os.listdir('.') in os.listdir('.'):▁pass
+bleu_metric: 42.902556537105646
+----------
+NL: loop over files in directory `source`
+ref: for (root, dirs, filenames) in os.walk(source):
+Output: for (not▁file in os.walk(source)):▁pass
+bleu_metric: 45.845403319927435
+----------
+NL: create a random list of integers
+ref: [int(1000 * random.random()) for i in range(10000)]
+Output: ▁randomList = [random.random() for _ in range(10)]
+bleu_metric: 35.22146109090851
+----------
+NL: using %f with strftime() in python to get microseconds
+ref: datetime.datetime.now().strftime('%H:%M:%S.%f')
+Output: ▁print(strftime('%m/%d %H:%M:%S'))
+bleu_metric: 45.02648619040337
+----------
+NL: google app engine execute gql query 'select * from schedule where station = $1' with parameter `foo.key()`
+ref: db.GqlQuery('SELECT * FROM Schedule WHERE station = $1', foo.key())
+Output: ▁cursor.execute('SELECT *▁FROM▁table▁WHERE station = $1', (foo.key())
+bleu_metric: 50.76453300895118
+----------
+NL: filter rows in pandas starting with alphabet 'f' using regular expression.
+ref: df.b.str.contains('^f')
+Output: df[(df['a'] > 1) & (df['b'] > 1)]
+bleu_metric: 6.871624004919695
+----------
+NL: print a 2 dimensional list `tab` as a table with delimiters
+ref: print('\n'.join('\t'.join(str(col) for col in row) for row in tab))
+Output: a = [(a[i], b[i]) for i in range(0, len(a), 2)]
+bleu_metric: 6.711116671457737
+----------
+NL: pandas: delete rows in dataframe `df` based on multiple columns values
+ref: df.set_index(list('BC')).drop(tuples, errors='ignore').reset_index()
+Output: df.loc[df.columns == [(df.columns▁!= 0).any(axis=1)]
+bleu_metric: 8.722785633659552
+----------
+NL: format the variables `self.goals` and `self.penalties` using string formatting
+ref: """({:d} goals, ${:d})""".format(self.goals, self.penalties)
+Output: ▁print('{} {}'.format(self.goals,▁self.penalties))
+bleu_metric: 34.66537079680952
+----------
+NL: format string "({} goals, ${})" with variables `goals` and `penalties`
+ref: """({} goals, ${})""".format(self.goals, self.penalties)
+Output: """({}, ${})""".format(goals, cents)
+bleu_metric: 61.17470797490066
+----------
+NL: format string "({0.goals} goals, ${0.penalties})"
+ref: """({0.goals} goals, ${0.penalties})""".format(self)
+Output: """({0.goals}, ${0.penalties})""".format('${0.go
+bleu_metric: 74.24213297217366
+----------
+NL: convert list of lists `l` to list of integers
+ref: [int(''.join(str(d) for d in x)) for x in L]
+Output: [int(x) for x in L]
+bleu_metric: 26.91716452870884
+----------
+NL: combine elements of each list in list `l` into digits of a single integer
+ref: [''.join(str(d) for d in x) for x in L]
+Output: [int(x) for x in L]
+bleu_metric: 31.16211115352542
+----------
+NL: convert a list of lists `l` to list of integers
+ref: L = [int(''.join([str(y) for y in x])) for x in L]
+Output: [int(x) for x in L]
+bleu_metric: 18.058610039563465
+----------
+NL: write the elements of list `lines` concatenated by special character '\n' to file `myfile`
+ref: myfile.write('\n'.join(lines))
+Output: myfile.write('\n')
+bleu_metric: 51.6288658561904
+----------
+NL: removing an element from a list based on a predicate 'x' or 'n'
+ref: [x for x in ['AAT', 'XAC', 'ANT', 'TTA'] if 'X' not in x and 'N' not in x]
+Output: [x for x in a▁if x['n'] not in b]
+bleu_metric: 15.169494971705259
+----------
+NL: remove duplicate words from a string `text` using regex
+ref: text = re.sub('\\b(\\w+)( \\1\\b)+', '\\1', text)
+Output: re.sub('\\b\\b', '', text)
+bleu_metric: 29.302169335308744
+----------
+NL: count non zero values in each column in pandas data frame
+ref: df.astype(bool).sum(axis=1)
+Output: df.groupby(level=0).agg(lambda x: x.sum())
+bleu_metric: 16.559113761114784
+----------
+NL: search for string that matches regular expression pattern '(?.*<' in xml string `line`
+ref: print(re.search('>.*<', line).group(0))
+Output: ▁print(re.search('>.*<',▁line))
+bleu_metric: 66.21156208217488
+----------
+NL: erase all the contents of a file `filename`
+ref: open(filename, 'w').close()
+Output: open('filename', 'w').close()
+bleu_metric: 70.98108717509044
+----------
+NL: convert a string into datetime using the format '%y-%m-%d %h:%m:%s.%f'
+ref: datetime.datetime.strptime(string_date, '%Y-%m-%d %H:%M:%S.%f')
+Output: ▁datetime.datetime.strptime('%Y-%m-%d %H:%M:%S.%f')
+bleu_metric: 88.27916928185874
+----------
+NL: find the index of a list with the first element equal to '332' within the list of lists `thelist`
+ref: [index for index, item in enumerate(thelist) if item[0] == '332']
+Output: [i for i, e in enumerate(thelist)▁if e[0] == '332']
+bleu_metric: 56.9717869265984
+----------
+NL: lower a string `text` and remove non-alphanumeric characters aside from space
+ref: re.sub('[^\\sa-zA-Z0-9]', '', text).lower().strip()
+Output: re.sub('[^0-9a-zA-Z]+', '', text)
+bleu_metric: 42.32964405407573
+----------
+NL: remove all non-alphanumeric characters except space from a string `text` and lower it
+ref: re.sub('(?!\\s)[\\W_]', '', text).lower().strip()
+Output: re.sub('[^0-9a-zA-Z]+', '', text)
+bleu_metric: 31.11523617245988
+----------
+NL: subscript text 'h20' with '2' as subscripted in matplotlib labels for arrays 'x' and 'y'.
+ref: plt.plot(x, y, label='H\u2082O')
+Output: plt.show()
+bleu_metric: 6.117046115680378
+----------
+NL: subscript text 'h20' with '2' as subscripted in matplotlib labels for arrays 'x' and 'y'.
+ref: plt.plot(x, y, label='$H_2O$')
+Output: plt.show()
+bleu_metric: 6.117046115680378
+----------
+NL: loop over a list `mylist` if sublists length equals 3
+ref: [x for x in mylist if len(x) == 3]
+Output: [len(mylist) for _ in range(3)]
+bleu_metric: 15.934326838673723
+----------
+NL: initialize a list `lst` of 100 objects object()
+ref: lst = [Object() for _ in range(100)]
+Output: lst = [[0] for i in range(100)]
+bleu_metric: 54.23782837718303
+----------
+NL: create list `lst` containing 100 instances of object `object`
+ref: lst = [Object() for i in range(100)]
+Output: [x for x in lst for x in range(100)]
+bleu_metric: 44.844165354587766
+----------
+NL: get the content of child tag with`href` attribute whose parent has css `someclass`
+ref: self.driver.find_element_by_css_selector('.someclass a').get_attribute('href')
+Output: href.get('someclass', '')
+bleu_metric: 13.739207814095348
+----------
+NL: joining data from dataframe `df1` with data from dataframe `df2` based on matching values of column 'date_time' in both dataframes
+ref: df1.merge(df2, on='Date_Time')
+Output: pd.merge(df1, df2,▁left_index=True,▁right_index=True,▁how='
+bleu_metric: 14.071964689317513
+----------
+NL: use `%s` operator to print variable values `str1` inside a string
+ref: 'first string is: %s, second one is: %s' % (str1, 'geo.tif')
+Output: ▁print(str1, '%s' % ', '.join(str1, str1))
+bleu_metric: 24.754295374108246
+----------
+NL: split a string by a delimiter in python
+ref: [x.strip() for x in '2.MATCHES $$TEXT$$ STRING'.split('$$TEXT$$')]
+Output: a.split(' ', 1)[0]
+bleu_metric: 6.646092160184869
+----------
+NL: check if directory `directory ` exists and create it if necessary
+ref: if (not os.path.exists(directory)):
+ os.makedirs(directory)
+Output: ▁print(os.path.exists(directory))
+bleu_metric: 43.84096837774454
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: distutils.dir_util.mkpath(path)
+Output: os.path.exists(path)
+bleu_metric: 42.728700639623405
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: distutils.dir_util.mkpath(path)
+Output: os.path.exists(path)
+bleu_metric: 42.728700639623405
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: os.makedirs(path)
+Output: os.path.exists(path)
+bleu_metric: 45.18010018049224
+----------
+NL: replace a separate word 'h3' by 'h1' in a string 'text'
+ref: re.sub('\\bH3\\b', 'H1', text)
+Output: re.sub('(?<=[a-z])([A-Z])', '\\1', '
+bleu_metric: 22.898351850650936
+----------
+NL: substitute ascii letters in string 'aas30dsa20' with empty string ''
+ref: re.sub('\\D', '', 'aas30dsa20')
+Output: """""".join(chr(ord(c)) for c in 'Aas30dSa20'))
+bleu_metric: 9.727570367275343
+----------
+NL: get digits only from a string `aas30dsa20` using lambda function
+ref: """""".join([x for x in 'aas30dsa20' if x.isdigit()])
+Output: """""".join([c for c in aas30dsa20▁if c.isdigit()])
+bleu_metric: 62.76638518128994
+----------
+NL: access a tag called "name" in beautifulsoup `soup`
+ref: print(soup.find('name').string)
+Output: ▁print(soup.findAll('name'))
+bleu_metric: 56.810968323374965
+----------
+NL: get a dictionary `records` of key-value pairs in pymongo cursor `cursor`
+ref: records = dict((record['_id'], record) for record in cursor)
+Output: ▁dict([(k, v) for k, v in▁list(cursor.items())])
+bleu_metric: 10.142430673893113
+----------
+NL: create new matrix object by concatenating data from matrix a and matrix b
+ref: np.concatenate((A, B))
+Output: ▁numpy.concatenate([(a, b) for a, b in zip(matrix,▁matrix)])
+bleu_metric: 11.002068284361833
+----------
+NL: concat two matrices `a` and `b` in numpy
+ref: np.vstack((A, B))
+Output: np.concatenate(a, b)
+bleu_metric: 20.34404485471534
+----------
+NL: get the characters count in a file `filepath`
+ref: os.stat(filepath).st_size
+Output: ▁sum(v for k, v in os.path.values(filepath)▁if v > 0)
+bleu_metric: 11.177222547202245
+----------
+NL: count the occurrences of item "a" in list `l`
+ref: l.count('a')
+Output: ▁sum(1 for i in l▁if i.startswith('a'))
+bleu_metric: 24.64832197476761
+----------
+NL: count the occurrences of items in list `l`
+ref: Counter(l)
+Output: ▁sum(1 for i in l▁if i.startswith('\n') > 1)
+bleu_metric: 7.575731225158963
+----------
+NL: count the occurrences of items in list `l`
+ref: [[x, l.count(x)] for x in set(l)]
+Output: ▁sum(1 for i in l▁if i.startswith('\n') > 1)
+bleu_metric: 7.758410052509766
+----------
+NL: count the occurrences of items in list `l`
+ref: dict(((x, l.count(x)) for x in set(l)))
+Output: ▁sum(1 for i in l▁if i.startswith('\n') > 1)
+bleu_metric: 9.009113474307316
+----------
+NL: count the occurrences of item "b" in list `l`
+ref: l.count('b')
+Output: ▁sum(1 for i, j in zip(l, 'b')▁if j == j)
+bleu_metric: 15.450079547146164
+----------
+NL: copy file `srcfile` to directory `dstdir`
+ref: shutil.copy(srcfile, dstdir)
+Output: ▁shutil.copy(srcfile, dstdir)
+bleu_metric: 80.70557274927982
+----------
+NL: find the key associated with the largest value in dictionary `x` whilst key is non-zero value
+ref: max(k for k, v in x.items() if v != 0)
+Output: max(x,▁key=lambda x: x[0])
+bleu_metric: 11.483394131362408
+----------
+NL: get the largest key whose not associated with value of 0 in dictionary `x`
+ref: (k for k, v in x.items() if v != 0)
+Output: max(k for k, v in x.items()▁if v▁!= 0)
+bleu_metric: 72.00242075875519
+----------
+NL: get the largest key in a dictionary `x` with non-zero value
+ref: max(k for k, v in x.items() if v != 0)
+Output: max(k for k, v in x.items()▁if v▁!= 0)
+bleu_metric: 76.74174160136337
+----------
+NL: put the curser at beginning of the file
+ref: file.seek(0)
+Output: time.sleep(1)
+bleu_metric: 27.776190340117914
+----------
+NL: combine values from column 'b' and column 'a' of dataframe `df` into column 'c' of datafram `df`
+ref: df['c'] = np.where(df['a'].isnull, df['b'], df['a'])
+Output: df['C'] = df['B'].apply(lambda x: x.astype(int))
+bleu_metric: 15.164081046360186
+----------
+NL: remove key 'ele' from dictionary `d`
+ref: del d['ele']
+Output: d =▁dict((k, v) for k, v in d.items()▁if v)
+bleu_metric: 5.377336385080629
+----------
+NL: update datetime field in `mymodel` to be the existing `timestamp` plus 100 years
+ref: MyModel.objects.update(timestamp=F('timestamp') + timedelta(days=36524.25))
+Output: time.strftime('%Y-%m-%d %H:%M:%S', time.gmtime(
+bleu_metric: 7.278748444764005
+----------
+NL: merge list `['it']` and list `['was']` and list `['annoying']` into one list
+ref: ['it'] + ['was'] + ['annoying']
+Output: ▁list(zip([1'it'], ['is'], ['annoYing']))
+bleu_metric: 19.158743349944555
+----------
+NL: increment a value with leading zeroes in a number `x`
+ref: str(int(x) + 1).zfill(len(x))
+Output: ▁numpy.zeros((x + 1) + x)
+bleu_metric: 19.950354926643225
+----------
+NL: check if a pandas dataframe `df`'s index is sorted
+ref: all(df.index[:-1] <= df.index[1:])
+Output: df.index.sort()
+bleu_metric: 7.510537410248026
+----------
+NL: convert tuple `t` to list
+ref: list(t)
+Output: ▁list(t)
+bleu_metric: 68.03749333171201
+----------
+NL: convert list `t` to tuple
+ref: tuple(l)
+Output: tuple(t)
+bleu_metric: 45.180100180492246
+----------
+NL: convert tuple `level1` to list
+ref: level1 = map(list, level1)
+Output: ▁list(level1.from_tuples())
+bleu_metric: 17.827531042796256
+----------
+NL: send the output of pprint object `dataobject` to file `logfile`
+ref: pprint.pprint(dataobject, logFile)
+Output: pprint(dataobject, logfile)
+bleu_metric: 38.73479811032991
+----------
+NL: get index of rows in column 'boolcol'
+ref: df.loc[df['BoolCol']]
+Output: df.index
+bleu_metric: 2.5225322607845957
+----------
+NL: create a list containing the indexes of rows where the value of column 'boolcol' in dataframe `df` are equal to true
+ref: df.iloc[np.flatnonzero(df['BoolCol'])]
+Output: df[df.groupby(['boolcol', 'value'])[0].tolist()]
+bleu_metric: 17.4932986566737
+----------
+NL: get list of indexes of rows where column 'boolcol' values match true
+ref: df[df['BoolCol'] == True].index.tolist()
+Output: [i for i, j in enumerate(['boolcol', 'true'])▁if j == j]
+bleu_metric: 9.579371674972275
+----------
+NL: get index of rows in dataframe `df` which column 'boolcol' matches value true
+ref: df[df['BoolCol']].index.tolist()
+Output: df.index = df.loc[df['boolcol'] ==▁value]
+bleu_metric: 23.891455818410613
+----------
+NL: change working directory to the directory `owd`
+ref: os.chdir(owd)
+Output: os.chdir('\\Users\\Desktop\\Desktop\\Desktop\\Desk
+bleu_metric: 17.001078098404225
+----------
+NL: insert data from a string `testfield` to sqlite db `c`
+ref: c.execute("INSERT INTO test VALUES (?, 'bar')", (testfield,))
+Output: ▁cursor.execute('INSERT▁INTO▁table▁VALUES (%s, %s, %s)', (testfield,)
+bleu_metric: 36.151920269854195
+----------
+NL: decode string "\\x89\\n" into a normal string
+ref: """\\x89\\n""".decode('string_escape')
+Output: """\\x89\\n""".decode('utf-8')
+bleu_metric: 79.47545184555568
+----------
+NL: convert a raw string `raw_string` into a normal string
+ref: raw_string.decode('string_escape')
+Output: ▁raw_string.decode('unicode_escape')
+bleu_metric: 53.107253497886994
+----------
+NL: convert a raw string `raw_byte_string` into a normal string
+ref: raw_byte_string.decode('unicode_escape')
+Output: ▁raw_byte_string.decode('unicode_escape')
+bleu_metric: 80.70557274927982
+----------
+NL: split a string `s` with into all strings of repeated characters
+ref: [m.group(0) for m in re.finditer('(\\d)\\1*', s)]
+Output: re.findall('[^a-zA-Z]*', s)
+bleu_metric: 17.036282985955026
+----------
+NL: scatter a plot with x, y position of `np.random.randn(100)` and face color equal to none
+ref: plt.scatter(np.random.randn(100), np.random.randn(100), facecolors='none')
+Output: plt.show()
+bleu_metric: 0.8278521686370256
+----------
+NL: do a scatter plot with empty circles
+ref: plt.plot(np.random.randn(100), np.random.randn(100), 'o', mfc='none')
+Output: plt.show()
+bleu_metric: 0.42503347540650505
+----------
+NL: remove a div with a id `main-content` using beautifulsoup
+ref: soup.find('div', id='main-content').decompose()
+Output: ▁print(soup.findAll('div', class_='maincontent'))
+bleu_metric: 34.193725219509716
+----------
+NL: filter rows containing key word `ball` in column `ids`
+ref: df[df['ids'].str.contains('ball')]
+Output: [(k, v) for k, v in df.ids]
+bleu_metric: 11.123500809946101
+----------
+NL: convert index at level 0 into a column in dataframe `df`
+ref: df.reset_index(level=0, inplace=True)
+Output: df['index'] = df['index'].apply(lambda x: x.tolist())
+bleu_metric: 7.592939129822969
+----------
+NL: add indexes in a data frame `df` to a column `index1`
+ref: df['index1'] = df.index
+Output: df['index1'] = df['index1'].index
+bleu_metric: 55.54570250728591
+----------
+NL: convert pandas index in a dataframe to columns
+ref: df.reset_index(level=['tick', 'obs'])
+Output: df.set_index(['index', 'index'])
+bleu_metric: 32.51279032103401
+----------
+NL: get reverse of list items from list 'b' using extended slicing
+ref: [x[::-1] for x in b]
+Output: [x[i:i + 2] for i in range(0, len(x), 2)]
+bleu_metric: 17.895706401541528
+----------
+NL: join each element in array `a` with element at the same index in array `b` as a tuple
+ref: np.array([zip(x, y) for x, y in zip(a, b)])
+Output: A = [(a + b) for a, b in zip(A, B)]
+bleu_metric: 18.114844247735682
+----------
+NL: zip two 2-d arrays `a` and `b`
+ref: np.array(zip(a.ravel(), b.ravel()), dtype='i4,i4').reshape(a.shape)
+Output: np.array(zip(a, b))
+bleu_metric: 10.156383094063225
+----------
+NL: convert list `list_of_ints` into a comma separated string
+ref: """,""".join([str(i) for i in list_of_ints])
+Output: """""".join(list_of_ints)
+bleu_metric: 31.139742129493257
+----------
+NL: send a post request with raw data `data` and basic authentication with `username` and `password`
+ref: requests.post(url, data=DATA, headers=HEADERS_DICT, auth=(username, password))
+Output: r =▁requests.post(url, data=data,▁headers=headers,▁files=passwords)
+bleu_metric: 38.57153091307505
+----------
+NL: find last occurrence of character '}' in string "abcd}def}"
+ref: 'abcd}def}'.rfind('}')
+Output: """abcd}def}""".rfind('}.rsplit('\\d+')
+bleu_metric: 42.26469376227503
+----------
+NL: iterate ove list `[1, 2, 3]` using list comprehension
+ref: print([item for item in [1, 2, 3]])
+Output: ▁print(list(itertools.chain.from_iterable([1, 2, 3])))
+bleu_metric: 35.3003111274967
+----------
+NL: extract all the values with keys 'x' and 'y' from a list of dictionaries `d` to list of tuples
+ref: [(x['x'], x['y']) for x in d]
+Output: [d['x'], d['y'], d['d'], d['x'], d[
+bleu_metric: 30.398382447928807
+----------
+NL: get the filename without the extension from file 'hemanth.txt'
+ref: print(os.path.splitext(os.path.basename('hemanth.txt'))[0])
+Output: ▁print(os.path.splitext('hemanth.txt'))
+bleu_metric: 51.55938317965318
+----------
+NL: create a dictionary by adding each two adjacent elements in tuple `x` as key/value pair to it
+ref: dict(x[i:i + 2] for i in range(0, len(x), 2))
+Output: ▁dict(zip(x[::2], x[1::2]))
+bleu_metric: 16.45452192914717
+----------
+NL: create a list containing flattened list `[['a', 'b', 'c'], ['d', 'e', 'f'], ['g', 'h', 'i']]`
+ref: values = sum([['A', 'B', 'C'], ['D', 'E', 'F'], ['G', 'H', 'I']], [])
+Output: [['a', 'b', 'c'], ['d', 'e', 'f'], ['
+bleu_metric: 22.929399628621095
+----------
+NL: select rows in a dataframe `df` column 'closing_price' between two values 99 and 101
+ref: df = df[(df['closing_price'] >= 99) & (df['closing_price'] <= 101)]
+Output: df.loc[df['cold_Price'] == 99]
+bleu_metric: 9.613222089577718
+----------
+NL: replace all occurences of newlines `\n` with `
` in dataframe `df`
+ref: df.replace({'\n': '
'}, regex=True)
+Output: df = pd.read_csv('\n').replace('\n', '
')
+bleu_metric: 33.87144423286913
+----------
+NL: replace all occurrences of a string `\n` by string `
` in a pandas data frame `df`
+ref: df.replace({'\n': '
'}, regex=True)
+Output: df = pd.read_csv('\n').replace('\n', '
')
+bleu_metric: 33.87144423286913
+----------
+NL: create a list containing each two adjacent letters in string `word` as its elements
+ref: [(x + y) for x, y in zip(word, word[1:])]
+Output: [word[i:i + 2] for i in range(0, len(word), 2)]
+bleu_metric: 15.253223052326929
+----------
+NL: get a list of pairs from a string `word` using lambda function
+ref: list(map(lambda x, y: x + y, word[:-1], word[1:]))
+Output: ▁list(word.split(','))
+bleu_metric: 7.965309748053327
+----------
+NL: extract a url from a string `mystring`
+ref: print(re.findall('(https?://[^\\s]+)', myString))
+Output: re.findall('[^a-zA-Z]*', myString)
+bleu_metric: 30.515239072390138
+----------
+NL: extract a url from a string `mystring`
+ref: print(re.search('(?Phttps?://[^\\s]+)', myString).group('url'))
+Output: re.findall('[^a-zA-Z]*', myString)
+bleu_metric: 12.326746139034004
+----------
+NL: remove all special characters, punctuation and spaces from a string `mystring` using regex
+ref: re.sub('[^A-Za-z0-9]+', '', mystring)
+Output: """""".join(re.sub('\\w+', '', mystring))
+bleu_metric: 40.6731552331291
+----------
+NL: create a datetimeindex containing 13 periods of the second friday of each month starting from date '2016-01-01'
+ref: pd.date_range('2016-01-01', freq='WOM-2FRI', periods=13)
+Output: ▁datetime.datetime.strptime('2012-01-01', '2012-01-01', '%Y-%m-%d
+bleu_metric: 21.54897801937284
+----------
+NL: create multidimensional array `matrix` with 3 rows and 2 columns in python
+ref: matrix = [[a, b], [c, d], [e, f]]
+Output: ▁matrix = [[[0, 0], [0, 0], [0, 0], [0, 0]], [[0, 0]], [
+bleu_metric: 15.448396990132697
+----------
+NL: replace spaces with underscore
+ref: mystring.replace(' ', '_')
+Output: re.sub('(?<=[a-z])(?=[A-Z])', '\\1'
+bleu_metric: 9.500266321476488
+----------
+NL: get an absolute file path of file 'mydir/myfile.txt'
+ref: os.path.abspath('mydir/myfile.txt')
+Output: os.path.abspath('mydir/myfile.txt')
+bleu_metric: 100.0
+----------
+NL: split string `my_string` on white spaces
+ref: """ """.join(my_string.split())
+Output: """ """.join(my_string.split())
+bleu_metric: 100.0
+----------
+NL: get filename without extension from file `filename`
+ref: os.path.splitext(filename)[0]
+Output: os.path.splitext(filename)
+bleu_metric: 64.74591278836638
+----------
+NL: get a list containing the sum of each element `i` in list `l` plus the previous elements
+ref: [sum(l[:i]) for i, _ in enumerate(l)]
+Output: [sum(i) for i, j in zip(l, l)]
+bleu_metric: 41.46278264707056
+----------
+NL: split a string `docs/src/scripts/temp` by `/` keeping `/` in the result
+ref: """Docs/src/Scripts/temp""".replace('/', '/\x00/').split('\x00')
+Output: ▁print(docs/src/scripts/temp.split('/'))
+bleu_metric: 17.379851709438395
+----------
+NL: shuffle columns of an numpy array 'r'
+ref: np.random.shuffle(np.transpose(r))
+Output: np.random.shuffle(np.random.randn(r))
+bleu_metric: 69.64705665515707
+----------
+NL: copy all values in a column 'b' to a new column 'd' in a pandas data frame 'df'
+ref: df['D'] = df['B']
+Output: df['d'] = df['d'].values.tolist()
+bleu_metric: 37.392149096896674
+----------
+NL: find a value within nested json 'data' where the key inside another key 'b' is unknown.
+ref: list(data['A']['B'].values())[0]['maindata'][0]['Info']
+Output: [item for item in data▁if item['b'] not in item]
+bleu_metric: 5.9434130655282855
+----------
+NL: check characters of string `string` are true predication of function `predicate`
+ref: all(predicate(x) for x in string)
+Output: ▁any(pred(x) for x in▁string.split())
+bleu_metric: 33.92375549949802
+----------
+NL: determine number of files on a drive with python
+ref: os.statvfs('/').f_files - os.statvfs('/').f_ffree
+Output: ▁sum(i for i in os.listdir('filename')▁if os.path.isfile(filename))
+bleu_metric: 10.12895233093219
+----------
+NL: how to get a single result from a sqlite query in python?
+ref: cursor.fetchone()[0]
+Output: ▁cursor.execute('INSERT▁INTO▁table▁VALUES (%s, %s, %s)', (var1, var2,
+bleu_metric: 5.912329024877923
+----------
+NL: convert string `user_input` into a list of integers `user_list`
+ref: user_list = [int(number) for number in user_input.split(',')]
+Output: user_list = [int(i) for i in▁user_input.split(',')]
+bleu_metric: 66.43548861507492
+----------
+NL: get a list of integers by splitting a string `user` with comma
+ref: [int(s) for s in user.split(',')]
+Output: [int(x) for x in▁user.split(',')]
+bleu_metric: 64.55651607183006
+----------
+NL: sorting a python list by two criteria
+ref: sorted(list, key=lambda x: (x[0], -x[1]))
+Output: a.sort(key=lambda x: (x[0], x[0]), reverse=True)
+bleu_metric: 49.45703137422567
+----------
+NL: sort a list of objects `ut`, based on a function `cmpfun` in descending order
+ref: ut.sort(key=cmpfun, reverse=True)
+Output: ▁sorted(ut,▁key=lambda x: func(x, -1))
+bleu_metric: 10.813005337959174
+----------
+NL: reverse list `ut` based on the `count` attribute of each object
+ref: ut.sort(key=lambda x: x.count, reverse=True)
+Output: ▁print(reversed(ut,▁key=lambda x: x.count(), reverse=True))
+bleu_metric: 45.26353562390146
+----------
+NL: sort a list of objects `ut` in reverse order by their `count` property
+ref: ut.sort(key=lambda x: x.count, reverse=True)
+Output: ▁sorted(ut,▁key=lambda x:▁count.count(x[0], x[0]))
+bleu_metric: 20.563241139732103
+----------
+NL: click a href button 'send' with selenium
+ref: driver.find_element_by_partial_link_text('Send').click()
+Output: ▁driver.find_element_by_css_selector('send')
+bleu_metric: 20.083909447408573
+----------
+NL: click a href button having text `send inmail` with selenium
+ref: driver.findElement(By.linkText('Send InMail')).click()
+Output: ▁driver.find_element_by_css_selector('send inmail.body.send').click()
+bleu_metric: 32.326275289498305
+----------
+NL: click a href button with text 'send inmail' with selenium
+ref: driver.find_element_by_link_text('Send InMail').click()
+Output: ▁driver.find_element_by_css_selector('send inmail').click()
+bleu_metric: 48.80217780091719
+----------
+NL: cast an int `i` to a string and concat to string 'me'
+ref: 'ME' + str(i)
+Output: """""".join(chr(i) for i in range(10))
+bleu_metric: 13.308442527111023
+----------
+NL: sorting data in dataframe pandas
+ref: df.sort_values(['System_num', 'Dis'])
+Output: df.sort_values(['year','month'],▁ascending=[True, False])
+bleu_metric: 34.52186471009711
+----------
+NL: prepend the line '#test firstline\n' to the contents of file 'infile' and save as the file 'outfile'
+ref: open('outfile', 'w').write('#test firstline\n' + open('infile').read())
+Output: open('infile', '#test firstline\n')
+bleu_metric: 23.71937737244433
+----------
+NL: sort a list `l` by length of value in tuple
+ref: l.sort(key=lambda t: len(t[1]), reverse=True)
+Output: ▁sorted(l,▁key=len)
+bleu_metric: 8.655647308002333
+----------
+NL: split string `s` by words that ends with 'd'
+ref: re.findall('\\b(\\w+)d\\b', s)
+Output: re.split('\\s+', s)
+bleu_metric: 28.449457729768916
+----------
+NL: return `true` if string `foobarrrr` contains regex `ba[rzd]`
+ref: bool(re.search('ba[rzd]', 'foobarrrr'))
+Output: ▁print(re.findall('\\b[A-Z]', 'foobarrarrrrrr'))
+bleu_metric: 29.452884374019465
+----------
+NL: removing duplicates in list `t`
+ref: list(set(t))
+Output: ▁list(set(t))
+bleu_metric: 78.60753021519787
+----------
+NL: removing duplicates in list `source_list`
+ref: list(set(source_list))
+Output: ▁print([item for item in▁list(source_list)▁if item.endswith('item')])
+bleu_metric: 11.295714543953919
+----------
+NL: removing duplicates in list `abracadabra`
+ref: list(OrderedDict.fromkeys('abracadabra'))
+Output: [dict((x / y) for x, y in▁itertools.groupby(lambda x: x.group(0))
+bleu_metric: 8.635800047213174
+----------
+NL: convert array `a` into a list
+ref: numpy.array(a).reshape(-1).tolist()
+Output: ▁list(array.from_iterable(a))
+bleu_metric: 16.778292801990535
+----------
+NL: convert the first row of numpy matrix `a` to a list
+ref: numpy.array(a)[0].tolist()
+Output: a = np.array(a)
+bleu_metric: 32.2645258949984
+----------
+NL: in `soup`, get the content of the sibling of the `td` tag with text content `address:`
+ref: print(soup.find(text='Address:').findNext('td').contents[0])
+Output: soup.find_element_by_xpath("//p[@id='id']/text()/
+bleu_metric: 9.699583930275937
+----------
+NL: convert elements of each tuple in list `l` into a string separated by character `@`
+ref: """ """.join([('%d@%d' % t) for t in l])
+Output: """""".join(l[0] for l in zip(l[::2], l[1::2]))
+bleu_metric: 32.31269177549822
+----------
+NL: convert each tuple in list `l` to a string with '@' separating the tuples' elements
+ref: """ """.join([('%d@%d' % (t[0], t[1])) for t in l])
+Output: [(x, y) for x, y in zip(l, l[::2], l[1::2])]
+bleu_metric: 8.274984205902886
+----------
+NL: get the html from the current web page of a selenium driver
+ref: driver.execute_script('return document.documentElement.outerHTML;')
+Output: ▁driver.find_element_by_css_selector('http://www.google.com/')
+bleu_metric: 16.24355752882384
+----------
+NL: get all matches with regex pattern `\\d+[xx]` in list of string `teststr`
+ref: [i for i in teststr if re.search('\\d+[xX]', i)]
+Output: re.findall('\\d+[xx]', teststr)
+bleu_metric: 29.180753193172116
+----------
+NL: select values from column 'a' for which corresponding values in column 'b' will be greater than 50, and in column 'c' - equal 900 in dataframe `df`
+ref: df['A'][(df['B'] > 50) & (df['C'] == 900)]
+Output: df['b'].str[0].isin(['c', 'a'])
+bleu_metric: 12.146852493064019
+----------
+NL: sort dictionary `o` in ascending order based on its keys and items
+ref: sorted(o.items())
+Output: ▁sorted(list(o.items()),▁key=lambda x: x[1])
+bleu_metric: 29.452884374019472
+----------
+NL: get sorted list of keys of dict `d`
+ref: sorted(d)
+Output: ▁sorted(list(d.items()),▁key=lambda x: x[1])
+bleu_metric: 9.043565779655339
+----------
+NL: how to sort dictionaries by keys in python
+ref: sorted(d.items())
+Output: ▁sorted(list(d.items()),▁key=lambda x: x[1])
+bleu_metric: 29.452884374019472
+----------
+NL: convert string "1" into integer
+ref: int('1')
+Output: int('1', 2)
+bleu_metric: 58.143073696821936
+----------
+NL: function to convert strings into integers
+ref: int()
+Output: int(''.join(map(str, s)))
+bleu_metric: 12.729922658368398
+----------
+NL: convert items in `t1` to integers
+ref: T2 = [map(int, x) for x in T1]
+Output: int(''.join(map(str, t1)))
+bleu_metric: 12.512236921161914
+----------
+NL: call a shell script `./test.sh` using subprocess
+ref: subprocess.call(['./test.sh'])
+Output: subprocess.call(['./test.sh'])
+bleu_metric: 100.0
+----------
+NL: call a shell script `notepad` using subprocess
+ref: subprocess.call(['notepad'])
+Output: subprocess.call(['notepad'])
+bleu_metric: 100.0
+----------
+NL: combine lists `l1` and `l2` by alternating their elements
+ref: [val for pair in zip(l1, l2) for val in pair]
+Output: [(x + y) for x, y in zip(l1, l2)]
+bleu_metric: 40.59264050928986
+----------
+NL: encode string 'data to be encoded'
+ref: encoded = base64.b64encode('data to be encoded')
+Output: 'data to be encoded'.encode('data to be encoded')
+bleu_metric: 50.495676290868666
+----------
+NL: encode a string `data to be encoded` to `ascii` encoding
+ref: encoded = 'data to be encoded'.encode('ascii')
+Output: ▁print(ascii.encode('data to be encoded'))
+bleu_metric: 51.09375879171261
+----------
+NL: parse tab-delimited csv file 'text.txt' into a list
+ref: lol = list(csv.reader(open('text.txt', 'rb'), delimiter='\t'))
+Output: ['text.txt', 'rb').split(',')
+bleu_metric: 35.4953120381003
+----------
+NL: get attribute `my_str` of object `my_object`
+ref: getattr(my_object, my_str)
+Output: ▁getattr(my_str, my_object)
+bleu_metric: 26.26909894424158
+----------
+NL: group a list of dicts `ld` into one dict by key
+ref: print(dict(zip(LD[0], zip(*[list(d.values()) for d in LD]))))
+Output: mydict = {k: v for k, v in▁list(d.items())}
+bleu_metric: 13.839118500778957
+----------
+NL: how do i sum the first value in each tuple in a list of tuples in python?
+ref: sum([pair[0] for pair in list_of_pairs])
+Output: ▁sum(x[0] for x in▁rows)
+bleu_metric: 30.130404892785684
+----------
+NL: convert unicode string u"{'code1':1,'code2':1}" into dictionary
+ref: d = ast.literal_eval("{'code1':1,'code2':1}")
+Output: ▁dict(('{'code1':1,'code2':1}'.encode('unicode_escape')
+bleu_metric: 61.32878268086118
+----------
+NL: find all words in a string `mystring` that start with the `$` sign
+ref: [word for word in mystring.split() if word.startswith('$')]
+Output: re.findall('\\w+', mystring)
+bleu_metric: 8.395097551291249
+----------
+NL: remove any url within string `text`
+ref: text = re.sub('^https?:\\/\\/.*[\\r\\n]*', '', text, flags=re.MULTILINE)
+Output: ▁url = re.sub('▁?\\s+', '', text)
+bleu_metric: 21.464898124935406
+----------
+NL: replace all elements in array `a` that are not present in array `[1, 3, 4]` with zeros
+ref: np.where(np.in1d(A, [1, 3, 4]).reshape(A.shape), A, 0)
+Output: np.where(np.isnan(a), [1, 3, 4])
+bleu_metric: 40.9574484442826
+----------
+NL: calculate mean across dimension in a 2d array `a`
+ref: np.mean(a, axis=1)
+Output: np.mean(a,▁axis=1)
+bleu_metric: 76.11606003349893
+----------
+NL: running r script '/pathto/myrscript.r' from python
+ref: subprocess.call(['/usr/bin/Rscript', '--vanilla', '/pathto/MyrScript.r'])
+Output: os.system('/pathto/myrscript.r')
+bleu_metric: 10.462313403993925
+----------
+NL: run r script '/usr/bin/rscript --vanilla /pathto/myrscript.r'
+ref: subprocess.call('/usr/bin/Rscript --vanilla /pathto/MyrScript.r', shell=True)
+Output: os.system('/usr/bin/rscript --vanilla /pathto/myrscript.r')
+bleu_metric: 45.63981491147585
+----------
+NL: add a header to a csv file
+ref: writer.writeheader()
+Output: ▁writer.writeheader()
+bleu_metric: 72.59795291154771
+----------
+NL: replacing nan in the dataframe `df` with row average
+ref: df.fillna(df.mean(axis=1), axis=1)
+Output: df.fillna(lambda x: x.fillna(x.mean(),▁axis=1))
+bleu_metric: 31.449494592022948
+----------
+NL: convert unix timestamp '1347517370' to formatted string '%y-%m-%d %h:%m:%s'
+ref: time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(1347517370))
+Output: """%Y-%m-%d %H:%M:%S')
+bleu_metric: 57.04022584361268
+----------
+NL: call a base class's class method `do` from derived class `derived`
+ref: super(Derived, cls).do(a)
+Output: super(Do,▁self).__init__(*args,▁**kwargs)
+bleu_metric: 13.26475916741222
+----------
+NL: selecting rows in numpy ndarray 'a', where the value in the first column is 0 and value in the second column is 1
+ref: a[np.where((a[:, (0)] == 0) * (a[:, (1)] == 1))]
+Output: A[np.isnan(a[:, (1)])]
+bleu_metric: 21.823239874972007
+----------
+NL: separate words delimited by one or more spaces into a list
+ref: re.split(' +', 'hello world sample text')
+Output: [' '.join(l) for l in re.split('\\s+','', s)]
+bleu_metric: 23.977362037775197
+----------
+NL: length of longest element in list `words`
+ref: len(max(words, key=len))
+Output: len(words) - len(words)
+bleu_metric: 21.258637840736228
+----------
+NL: get the value associated with unicode key 'from_user' of first dictionary in list `result`
+ref: result[0]['from_user']
+Output: ▁print([x['from_user'] for x in▁list(result.items())])
+bleu_metric: 19.427446513842177
+----------
+NL: retrieve each line from a file 'file.txt' as a list
+ref: [line.split() for line in open('File.txt')]
+Output: open('file.txt', 'r')
+bleu_metric: 16.725016271836772
+----------
+NL: swap keys with values in a dictionary `a`
+ref: res = dict((v, k) for k, v in a.items())
+Output: ▁dict((v, k) for k, v in a.items())
+bleu_metric: 90.21315316859231
+----------
+NL: open a file `path/to/file_name.ext` in write mode
+ref: new_file = open('path/to/FILE_NAME.ext', 'w')
+Output: os.system('path/to/file_name.ext')
+bleu_metric: 40.33687971800765
+----------
+NL: how to count distinct values in a column of a pandas group by object?
+ref: df.groupby(['col1', 'col2'])['col3'].nunique().reset_index()
+Output: df.groupby(['col1', 'col2']).apply(lambda x: x.groupby('
+bleu_metric: 54.116466213090966
+----------
+NL: check if any key in the dictionary `dict1` starts with the string `emp$$`
+ref: any(key.startswith('EMP$$') for key in dict1)
+Output: ▁any(k for k, v in▁dict1.items()▁if 'emp$$' in k)
+bleu_metric: 13.217947626377288
+----------
+NL: create list of values from dictionary `dict1` that have a key that starts with 'emp$$'
+ref: [value for key, value in list(dict1.items()) if key.startswith('EMP$$')]
+Output: [d['emp$'] for d in▁dict1.values()]
+bleu_metric: 12.254981271073301
+----------
+NL: convert a pandas series `sf` into a pandas dataframe `df` with columns `email` and `list`
+ref: pd.DataFrame({'email': sf.index, 'list': sf.values})
+Output: df = pd.DataFrame(sf,▁columns=['email', 'list'])
+bleu_metric: 35.69768415530451
+----------
+NL: print elements of list `list` seperated by tabs `\t`
+ref: print('\t'.join(map(str, list)))
+Output: ▁print('\t'.join(map(str,▁list)))
+bleu_metric: 77.39321540474097
+----------
+NL: print unicode string '\xd0\xbf\xd1\x80\xd0\xb8' with utf-8
+ref: print('\xd0\xbf\xd1\x80\xd0\xb8'.encode('raw_unicode_escape'))
+Output: ▁print('\xd0\xbf\xd1\x80\xd0\xb8'.encode(
+bleu_metric: 75.96292058519337
+----------
+NL: encode a latin character in string `sopet\xc3\xb3n` properly
+ref: 'Sopet\xc3\xb3n'.encode('latin-1').decode('utf-8')
+Output: """sopet\xc3\xb3n""".encode('latin')
+bleu_metric: 38.377836342588495
+----------
+NL: resized image `image` to width, height of `(x, y)` with filter of `antialias`
+ref: image = image.resize((x, y), Image.ANTIALIAS)
+Output: ▁image.reset_image(image, (x, y))
+bleu_metric: 34.79998616163817
+----------
+NL: regex, find "n"s only in the middle of string `s`
+ref: re.findall('n(?<=[^n]n)n+(?=[^n])(?i)', s)
+Output: re.findall('\\d+', s)
+bleu_metric: 13.004547755725982
+----------
+NL: display the float `1/3*100` as a percentage
+ref: print('{0:.0f}%'.format(1.0 / 3 * 100))
+Output: ▁print('{0:.2f}'.format(1/3*100))
+bleu_metric: 65.24084871752925
+----------
+NL: sort a list of dictionary `mylist` by the key `title`
+ref: mylist.sort(key=lambda x: x['title'])
+Output: mylist.sort(key=lambda x: x['title'])
+bleu_metric: 100.0
+----------
+NL: sort a list `l` of dicts by dict value 'title'
+ref: l.sort(key=lambda x: x['title'])
+Output: ▁sorted(l,▁key=itemgetter('title', 'title'))
+bleu_metric: 15.448759309616733
+----------
+NL: sort a list of dictionaries by the value of keys 'title', 'title_url', 'id' in ascending order.
+ref: l.sort(key=lambda x: (x['title'], x['title_url'], x['id']))
+Output: ▁sorted(list_of_dicts,▁key=lambda x: x['title']['title'], reverse=Tru
+bleu_metric: 34.214468780181804
+----------
+NL: find 10 largest differences between each respective elements of list `l1` and list `l2`
+ref: heapq.nlargest(10, range(len(l1)), key=lambda i: abs(l1[i] - l2[i]))
+Output: max(l1,▁key=lambda x: (x[1], x[0]), reverse=True)
+bleu_metric: 11.878881948379775
+----------
+NL: beautifulsoup find all 'span' elements in html string `soup` with class of 'stargryb sp'
+ref: soup.find_all('span', {'class': 'starGryB sp'})
+Output: soup.find_all('span', class_='stargryb.txt')
+bleu_metric: 37.90023910380332
+----------
+NL: write records in dataframe `df` to table 'test' in schema 'a_schema'
+ref: df.to_sql('test', engine, schema='a_schema')
+Output: df.to_csv('test')
+bleu_metric: 23.118543024585154
+----------
+NL: extract brackets from string `s`
+ref: brackets = re.sub('[^(){}[\\]]', '', s)
+Output: re.findall('\\[(\\w+)\\]', s)
+bleu_metric: 28.446938709683227
+----------
+NL: remove duplicate elements from list 'l'
+ref: list(dict((x[0], x) for x in L).values())
+Output: [x for x in l▁if x not in x]
+bleu_metric: 11.000288081592606
+----------
+NL: read a file `file` without newlines
+ref: [line.rstrip('\n') for line in file]
+Output: open('file', 'w').readline()
+bleu_metric: 13.399118318121793
+----------
+NL: get the position of item 1 in `testlist`
+ref: [i for (i, x) in enumerate(testlist) if (x == 1)]
+Output: ▁print(testlist.position())
+bleu_metric: 6.435022756527756
+----------
+NL: get the position of item 1 in `testlist`
+ref: [i for (i, x) in enumerate(testlist) if (x == 1)]
+Output: ▁print(testlist.position())
+bleu_metric: 6.435022756527756
+----------
+NL: get the position of item 1 in `testlist`
+ref: for i in [i for (i, x) in enumerate(testlist) if (x == 1)]:
+ pass
+Output: ▁print(testlist.position())
+bleu_metric: 4.559019536899008
+----------
+NL: get the position of item 1 in `testlist`
+ref: for i in (i for (i, x) in enumerate(testlist) if (x == 1)):
+ pass
+Output: ▁print(testlist.position())
+bleu_metric: 5.045384493046664
+----------
+NL: get the position of item 1 in `testlist`
+ref: gen = (i for (i, x) in enumerate(testlist) if (x == 1))
+for i in gen:
+ pass
+Output: ▁print(testlist.position())
+bleu_metric: 3.8410531655908353
+----------
+NL: get the position of item `element` in list `testlist`
+ref: print(testlist.index(element))
+Output: ▁print(testlist.index(element))
+bleu_metric: 82.42367502646054
+----------
+NL: get the position of item `element` in list `testlist`
+ref: try:
+ print(testlist.index(element))
+except ValueError:
+ pass
+Output: ▁print(testlist.index(element))
+bleu_metric: 52.74890952658116
+----------
+NL: find the first element of the tuple with the maximum second element in a list of tuples `lis`
+ref: max(lis, key=lambda item: item[1])[0]
+Output: max(lis,▁key=lambda x: x[1])[0]
+bleu_metric: 60.39435155169263
+----------
+NL: get the item at index 0 from the tuple that has maximum value at index 1 in list `lis`
+ref: max(lis, key=itemgetter(1))[0]
+Output: max(enumerate(a),▁key=lambda x: x[1])[0]
+bleu_metric: 27.274191069381914
+----------
+NL: make a delay of 1 second
+ref: time.sleep(1)
+Output: time.sleep(1)
+bleu_metric: 100.0
+----------
+NL: convert list of tuples `l` to a string
+ref: """, """.join('(' + ', '.join(i) + ')' for i in L)
+Output: """""".join(l)
+bleu_metric: 14.164334116374599
+----------
+NL: django set default value of field `b` equal to '0000000'
+ref: b = models.CharField(max_length=7, default='0000000', editable=False)
+Output: super(B,▁self).save(*args,▁**kwargs)
+bleu_metric: 8.945053454772305
+----------
+NL: sort lis `list5` in ascending order based on the degrees value of its elements
+ref: sorted(list5, lambda x: (degree(x), x))
+Output: ▁sorted(list5,▁key=lambda x: int(x[0]))
+bleu_metric: 28.571962561926444
+----------
+NL: how do i perform secondary sorting in python?
+ref: sorted(list5, key=lambda vertex: (degree(vertex), vertex))
+Output: ▁sorted(list(range(10)),▁key=lambda x: x[1])
+bleu_metric: 15.94000950514835
+----------
+NL: convert a list into a generator object
+ref: (n for n in [1, 2, 3, 5])
+Output: ▁list(itertools.product(list(range(10)),▁repeat=4))
+bleu_metric: 8.282282660969601
+----------
+NL: remove elements from list `oldlist` that have an index number mentioned in list `removelist`
+ref: newlist = [v for i, v in enumerate(oldlist) if i not in removelist]
+Output: [x for x in a▁if x not in b]
+bleu_metric: 11.298101090980193
+----------
+NL: open a file `yourfile.txt` in write mode
+ref: f = open('yourfile.txt', 'w')
+Output: open('yourfile.txt', 'r')
+bleu_metric: 67.09548263862051
+----------
+NL: get attribute 'attr' from object `obj`
+ref: getattr(obj, 'attr')
+Output: ▁getattr(obj, 'attr')
+bleu_metric: 80.70557274927982
+----------
+NL: convert tuple of tuples `(('aa',), ('bb',), ('cc',))` to tuple
+ref: from functools import reduce
+reduce(lambda a, b: a + b, (('aa',), ('bb',), ('cc',)))
+Output: tuple(('aa',), ('bb',), ('cc',))
+bleu_metric: 50.621900464488824
+----------
+NL: convert tuple of tuples `(('aa',), ('bb',), ('cc',))` to list in one line
+ref: map(lambda a: a[0], (('aa',), ('bb',), ('cc',)))
+Output: ▁list(('aa',), ('bb',), ('cc',))
+bleu_metric: 64.0614517827111
+----------
+NL: python pandas: how to replace a characters in a column of a dataframe?
+ref: df['range'].replace(',', '-', inplace=True)
+Output: df.apply(lambda x: x.replace(',', '').replace(',', ''))
+bleu_metric: 30.713730826344694
+----------
+NL: unzip the list `[('a', 1), ('b', 2), ('c', 3), ('d', 4)]`
+ref: zip(*[('a', 1), ('b', 2), ('c', 3), ('d', 4)])
+Output: zip([('a', 1), ('b', 2), ('c', 3), ('d', 4)]
+bleu_metric: 89.10617687473993
+----------
+NL: unzip the list `[('a', 1), ('b', 2), ('c', 3), ('d', 4)]`
+ref: zip(*[('a', 1), ('b', 2), ('c', 3), ('d', 4)])
+Output: zip([('a', 1), ('b', 2), ('c', 3), ('d', 4)]
+bleu_metric: 89.10617687473993
+----------
+NL: unzip list `original`
+ref: result = ([a for (a, b) in original], [b for (a, b) in original])
+Output: zip(*Ordered)
+bleu_metric: 0.7985269443272172
+----------
+NL: unzip list `original` and return a generator
+ref: result = ((a for (a, b) in original), (b for (a, b) in original))
+Output: ▁return (i for i in range(10)▁if i < 0 for i in range(10))
+bleu_metric: 10.587752134048651
+----------
+NL: unzip list `[('a', 1), ('b', 2), ('c', 3), ('d', 4), ('e', )]`
+ref: zip(*[('a', 1), ('b', 2), ('c', 3), ('d', 4), ('e',)])
+Output: zip([('a', 1), ('b', 2), ('c', 3), ('d', 4),
+bleu_metric: 73.36039503547302
+----------
+NL: unzip list `[('a', 1), ('b', 2), ('c', 3), ('d', 4), ('e', )]` and fill empty results with none
+ref: map(None, *[('a', 1), ('b', 2), ('c', 3), ('d', 4), ('e',)])
+Output: zip(*[('a', 1), ('b', 2), ('c', 3), ('d', 4
+bleu_metric: 64.62811194838824
+----------
+NL: encode `decimal('3.9')` to a json string
+ref: json.dumps(Decimal('3.9'))
+Output: ▁print(json.dumps(decimal('Temperature (\\d+))))
+bleu_metric: 22.656720908801994
+----------
+NL: add key "mynewkey" to dictionary `d` with value "mynewvalue"
+ref: d['mynewkey'] = 'mynewvalue'
+Output: d = {'mynewkey':'mynewvalue','mynewkey':'mynewvalue'}
+bleu_metric: 17.976451428761386
+----------
+NL: add key 'a' to dictionary `data` with value 1
+ref: data.update({'a': 1, })
+Output: data.update({'a': 1, 'b': 1})
+bleu_metric: 63.83240325919925
+----------
+NL: add key 'a' to dictionary `data` with value 1
+ref: data.update(dict(a=1))
+Output: data.update({'a': 1, 'b': 1})
+bleu_metric: 21.951524426618455
+----------
+NL: add key 'a' to dictionary `data` with value 1
+ref: data.update(a=1)
+Output: data.update({'a': 1, 'b': 1})
+bleu_metric: 21.951524426618455
+----------
+NL: find maximal value in matrix `matrix`
+ref: max([max(i) for i in matrix])
+Output: np.max(matrix,▁axis=0)
+bleu_metric: 15.233734832972273
+----------
+NL: round number `answer` to 2 precision after the decimal point
+ref: answer = str(round(answer, 2))
+Output: ▁round(answer, 2)
+bleu_metric: 49.196255036686594
+----------
+NL: extract ip address from an html string
+ref: ip = re.findall('[0-9]+(?:\\.[0-9]+){3}', s)
+Output: re.findall('[a-zA-Z0-9-]+', text)
+bleu_metric: 20.302505814938357
+----------
+NL: filter dataframe `df` by values in column `a` that appear more than once
+ref: df.groupby('A').filter(lambda x: len(x) > 1)
+Output: df.filter(['a', 'b'])
+bleu_metric: 13.360392338069724
+----------
+NL: append each line in file `myfile` into a list
+ref: [x for x in myfile.splitlines() if x != '']
+Output: with open('myfile', 'w').append(line)
+bleu_metric: 9.082300379267782
+----------
+NL: get a list of integers `lst` from a file `filename.txt`
+ref: lst = map(int, open('filename.txt').readlines())
+Output: [int(x) for x in open('filename.txt', 'lst')]
+bleu_metric: 34.46913316729032
+----------
+NL: add color bar with image `mappable` to plot `plt`
+ref: plt.colorbar(mappable=mappable, cax=ax3)
+Output: plt.show()
+bleu_metric: 11.914378950211951
+----------
+NL: count most frequent 100 words in column 'text' of dataframe `df`
+ref: Counter(' '.join(df['text']).split()).most_common(100)
+Output: df['text'].str.contains(100)
+bleu_metric: 27.101674478567933
+----------
+NL: python split a string using regex
+ref: re.findall('(.+?):(.+?)\\b ?', text)
+Output: re.split('\\s+', s)
+bleu_metric: 9.868832370006283
+----------
+NL: generate all 2-element subsets of tuple `(1, 2, 3)`
+ref: list(itertools.combinations((1, 2, 3), 2))
+Output: ▁list(itertools.product(list(range(1, 2, 3)),▁repeat=4))
+bleu_metric: 39.47641631338841
+----------
+NL: get a value of datetime.today() in the utc time zone
+ref: datetime.now(pytz.utc)
+Output: ▁datetime.utcnow().strftime('%Y-%m-%d %H:%M')
+bleu_metric: 8.085298080223222
+----------
+NL: get a new list `list2`by removing empty list from a list of lists `list1`
+ref: list2 = [x for x in list1 if x != []]
+Output: ▁list2 = [x for x in▁list1▁if x is not▁None]
+bleu_metric: 39.5373839333436
+----------
+NL: create `list2` to contain the lists from list `list1` excluding the empty lists from `list1`
+ref: list2 = [x for x in list1 if x]
+Output: ▁list2 = [[] for i in▁list1, 0]
+bleu_metric: 18.842393723950337
+----------
+NL: django response with json `data`
+ref: return HttpResponse(data, mimetype='application/json')
+Output: ▁return HttpResponse(json.dumps(data))
+bleu_metric: 29.361568020583483
+----------
+NL: get all text that is not enclosed within square brackets in string `example_str`
+ref: re.findall('(.*?)\\[.*?\\]', example_str)
+Output: ▁example_str.split('\\[(\\w+)\\]', '\\[\\]]')
+bleu_metric: 23.62618715634458
+----------
+NL: use a regex to get all text in a string `example_str` that is not surrounded by square brackets
+ref: re.findall('(.*?)(?:\\[.*?\\]|$)', example_str)
+Output: ▁print(re.findall('\\[(\\w+)\\]',▁example_str))
+bleu_metric: 27.808268092146015
+----------
+NL: get whatever is between parentheses as a single match, and any char outside as an individual match in string '(zyx)bc'
+ref: re.findall('\\(.+?\\)|\\w', '(zyx)bc')
+Output: re.search('(zyx)bc', '(zyx)bc')
+bleu_metric: 37.12324840533176
+----------
+NL: match regex '\\((.*?)\\)|(\\w)' with string '(zyx)bc'
+ref: re.findall('\\((.*?)\\)|(\\w)', '(zyx)bc')
+Output: re.findall('\\((.*?)\\)(\\w)', '(zyx)
+bleu_metric: 79.99881739549791
+----------
+NL: match multiple regex patterns with the alternation operator `|` in a string `(zyx)bc`
+ref: re.findall('\\(.*?\\)|\\w', '(zyx)bc')
+Output: re.findall('(?<=[a-z])(?=[A-Z])', '(zy
+bleu_metric: 25.2840740779747
+----------
+NL: formate each string cin list `elements` into pattern '%{0}%'
+ref: elements = ['%{0}%'.format(element) for element in elements]
+Output: re.sub('[\\W_]+', '', '\\1', '\\1', '\
+bleu_metric: 5.517784889816492
+----------
+NL: open a background process 'background-process' with arguments 'arguments'
+ref: subprocess.Popen(['background-process', 'arguments'])
+Output: subprocess.call(['background-process', 'arguments'])
+bleu_metric: 85.78928092681434
+----------
+NL: get list of values from dictionary 'mydict' w.r.t. list of keys 'mykeys'
+ref: [mydict[x] for x in mykeys]
+Output: [d['mykey'] for d in mydict]
+bleu_metric: 18.69300079996002
+----------
+NL: convert list `[('name', 'joe'), ('age', 22)]` into a dictionary
+ref: dict([('Name', 'Joe'), ('Age', 22)])
+Output: ▁dict([('name', 'joe'), ('age', 22)])
+bleu_metric: 58.41177597069678
+----------
+NL: average each two columns of array `data`
+ref: data.reshape(-1, j).mean(axis=1).reshape(data.shape[0], -1)
+Output: np.mean(np.mean(a.shape[0],▁axis=1))
+bleu_metric: 31.214272516254898
+----------
+NL: double backslash escape all double quotes in string `s`
+ref: print(s.encode('unicode-escape').replace('"', '\\"'))
+Output: ▁print(s.replace('"', ''))
+bleu_metric: 42.801426144397794
+----------
+NL: split a string into a list of words and whitespace
+ref: re.split('(\\W+)', s)
+Output: [s.split() for s in re.findall('\\S+[^\\S+]+',
+bleu_metric: 12.913533075470381
+----------
+NL: plotting stacked barplots on a panda data frame
+ref: df.plot(kind='barh', stacked=True)
+Output: df.plot(kind='bar',▁stacked=True)
+bleu_metric: 66.75075987129311
+----------
+NL: reverse the keys and values in a dictionary `mydictionary`
+ref: {i[1]: i[0] for i in list(myDictionary.items())}
+Output: ▁dict((v, k) for k, v in mydict.items())
+bleu_metric: 20.467124548606286
+----------
+NL: finding the index of elements containing substring 'how' and 'what' in a list of strings 'mylist'.
+ref: [i for i, j in enumerate(myList) if 'how' in j.lower() or 'what' in j.lower()]
+Output: [item for item in mylist▁if 'how' in item]
+bleu_metric: 11.11586813181086
+----------
+NL: check if object `obj` is a string
+ref: isinstance(obj, str)
+Output: isinstance(obj, str)
+bleu_metric: 100.0
+----------
+NL: check if object `o` is a string
+ref: isinstance(o, str)
+Output: isinstance(o, str)
+bleu_metric: 100.0
+----------
+NL: check if object `o` is a string
+ref: (type(o) is str)
+Output: isinstance(o, str)
+bleu_metric: 26.78284959130086
+----------
+NL: check if object `o` is a string
+ref: isinstance(o, str)
+Output: isinstance(o, str)
+bleu_metric: 100.0
+----------
+NL: check if `obj_to_test` is a string
+ref: isinstance(obj_to_test, str)
+Output: isinstance(obj_to_test, str)
+bleu_metric: 100.0
+----------
+NL: append list `list1` to `list2`
+ref: list2.extend(list1)
+Output: ▁list2 = [[] for i in range(1)]
+bleu_metric: 9.672649511413093
+----------
+NL: append list `mylog` to `list1`
+ref: list1.extend(mylog)
+Output: ▁list1.append(mylog)
+bleu_metric: 45.18010018049224
+----------
+NL: append list `a` to `c`
+ref: c.extend(a)
+Output: a.append(c)
+bleu_metric: 30.21375397356768
+----------
+NL: append items in list `mylog` to `list1`
+ref: for line in mylog:
+ list1.append(line)
+Output: ▁list1.append({'mylog': [1, 2, 3, 4], 'list1': [1, 2, 3, 4]})
+bleu_metric: 11.110969989860408
+----------
+NL: append a tuple of elements from list `a` with indexes '[0][0] [0][2]' to list `b`
+ref: b.append((a[0][0], a[0][2]))
+Output: [(i, j) for i, j in zip(a, b)]
+bleu_metric: 9.575085238714179
+----------
+NL: initialize `secret_key` in flask config with `your_secret_string `
+ref: app.config['SECRET_KEY'] = 'Your_secret_string'
+Output: {'SECURITY_KEY': 'your_SECURITY_string'}
+bleu_metric: 12.498879161997976
+----------
+NL: unpack a series of tuples in pandas into a dataframe with column names 'out-1' and 'out-2'
+ref: pd.DataFrame(out.tolist(), columns=['out-1', 'out-2'], index=out.index)
+Output: pd.DataFrame(df.columns,▁columns=['out-1', 'out-2'])
+bleu_metric: 55.07111790934335
+----------
+NL: find the index of an element 'msft' in a list `stocks_list`
+ref: [x for x in range(len(stocks_list)) if stocks_list[x] == 'MSFT']
+Output: [i for i, v in enumerate(stocks_list.items())▁if v =='msft')▁if v
+bleu_metric: 13.33045094117443
+----------
+NL: rotate the xtick labels of matplotlib plot `ax` by `45` degrees to make long labels readable
+ref: ax.set_xticklabels(labels, rotation=45)
+Output: ax.set_xticklabels(ax.xaxis.get_majorticklabels(45))
+bleu_metric: 30.620490882364876
+----------
+NL: remove symbols from a string `s`
+ref: re.sub('[^\\w]', ' ', s)
+Output: re.sub('(?<=[a-z])(?=[A-Z])', '', s)
+bleu_metric: 35.3639760061202
+----------
+NL: get the current directory of a script
+ref: os.path.basename(os.path.dirname(os.path.realpath(__file__)))
+Output: os.path.dirname(os.path.realpath(__file__))
+bleu_metric: 63.37967287681905
+----------
+NL: find octal characters matches from a string `str` using regex
+ref: print(re.findall("'\\\\[0-7]{1,3}'", str))
+Output: re.findall(pattern,▁string).findall(pattern, re.UNICODE)
+bleu_metric: 13.635660702165275
+----------
+NL: split string `input` based on occurrences of regex pattern '[ ](?=[a-z]+\\b)'
+ref: re.split('[ ](?=[A-Z]+\\b)', input)
+Output: input.split('[ ](?=[a-z]+\\b)')
+bleu_metric: 64.58552885189877
+----------
+NL: split string `input` at every space followed by an upper-case letter
+ref: re.split('[ ](?=[A-Z])', input)
+Output: input.split('.')[1]
+bleu_metric: 16.636072310752336
+----------
+NL: send multipart encoded file `files` to url `url` with headers `headers` and metadata `data`
+ref: r = requests.post(url, files=files, headers=headers, data=data)
+Output: ▁urllib.request.urlopen(url,▁headers=headers,▁headers=data)
+bleu_metric: 33.09468095382839
+----------
+NL: write bytes `bytes_` to a file `filename` in python 3
+ref: open('filename', 'wb').write(bytes_)
+Output: open('filename', 'rb').write(bytes(bytes_))
+bleu_metric: 60.59827460443867
+----------
+NL: get a list from a list `lst` with values mapped into a dictionary `dct`
+ref: [dct[k] for k in lst]
+Output: ▁list(itertools.chain.from_iterable(lst))
+bleu_metric: 10.511846841633776
+----------
+NL: find duplicate names in column 'name' of the dataframe `x`
+ref: x.set_index('name').index.get_duplicates()
+Output: df.groupby('name')['x'].cumsum()
+bleu_metric: 34.11488281065382
+----------
+NL: truncate float 1.923328437452 to 3 decimal places
+ref: round(1.923328437452, 3)
+Output: ▁float('{0:.2f}'.format(env))
+bleu_metric: 9.629943614188138
+----------
+NL: sort list `li` in descending order based on the date value in second element of each list in list `li`
+ref: sorted(li, key=lambda x: datetime.strptime(x[1], '%d/%m/%Y'), reverse=True)
+Output: ▁sorted(li,▁key=lambda x: (x[1], x[0]))
+bleu_metric: 31.993666822129963
+----------
+NL: place the radial ticks in plot `ax` at 135 degrees
+ref: ax.set_rlabel_position(135)
+Output: ax.set_visible(False)
+bleu_metric: 32.17294420803808
+----------
+NL: check if path `my_path` is an absolute path
+ref: os.path.isabs(my_path)
+Output: ▁print(os.path.abspath(__file__))
+bleu_metric: 29.256127307315065
+----------
+NL: get number of keys in dictionary `yourdict`
+ref: len(list(yourdict.keys()))
+Output: len(yourdict.keys())
+bleu_metric: 63.70964381207871
+----------
+NL: count the number of keys in dictionary `yourdictfile`
+ref: len(set(open(yourdictfile).read().split()))
+Output: len(yourdictfile.values())
+bleu_metric: 13.938681520780069
+----------
+NL: pandas dataframe get first row of each group by 'id'
+ref: df.groupby('id').first()
+Output: df['id'] = df['id'].apply(lambda x: x.tolist())
+bleu_metric: 12.991916506579942
+----------
+NL: split a list in first column into multiple columns keeping other columns as well in pandas data frame
+ref: pd.concat([df[0].apply(pd.Series), df[1]], axis=1)
+Output: df['A'].str.split(',', 1)['B'].str.split(',')
+bleu_metric: 9.64535051675695
+----------
+NL: extract attributes 'src="js/([^"]*\\bjquery\\b[^"]*)"' from string `data`
+ref: re.findall('src="js/([^"]*\\bjquery\\b[^"]*)"', data)
+Output: re.findall('\\bjquery\\b[^"]*)', data)
+bleu_metric: 51.296752626995115
+----------
+NL: sum integers contained in strings in list `['', '3.4', '', '', '1.0']`
+ref: sum(int(float(item)) for item in [_f for _f in ['', '3.4', '', '', '1.0'] if _f])
+Output: ▁sum(map(int, ['', '3.4', '', '1.0']))
+bleu_metric: 45.29937105346175
+----------
+NL: call a subprocess with arguments `c:\\program files\\vmware\\vmware server\\vmware-cmd.bat` that may contain spaces
+ref: subprocess.Popen(['c:\\Program Files\\VMware\\VMware Server\\vmware-cmd.bat'])
+Output: subprocess.call(['c:\\programfile\\vmware\\vmware-server\\
+bleu_metric: 33.78386474477107
+----------
+NL: reverse a priority queue `q` in python without using classes
+ref: q.put((-n, n))
+Output: time.sleep(1)
+bleu_metric: 13.597602315271134
+----------
+NL: make a barplot of data in column `group` of dataframe `df` colour-coded according to list `color`
+ref: df['group'].plot(kind='bar', color=['r', 'g', 'b', 'r', 'g', 'b', 'r'])
+Output: plt.show()
+bleu_metric: 0.030120959708599737
+----------
+NL: find all matches of regex pattern '([a-fa-f\\d]{32})' in string `data`
+ref: re.findall('([a-fA-F\\d]{32})', data)
+Output: re.findall('([a-fa-f\\d]{32})', data)
+bleu_metric: 80.59468738950976
+----------
+NL: get the length of list `my_list`
+ref: len(my_list)
+Output: len(my_list)
+bleu_metric: 100.0
+----------
+NL: getting the length of array `l`
+ref: len(l)
+Output: len(l)
+bleu_metric: 100.0
+----------
+NL: getting the length of array `s`
+ref: len(s)
+Output: len(s)
+bleu_metric: 100.0
+----------
+NL: getting the length of `my_tuple`
+ref: len(my_tuple)
+Output: len(my_tuple)
+bleu_metric: 100.0
+----------
+NL: getting the length of `my_string`
+ref: len(my_string)
+Output: len(my_string)
+bleu_metric: 100.0
+----------
+NL: remove escape character from string "\\a"
+ref: """\\a""".decode('string_escape')
+Output: """\\a""".decode('string_escape')
+bleu_metric: 100.0
+----------
+NL: replace each 'a' with 'b' and each 'b' with 'a' in the string 'obama' in a single pass.
+ref: """obama""".replace('a', '%temp%').replace('b', 'a').replace('%temp%', 'b')
+Output: """""".join(['a', 'b', 'a'])
+bleu_metric: 28.029141246256657
+----------
+NL: remove directory tree '/folder_name'
+ref: shutil.rmtree('/folder_name')
+Output: os.path.dirname('/folder_name')
+bleu_metric: 57.608442016038964
+----------
+NL: create a new column `weekday` in pandas data frame `data` based on the values in column `my_dt`
+ref: data['weekday'] = data['my_dt'].apply(lambda x: x.weekday())
+Output: data.apply(lambda x: x['weekday'].str.contains(x['my_dt'
+bleu_metric: 51.68669917653163
+----------
+NL: reverse sort counter `x` by values
+ref: sorted(x, key=x.get, reverse=True)
+Output: ▁sorted(x, reverse=True)
+bleu_metric: 49.88250584787245
+----------
+NL: reverse sort counter `x` by value
+ref: sorted(list(x.items()), key=lambda pair: pair[1], reverse=True)
+Output: ▁sorted(x,▁key=itemgetter(1), reverse=True)
+bleu_metric: 28.177860103739523
+----------
+NL: append a numpy array 'b' to a numpy array 'a'
+ref: np.vstack((a, b))
+Output: ▁numpy.append(a, b)
+bleu_metric: 46.92470064105599
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: print(concatenate((a, b), axis=0))
+Output: np.concatenate((a, b),▁axis=0)
+bleu_metric: 65.60930058975354
+----------
+NL: numpy concatenate two arrays `a` and `b` along the second axis
+ref: print(concatenate((a, b), axis=1))
+Output: np.concatenate((a, b),▁axis=0)
+bleu_metric: 55.60336961016132
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: c = np.r_[(a[None, :], b[None, :])]
+Output: np.concatenate((a, b),▁axis=0)
+bleu_metric: 10.756284573187752
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: np.array((a, b))
+Output: np.concatenate((a, b),▁axis=0)
+bleu_metric: 41.896825433302276
+----------
+NL: fetch address information for host 'google.com' ion port 80
+ref: print(socket.getaddrinfo('google.com', 80))
+Output: db.request.urlretrieve('google.com')
+bleu_metric: 39.00756812161481
+----------
+NL: add a column 'day' with value 'sat' to dataframe `df`
+ref: df.xs('sat', level='day', drop_level=False)
+Output: df['Date'] = df['Date'].apply(lambda x: x.to_date(
+bleu_metric: 7.575731225158963
+----------
+NL: return a 401 unauthorized in django
+ref: return HttpResponse('Unauthorized', status=401)
+Output: MyModel.objects.filter(author__id=1)
+bleu_metric: 14.294845713017917
+----------
+NL: flask set folder 'wherever' as the default template folder
+ref: Flask(__name__, template_folder='wherever')
+Output: ▁app.run(debug=True)
+bleu_metric: 17.03318603763928
+----------
+NL: how do i insert into t1 (select * from t2) in sqlalchemy?
+ref: session.execute('INSERT INTO t1 (SELECT * FROM t2)')
+Output: ▁session.commit()
+bleu_metric: 9.01760872773444
+----------
+NL: sort a list of lists 'c2' such that third row comes first
+ref: c2.sort(key=lambda row: row[2])
+Output: c2.sort(key=lambda x: x[0])
+bleu_metric: 55.33409598501608
+----------
+NL: sorting a list of lists in python
+ref: c2.sort(key=lambda row: (row[2], row[1], row[0]))
+Output: ▁sorted(list_of_lists,▁key=lambda x: x[1])
+bleu_metric: 13.837388732992803
+----------
+NL: sorting a list of lists in python
+ref: c2.sort(key=lambda row: (row[2], row[1]))
+Output: ▁sorted(list_of_lists,▁key=lambda x: x[1])
+bleu_metric: 23.34445185729352
+----------
+NL: set font `arial` to display non-ascii characters in matplotlib
+ref: matplotlib.rc('font', **{'sans-serif': 'Arial', 'family': 'sans-serif'})
+Output: plt.show()
+bleu_metric: 0.3669484097406089
+----------
+NL: convert datetime column 'date' of pandas dataframe 'df' to ordinal
+ref: df['date'].apply(lambda x: x.toordinal())
+Output: df['date'] = df['date'].dt.to_datetime(df['date']
+bleu_metric: 28.326302347195153
+----------
+NL: get html source of selenium webelement `element`
+ref: element.get_attribute('innerHTML')
+Output: browser.find_element_by_xpath("//a[@id='a']//a[
+bleu_metric: 9.092617426809149
+----------
+NL: get the integer location of a key `bob` in a pandas data frame
+ref: df.index.get_loc('bob')
+Output: df.loc[df['bob'].isin()]
+bleu_metric: 19.345299022826186
+----------
+NL: open a 'gnome' terminal from python script and run 'sudo apt-get update' command.
+ref: os.system('gnome-terminal -e \'bash -c "sudo apt-get update; exec bash"\'')
+Output: os.system('sudo apt-get update')
+bleu_metric: 21.20024487612511
+----------
+NL: add an item with key 'third_key' and value 1 to an dictionary `my_dict`
+ref: my_dict.update({'third_key': 1})
+Output: ▁dict((k, v) for k, v in my_dict.items()▁if v[0] == 1)
+bleu_metric: 7.261813302549417
+----------
+NL: declare an array
+ref: my_list = []
+Output: intarray = []
+bleu_metric: 75.98356856515926
+----------
+NL: insert item `12` to a list `my_list`
+ref: my_list.append(12)
+Output: my_list.insert(0, 12)
+bleu_metric: 32.46679154750989
+----------
+NL: add an entry 'wuggah' at the beginning of list `mylist`
+ref: myList.insert(0, 'wuggah')
+Output: ▁list(['wuggah', 'wuggah'])
+bleu_metric: 27.22589423069702
+----------
+NL: convert a hex-string representation to actual bytes
+ref: """\\xF3\\xBE\\x80\\x80""".replace('\\x', '').decode('hex')
+Output: ▁struct.unpack('I', 16))[0]
+bleu_metric: 6.258020733367848
+----------
+NL: select the last column of dataframe `df`
+ref: df[df.columns[-1]]
+Output: df.iloc[:, (-1)]
+bleu_metric: 20.78060434846712
+----------
+NL: get the first value from dataframe `df` where column 'letters' is equal to 'c'
+ref: df.loc[df['Letters'] == 'C', 'Letters'].values[0]
+Output: df['letters'].str.contains('c')
+bleu_metric: 12.409088593346615
+----------
+NL: converting two lists `[1, 2, 3]` and `[4, 5, 6]` into a matrix
+ref: np.column_stack(([1, 2, 3], [4, 5, 6]))
+Output: ▁matrix([[1, 2, 3], [4, 5, 6]))
+bleu_metric: 79.78897056262763
+----------
+NL: get the type of `i`
+ref: type(i)
+Output: type(i)
+bleu_metric: 100.0
+----------
+NL: determine the type of variable `v`
+ref: type(v)
+Output: type(v)
+bleu_metric: 100.0
+----------
+NL: determine the type of variable `v`
+ref: type(v)
+Output: type(v)
+bleu_metric: 100.0
+----------
+NL: determine the type of variable `v`
+ref: type(v)
+Output: type(v)
+bleu_metric: 100.0
+----------
+NL: determine the type of variable `v`
+ref: type(v)
+Output: type(v)
+bleu_metric: 100.0
+----------
+NL: get the type of variable `variable_name`
+ref: print(type(variable_name))
+Output: ▁globals()['factor']
+bleu_metric: 13.54599427337814
+----------
+NL: get the 5th item of a generator
+ref: next(itertools.islice(range(10), 5, 5 + 1))
+Output: ▁list(itertools.islice(it, 5))
+bleu_metric: 31.347458595184502
+----------
+NL: print a string `word` with string format
+ref: print('"{}"'.format(word))
+Output: ▁print(' '.join(word))
+bleu_metric: 49.97150634074441
+----------
+NL: join a list of strings `list` using a space ' '
+ref: """ """.join(list)
+Output: """ """.join(map(str,▁list))
+bleu_metric: 56.79161104357995
+----------
+NL: create list `y` containing two empty lists
+ref: y = [[] for n in range(2)]
+Output: y = [[] for i in range(2)]
+bleu_metric: 80.03203203844998
+----------
+NL: read a file 'c:/name/mydocuments/numbers' into a list `data`
+ref: data = [line.strip() for line in open('C:/name/MyDocuments/numbers', 'r')]
+Output: data = [line.read() for▁line in open('C:/name/mydocuments/numbers')]
+bleu_metric: 56.78312391912661
+----------
+NL: delete all occurrences of character 'i' in string 'it is icy'
+ref: """""".join([char for char in 'it is icy' if char != 'i'])
+Output: re.sub('i', 'it is icy')
+bleu_metric: 15.99248714366856
+----------
+NL: delete all instances of a character 'i' in a string 'it is icy'
+ref: re.sub('i', '', 'it is icy')
+Output: re.sub('i', '', 'it is icy')
+bleu_metric: 100.0
+----------
+NL: delete all characters "i" in string "it is icy"
+ref: """it is icy""".replace('i', '')
+Output: """it is icy""".replace('i', '')
+bleu_metric: 100.0
+----------
+NL: how to delete all instances of a character in a string in python?
+ref: """""".join([char for char in 'it is icy' if char != 'i'])
+Output: re.sub('[^a-zA-Z]*', '', text)
+bleu_metric: 7.866954533441649
+----------
+NL: drop rows of pandas dataframe `df` having nan in column at index "1"
+ref: df.dropna(subset=[1])
+Output: df.dropna(df.columns,▁axis=1)
+bleu_metric: 30.130404892785684
+----------
+NL: get elements from list `mylist`, that have a field `n` value 30
+ref: [x for x in myList if x.n == 30]
+Output: [i for i, v in enumerate(mylist)▁if v == n]
+bleu_metric: 13.229527092256335
+----------
+NL: converting list of strings `intstringlist` to list of integer `nums`
+ref: nums = [int(x) for x in intstringlist]
+Output: [int(x) for x in intstringlist]
+bleu_metric: 83.9234989944319
+----------
+NL: convert list of string numbers into list of integers
+ref: map(int, eval(input('Enter the unfriendly numbers: ')))
+Output: [int(x) for x in lst]
+bleu_metric: 7.290792403968443
+----------
+NL: print "." without newline
+ref: sys.stdout.write('.')
+Output: ▁sys.stdout.write('.')
+bleu_metric: 83.85766789076257
+----------
+NL: round off the float that is the product of `2.52 * 100` and convert it to an int
+ref: int(round(2.51 * 100))
+Output: int(round(1.0 *▁100))
+bleu_metric: 46.82568791024402
+----------
+NL: find all files in directory "/mydir" with extension ".txt"
+ref: for file in glob.glob('*.txt'):
+ pass
+Output: for▁file in os.listdir('/mydir'):▁if▁file.endswith('.txt'):▁pass
+bleu_metric: 20.1857735807738
+----------
+NL: find all files in directory "/mydir" with extension ".txt"
+ref: for file in os.listdir('/mydir'):
+ if file.endswith('.txt'):
+ pass
+Output: for▁file in os.listdir('/mydir'):▁if▁file.endswith('.txt'):▁pass
+bleu_metric: 71.95956498085113
+----------
+NL: find all files in directory "/mydir" with extension ".txt"
+ref: for (root, dirs, files) in os.walk('/mydir'):
+ for file in files:
+ if file.endswith('.txt'):
+ pass
+Output: for▁file in os.listdir('/mydir'):▁if▁file.endswith('.txt'):▁pass
+bleu_metric: 47.1684455439837
+----------
+NL: plot dataframe `df` without a legend
+ref: df.plot(legend=False)
+Output: plt.show()
+bleu_metric: 20.034704329441453
+----------
+NL: loop through the ip address range "192.168.x.x"
+ref: for i in range(256):
+ for j in range(256):
+ ip = ('192.168.%d.%d' % (i, j))
+ print(ip)
+Output: for i in range('192.168.x.x'):▁pass
+bleu_metric: 14.238794877552813
+----------
+NL: loop through the ip address range "192.168.x.x"
+ref: for (i, j) in product(list(range(256)), list(range(256))):
+ pass
+Output: for i in range('192.168.x.x'):▁pass
+bleu_metric: 7.916713848850043
+----------
+NL: loop through the ip address range "192.168.x.x"
+ref: generator = iter_iprange('192.168.1.1', '192.168.255.255', step=1)
+Output: for i in range('192.168.x.x'):▁pass
+bleu_metric: 18.291291695991816
+----------
+NL: sum the corresponding decimal values for binary values of each boolean element in list `x`
+ref: sum(1 << i for i, b in enumerate(x) if b)
+Output: ▁sum([int(x) for x in▁list(bool(e))])
+bleu_metric: 14.071964689317513
+----------
+NL: write multiple strings `line1`, `line2` and `line3` in one line in a file `target`
+ref: target.write('%r\n%r\n%r\n' % (line1, line2, line3))
+Output: f.write(line1,▁line2 +▁line3.split(','))
+bleu_metric: 16.116608915070366
+----------
+NL: convert list of lists `data` into a flat list
+ref: [y for x in data for y in (x if isinstance(x, list) else [x])]
+Output: [j for i in data for j in i]
+bleu_metric: 9.344244386017191
+----------
+NL: print new line character as `\n` in a string `foo\nbar`
+ref: print('foo\nbar'.encode('string_escape'))
+Output: """foo\nbar""".replace('\n')
+bleu_metric: 32.49189152106538
+----------
+NL: remove last comma character ',' in string `s`
+ref: """""".join(s.rsplit(',', 1))
+Output: s = s.replace(',', '')
+bleu_metric: 24.4545047029932
+----------
+NL: calculate the mean of each element in array `x` with the element previous to it
+ref: (x[1:] + x[:-1]) / 2
+Output: np.mean(x)
+bleu_metric: 6.656592803413297
+----------
+NL: get an array of the mean of each two consecutive values in numpy array `x`
+ref: x[:-1] + (x[1:] - x[:-1]) / 2
+Output: ▁numpy.array([[1, 2], [2, 4]])
+bleu_metric: 10.64536882179483
+----------
+NL: load data containing `utf-8` from file `new.txt` into numpy array `arr`
+ref: arr = numpy.fromiter(codecs.open('new.txt', encoding='utf-8'), dtype=' 1]
+Output: set(['a', 'b', 'c'])
+bleu_metric: 6.171879938074473
+----------
+NL: Insert directory 'apps' into directory `__file__`
+ref: sys.path.insert(1, os.path.join(os.path.dirname(__file__), 'apps'))
+Output: ▁sys.path.apps(os.path.dirname(__file__))
+bleu_metric: 37.386427821086265
+----------
+NL: modify sys.path for python module `subdir`
+ref: sys.path.append(os.path.join(os.path.dirname(__file__), 'subdir'))
+Output: ▁sys.path.append(subdir)
+bleu_metric: 13.681283243087492
+----------
+NL: Insert a 'None' value into a SQLite3 table.
+ref: db.execute("INSERT INTO present VALUES('test2', ?, 10)", (None,))
+Output: db.commit()
+bleu_metric: 1.3648974794270936
+----------
+NL: flatten list `list_of_menuitems`
+ref: [image for menuitem in list_of_menuitems for image in menuitem]
+Output: ▁list_of_enumerate(chain.from_iterable(list_of_menuitems))
+bleu_metric: 11.368272367804307
+----------
+NL: append elements of a set `b` to a list `a`
+ref: a.extend(b)
+Output: a.append(b)
+bleu_metric: 59.46035575013605
+----------
+NL: Append elements of a set to a list in Python
+ref: a.extend(list(b))
+Output: a.extend([1, 2, 3])
+bleu_metric: 30.576902884505113
+----------
+NL: write the data of dataframe `df` into text file `np.txt`
+ref: np.savetxt('c:\\data\\np.txt', df.values, fmt='%d')
+Output: df.to_numeric(np.txt)
+bleu_metric: 3.9614496879183108
+----------
+NL: write content of DataFrame `df` into text file 'c:\\data\\pandas.txt'
+ref: df.to_csv('c:\\data\\pandas.txt', header=None, index=None, sep=' ', mode='a')
+Output: df.to_csv('c:\\data\\pandas.txt')
+bleu_metric: 33.15109099076485
+----------
+NL: Split a string `x` by last occurrence of character `-`
+ref: print(x.rpartition('-')[0])
+Output: x.rsplit('-', 1)
+bleu_metric: 28.24924701979438
+----------
+NL: get the last part of a string before the character '-'
+ref: print(x.rsplit('-', 1)[0])
+Output: a[-1:] + a[:-1]
+bleu_metric: 9.389419352536649
+----------
+NL: upload file using FTP
+ref: ftp.storlines('STOR ' + filename, open(filename, 'r'))
+Output: upload_file = open('file.txt', 'wb')
+bleu_metric: 14.12701802079611
+----------
+NL: add one to the hidden web element with id 'XYZ' with selenium python script
+ref: browser.execute_script("document.getElementById('XYZ').value+='1'")
+Output: ▁driver.get('XYZ')
+bleu_metric: 13.581073322306715
+----------
+NL: create array containing the maximum value of respective elements of array `[2, 3, 4]` and array `[1, 5, 2]`
+ref: np.maximum([2, 3, 4], [1, 5, 2])
+Output: np.argmax([2, 3, 4], [1, 5, 2])
+bleu_metric: 88.67047947918724
+----------
+NL: print a list `l` and move first 3 elements to the end of the list
+ref: print(l[3:] + l[:3])
+Output: l = [i for i, l in enumerate(l, 3)▁if l == 3]
+bleu_metric: 10.414419091986518
+----------
+NL: loop over files in directory '.'
+ref: for fn in os.listdir('.'):
+ if os.path.isfile(fn):
+ pass
+Output: for item in os.listdir('.') in os.listdir('.'):▁pass
+bleu_metric: 48.62438913464417
+----------
+NL: loop over files in directory `source`
+ref: for (root, dirs, filenames) in os.walk(source):
+ for f in filenames:
+ pass
+Output: for (not▁file in os.walk(source)):▁pass
+bleu_metric: 34.21915639793568
+----------
+NL: create a random list of integers
+ref: [int(1000 * random.random()) for i in range(10000)]
+Output: ▁randomList = [random.random() for _ in range(10)]
+bleu_metric: 35.22146109090851
+----------
+NL: Using %f with strftime() in Python to get microseconds
+ref: datetime.datetime.now().strftime('%H:%M:%S.%f')
+Output: ▁print(os.path.getmtime('/tmp/test.txt'))
+bleu_metric: 10.400016243706089
+----------
+NL: Google App Engine execute GQL query 'SELECT * FROM Schedule WHERE station = $1' with parameter `foo.key()`
+ref: db.GqlQuery('SELECT * FROM Schedule WHERE station = $1', foo.key())
+Output: ▁cursor.execute('SELECT *▁FROM▁your Station = $1,')
+bleu_metric: 22.409416775129817
+----------
+NL: filter rows in pandas starting with alphabet 'f' using regular expression.
+ref: df.b.str.contains('^f')
+Output: df[(df['a'] > 1) & (df['b'] > 1)]
+bleu_metric: 6.871624004919695
+----------
+NL: print a 2 dimensional list `tab` as a table with delimiters
+ref: print('\n'.join('\t'.join(str(col) for col in row) for row in tab))
+Output: a = [(a[i], b[i]) for i in range(0, len(a), 2)]
+bleu_metric: 6.711116671457737
+----------
+NL: pandas: delete rows in dataframe `df` based on multiple columns values
+ref: df.set_index(list('BC')).drop(tuples, errors='ignore').reset_index()
+Output: df.loc[df.columns == [(df.columns▁!= 0).any(axis=1)]
+bleu_metric: 8.722785633659552
+----------
+NL: format the variables `self.goals` and `self.penalties` using string formatting
+ref: """({:d} goals, ${:d})""".format(self.goals, self.penalties)
+Output: ▁print('{} {}'.format(self.goals,▁self.penalties))
+bleu_metric: 34.66537079680952
+----------
+NL: format string "({} goals, ${})" with variables `goals` and `penalties`
+ref: """({} goals, ${})""".format(self.goals, self.penalties)
+Output: """({}, ${})""".format(goals, cents)
+bleu_metric: 61.17470797490066
+----------
+NL: format string "({0.goals} goals, ${0.penalties})"
+ref: """({0.goals} goals, ${0.penalties})""".format(self)
+Output: """({0.goals}, ${0.penalties})""".format('${0.go
+bleu_metric: 74.24213297217366
+----------
+NL: convert list of lists `L` to list of integers
+ref: [int(''.join(str(d) for d in x)) for x in L]
+Output: [int(x) for x in L]
+bleu_metric: 26.91716452870884
+----------
+NL: combine elements of each list in list `L` into digits of a single integer
+ref: [''.join(str(d) for d in x) for x in L]
+Output: [int(x) for x in L]
+bleu_metric: 31.16211115352542
+----------
+NL: convert a list of lists `L` to list of integers
+ref: L = [int(''.join([str(y) for y in x])) for x in L]
+Output: [int(x) for x in L]
+bleu_metric: 18.058610039563465
+----------
+NL: write the elements of list `lines` concatenated by special character '\n' to file `myfile`
+ref: myfile.write('\n'.join(lines))
+Output: myfile.write('\n')
+bleu_metric: 51.6288658561904
+----------
+NL: removing an element from a list based on a predicate 'X' or 'N'
+ref: [x for x in ['AAT', 'XAC', 'ANT', 'TTA'] if 'X' not in x and 'N' not in x]
+Output: [x for x in lst▁if x▁!= 'N')
+bleu_metric: 12.603335496309672
+----------
+NL: Remove duplicate words from a string `text` using regex
+ref: text = re.sub('\\b(\\w+)( \\1\\b)+', '\\1', text)
+Output: re.sub('\\b\\b', '', text)
+bleu_metric: 29.302169335308744
+----------
+NL: count non zero values in each column in pandas data frame
+ref: df.astype(bool).sum(axis=1)
+Output: df.groupby(level=0).agg(lambda x: x.sum())
+bleu_metric: 16.559113761114784
+----------
+NL: search for string that matches regular expression pattern '(?.*<' in xml string `line`
+ref: print(re.search('>.*<', line).group(0))
+Output: ▁print(re.search('>.*<',▁line))
+bleu_metric: 66.21156208217488
+----------
+NL: erase all the contents of a file `filename`
+ref: open(filename, 'w').close()
+Output: open('filename', 'w').close()
+bleu_metric: 70.98108717509044
+----------
+NL: convert a string into datetime using the format '%Y-%m-%d %H:%M:%S.%f'
+ref: datetime.datetime.strptime(string_date, '%Y-%m-%d %H:%M:%S.%f')
+Output: ▁datetime.datetime.strptime('%Y-%m-%d %H:%M:%S.%f')
+bleu_metric: 88.27916928185874
+----------
+NL: find the index of a list with the first element equal to '332' within the list of lists `thelist`
+ref: [index for index, item in enumerate(thelist) if item[0] == '332']
+Output: [i for i, e in enumerate(thelist)▁if e[0] == '332']
+bleu_metric: 56.9717869265984
+----------
+NL: lower a string `text` and remove non-alphanumeric characters aside from space
+ref: re.sub('[^\\sa-zA-Z0-9]', '', text).lower().strip()
+Output: re.sub('[^0-9a-zA-Z]+', '', text)
+bleu_metric: 42.32964405407573
+----------
+NL: remove all non-alphanumeric characters except space from a string `text` and lower it
+ref: re.sub('(?!\\s)[\\W_]', '', text).lower().strip()
+Output: re.sub('[^0-9a-zA-Z]+', '', text)
+bleu_metric: 31.11523617245988
+----------
+NL: subscript text 'H20' with '2' as subscripted in matplotlib labels for arrays 'x' and 'y'.
+ref: plt.plot(x, y, label='H\u2082O')
+Output: plt.show()
+bleu_metric: 6.117046115680378
+----------
+NL: subscript text 'H20' with '2' as subscripted in matplotlib labels for arrays 'x' and 'y'.
+ref: plt.plot(x, y, label='$H_2O$')
+Output: plt.show()
+bleu_metric: 6.117046115680378
+----------
+NL: loop over a list `mylist` if sublists length equals 3
+ref: [x for x in mylist if len(x) == 3]
+Output: [len(mylist) for _ in range(3)]
+bleu_metric: 15.934326838673723
+----------
+NL: initialize a list `lst` of 100 objects Object()
+ref: lst = [Object() for _ in range(100)]
+Output: [object(lst) for l in range(100)]
+bleu_metric: 51.697315395717055
+----------
+NL: create list `lst` containing 100 instances of object `Object`
+ref: lst = [Object() for i in range(100)]
+Output: [x for x in lst for x in range(100)]
+bleu_metric: 44.844165354587766
+----------
+NL: get the content of child tag with`href` attribute whose parent has css `someclass`
+ref: self.driver.find_element_by_css_selector('.someclass a').get_attribute('href')
+Output: href.get('someclass', '')
+bleu_metric: 13.739207814095348
+----------
+NL: joining data from dataframe `df1` with data from dataframe `df2` based on matching values of column 'Date_Time' in both dataframes
+ref: df1.merge(df2, on='Date_Time')
+Output: pd.merge(df1, df2,▁how='Date_Time').merge(df2, on='D
+bleu_metric: 44.05863767358845
+----------
+NL: use `%s` operator to print variable values `str1` inside a string
+ref: 'first string is: %s, second one is: %s' % (str1, 'geo.tif')
+Output: ▁print(str1, '%s' % ', '.join(str1, str1))
+bleu_metric: 24.754295374108246
+----------
+NL: Split a string by a delimiter in python
+ref: [x.strip() for x in '2.MATCHES $$TEXT$$ STRING'.split('$$TEXT$$')]
+Output: re.split('\\s+', s)
+bleu_metric: 6.630841478192585
+----------
+NL: check if directory `directory ` exists and create it if necessary
+ref: if (not os.path.exists(directory)):
+ os.makedirs(directory)
+Output: ▁print(os.path.exists(directory))
+bleu_metric: 43.84096837774454
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: os.makedirs(path)
+Output: os.path.exists(path)
+bleu_metric: 45.18010018049224
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: distutils.dir_util.mkpath(path)
+Output: os.path.exists(path)
+bleu_metric: 42.728700639623405
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: os.makedirs(path)
+Output: os.path.exists(path)
+bleu_metric: 45.18010018049224
+----------
+NL: Replace a separate word 'H3' by 'H1' in a string 'text'
+ref: re.sub('\\bH3\\b', 'H1', text)
+Output: """""".join(['H3', 'H1', 'text'])
+bleu_metric: 27.186350343875553
+----------
+NL: substitute ASCII letters in string 'aas30dsa20' with empty string ''
+ref: re.sub('\\D', '', 'aas30dsa20')
+Output: """Aas30dSa20""".replace(' ', '')
+bleu_metric: 32.263864160302525
+----------
+NL: get digits only from a string `aas30dsa20` using lambda function
+ref: """""".join([x for x in 'aas30dsa20' if x.isdigit()])
+Output: """""".join([c for c in aas30dsa20▁if c.isdigit()])
+bleu_metric: 62.76638518128994
+----------
+NL: access a tag called "name" in beautifulsoup `soup`
+ref: print(soup.find('name').string)
+Output: ▁print(soup.findAll('name'))
+bleu_metric: 56.810968323374965
+----------
+NL: get a dictionary `records` of key-value pairs in PyMongo cursor `cursor`
+ref: records = dict((record['_id'], record) for record in cursor)
+Output: ▁dict([(k, v) for k, v in▁list(cursor.items())])
+bleu_metric: 10.142430673893113
+----------
+NL: Create new matrix object by concatenating data from matrix A and matrix B
+ref: np.concatenate((A, B))
+Output: ▁numpy.concatenate([(A, B)])
+bleu_metric: 41.69392927528885
+----------
+NL: concat two matrices `A` and `B` in numpy
+ref: np.vstack((A, B))
+Output: np.concatenate(A, B)
+bleu_metric: 49.832529198346116
+----------
+NL: Get the characters count in a file `filepath`
+ref: os.stat(filepath).st_size
+Output: ▁filepath.count('/path/to/filepath')
+bleu_metric: 9.672649511413093
+----------
+NL: count the occurrences of item "a" in list `l`
+ref: l.count('a')
+Output: ▁sum(1 for i in l▁if i.startswith('a'))
+bleu_metric: 24.64832197476761
+----------
+NL: count the occurrences of items in list `l`
+ref: Counter(l)
+Output: ▁sum(1 for i in l▁if i.startswith('\n') > 1)
+bleu_metric: 7.575731225158963
+----------
+NL: count the occurrences of items in list `l`
+ref: [[x, l.count(x)] for x in set(l)]
+Output: ▁sum(1 for i in l▁if i.startswith('\n') > 1)
+bleu_metric: 7.758410052509766
+----------
+NL: count the occurrences of items in list `l`
+ref: dict(((x, l.count(x)) for x in set(l)))
+Output: ▁sum(1 for i in l▁if i.startswith('\n') > 1)
+bleu_metric: 9.009113474307316
+----------
+NL: count the occurrences of item "b" in list `l`
+ref: l.count('b')
+Output: ▁sum(1 for i, j in zip(l, 'b')▁if j == j)
+bleu_metric: 15.450079547146164
+----------
+NL: copy file `srcfile` to directory `dstdir`
+ref: shutil.copy(srcfile, dstdir)
+Output: ▁shutil.copy(srcfile, dstdir)
+bleu_metric: 80.70557274927982
+----------
+NL: find the key associated with the largest value in dictionary `x` whilst key is non-zero value
+ref: max(k for k, v in x.items() if v != 0)
+Output: max(x,��key=lambda x: x[0])
+bleu_metric: 11.483394131362408
+----------
+NL: get the largest key whose not associated with value of 0 in dictionary `x`
+ref: (k for k, v in x.items() if v != 0)
+Output: max(k for k, v in x.items()▁if v▁!= 0)
+bleu_metric: 72.00242075875519
+----------
+NL: get the largest key in a dictionary `x` with non-zero value
+ref: max(k for k, v in x.items() if v != 0)
+Output: max(k for k, v in x.items()▁if v▁!= 0)
+bleu_metric: 76.74174160136337
+----------
+NL: Put the curser at beginning of the file
+ref: file.seek(0)
+Output: time.sleep(1)
+bleu_metric: 27.776190340117914
+----------
+NL: combine values from column 'b' and column 'a' of dataframe `df` into column 'c' of datafram `df`
+ref: df['c'] = np.where(df['a'].isnull, df['b'], df['a'])
+Output: df['C'] = df['B'].apply(lambda x: x.astype(int))
+bleu_metric: 15.164081046360186
+----------
+NL: remove key 'ele' from dictionary `d`
+ref: del d['ele']
+Output: d =▁dict((k, v) for k, v in d.items()▁if v)
+bleu_metric: 5.377336385080629
+----------
+NL: Update datetime field in `MyModel` to be the existing `timestamp` plus 100 years
+ref: MyModel.objects.update(timestamp=F('timestamp') + timedelta(days=36524.25))
+Output: time.strftime('%Y-%m-%d %H:%M:%S', time.gmtime(
+bleu_metric: 7.278748444764005
+----------
+NL: merge list `['it']` and list `['was']` and list `['annoying']` into one list
+ref: ['it'] + ['was'] + ['annoying']
+Output: ▁list(zip([1'it'], ['is'], ['annoYing']))
+bleu_metric: 19.158743349944555
+----------
+NL: increment a value with leading zeroes in a number `x`
+ref: str(int(x) + 1).zfill(len(x))
+Output: ▁numpy.zeros((x + 1) + x)
+bleu_metric: 19.950354926643225
+----------
+NL: check if a pandas dataframe `df`'s index is sorted
+ref: all(df.index[:-1] <= df.index[1:])
+Output: df.index.sort()
+bleu_metric: 7.510537410248026
+----------
+NL: Convert tuple `t` to list
+ref: list(t)
+Output: tuple(t)
+bleu_metric: 75.98356856515926
+----------
+NL: Convert list `t` to tuple
+ref: tuple(l)
+Output: tuple(t)
+bleu_metric: 45.180100180492246
+----------
+NL: Convert tuple `level1` to list
+ref: level1 = map(list, level1)
+Output: ▁list(level1.decode('utf-8'))
+bleu_metric: 12.03921753741131
+----------
+NL: send the output of pprint object `dataobject` to file `logFile`
+ref: pprint.pprint(dataobject, logFile)
+Output: pprint(dataobject, logFile)
+bleu_metric: 77.8800783071405
+----------
+NL: get index of rows in column 'BoolCol'
+ref: df.loc[df['BoolCol']]
+Output: df.index
+bleu_metric: 2.5225322607845957
+----------
+NL: Create a list containing the indexes of rows where the value of column 'BoolCol' in dataframe `df` are equal to True
+ref: df.iloc[np.flatnonzero(df['BoolCol'])]
+Output: df[(df['Boolcol'] == 'true'])
+bleu_metric: 28.059228101129342
+----------
+NL: get list of indexes of rows where column 'BoolCol' values match True
+ref: df[df['BoolCol'] == True].index.tolist()
+Output: [i for i, v in enumerate(['Bool', 'Bool'])▁if v == 'true')
+bleu_metric: 11.751163762525296
+----------
+NL: get index of rows in dataframe `df` which column 'BoolCol' matches value True
+ref: df[df['BoolCol']].index.tolist()
+Output: df.index.tolist()
+bleu_metric: 28.866139906084342
+----------
+NL: change working directory to the directory `owd`
+ref: os.chdir(owd)
+Output: os.chdir('\\Users\\Desktop\\Desktop\\Desktop\\Desk
+bleu_metric: 17.001078098404225
+----------
+NL: insert data from a string `testfield` to sqlite db `c`
+ref: c.execute("INSERT INTO test VALUES (?, 'bar')", (testfield,))
+Output: ▁cursor.execute('INSERT▁INTO▁table▁VALUES (%s, %s, %s)', (testfield,)
+bleu_metric: 36.151920269854195
+----------
+NL: decode string "\\x89\\n" into a normal string
+ref: """\\x89\\n""".decode('string_escape')
+Output: """\\x89\\n""".decode('utf-8')
+bleu_metric: 79.47545184555568
+----------
+NL: convert a raw string `raw_string` into a normal string
+ref: raw_string.decode('string_escape')
+Output: ▁raw_string.decode('unicode_escape')
+bleu_metric: 53.107253497886994
+----------
+NL: convert a raw string `raw_byte_string` into a normal string
+ref: raw_byte_string.decode('unicode_escape')
+Output: ▁raw_byte_string.decode('unicode_escape')
+bleu_metric: 80.70557274927982
+----------
+NL: split a string `s` with into all strings of repeated characters
+ref: [m.group(0) for m in re.finditer('(\\d)\\1*', s)]
+Output: re.findall('[^a-zA-Z]*', s)
+bleu_metric: 17.036282985955026
+----------
+NL: scatter a plot with x, y position of `np.random.randn(100)` and face color equal to none
+ref: plt.scatter(np.random.randn(100), np.random.randn(100), facecolors='none')
+Output: plt.show()
+bleu_metric: 0.8278521686370256
+----------
+NL: do a scatter plot with empty circles
+ref: plt.plot(np.random.randn(100), np.random.randn(100), 'o', mfc='none')
+Output: plt.show()
+bleu_metric: 0.42503347540650505
+----------
+NL: remove a div with a id `main-content` using beautifulsoup
+ref: soup.find('div', id='main-content').decompose()
+Output: ▁print(soup.findAll('div', class_='maincontent'))
+bleu_metric: 34.193725219509716
+----------
+NL: filter rows containing key word `ball` in column `ids`
+ref: df[df['ids'].str.contains('ball')]
+Output: [(k, v) for k, v in df.ids]
+bleu_metric: 11.123500809946101
+----------
+NL: convert index at level 0 into a column in dataframe `df`
+ref: df.reset_index(level=0, inplace=True)
+Output: df['index'] = df['index'].apply(lambda x: x.tolist())
+bleu_metric: 7.592939129822969
+----------
+NL: Add indexes in a data frame `df` to a column `index1`
+ref: df['index1'] = df.index
+Output: df['index1'] = df['index1'].index
+bleu_metric: 55.54570250728591
+----------
+NL: convert pandas index in a dataframe to columns
+ref: df.reset_index(level=['tick', 'obs'])
+Output: df.set_index(['index', 'index'])
+bleu_metric: 32.51279032103401
+----------
+NL: Get reverse of list items from list 'b' using extended slicing
+ref: [x[::-1] for x in b]
+Output: [x[i:i + 2] for i in range(0, len(x), 2)]
+bleu_metric: 17.895706401541528
+----------
+NL: join each element in array `a` with element at the same index in array `b` as a tuple
+ref: np.array([zip(x, y) for x, y in zip(a, b)])
+Output: A = [(a + b) for a, b in zip(A, B)]
+bleu_metric: 18.114844247735682
+----------
+NL: zip two 2-d arrays `a` and `b`
+ref: np.array(zip(a.ravel(), b.ravel()), dtype='i4,i4').reshape(a.shape)
+Output: np.array(zip(a, b))
+bleu_metric: 10.156383094063225
+----------
+NL: convert list `list_of_ints` into a comma separated string
+ref: """,""".join([str(i) for i in list_of_ints])
+Output: """""".join(list_of_ints)
+bleu_metric: 31.139742129493257
+----------
+NL: Send a post request with raw data `DATA` and basic authentication with `username` and `password`
+ref: requests.post(url, data=DATA, headers=HEADERS_DICT, auth=(username, password))
+Output: r =▁requests.post(url, data=data,▁headers=headers,▁files=passwords)
+bleu_metric: 38.57153091307505
+----------
+NL: Find last occurrence of character '}' in string "abcd}def}"
+ref: 'abcd}def}'.rfind('}')
+Output: """abcd}def}""".rfind('}.rsplit('\\d+')
+bleu_metric: 42.26469376227503
+----------
+NL: Iterate ove list `[1, 2, 3]` using list comprehension
+ref: print([item for item in [1, 2, 3]])
+Output: [(i, j) for i, j in zip([1, 2, 3])▁if j == j]
+bleu_metric: 27.061633707336412
+----------
+NL: extract all the values with keys 'x' and 'y' from a list of dictionaries `d` to list of tuples
+ref: [(x['x'], x['y']) for x in d]
+Output: [d['x'], d['y'], d['d'], d['x'], d[
+bleu_metric: 30.398382447928807
+----------
+NL: get the filename without the extension from file 'hemanth.txt'
+ref: print(os.path.splitext(os.path.basename('hemanth.txt'))[0])
+Output: ▁print(os.path.splitext('hemanth.txt'))
+bleu_metric: 51.55938317965318
+----------
+NL: create a dictionary by adding each two adjacent elements in tuple `x` as key/value pair to it
+ref: dict(x[i:i + 2] for i in range(0, len(x), 2))
+Output: ▁dict(zip(x[::2], x[1::2]))
+bleu_metric: 16.45452192914717
+----------
+NL: create a list containing flattened list `[['A', 'B', 'C'], ['D', 'E', 'F'], ['G', 'H', 'I']]`
+ref: values = sum([['A', 'B', 'C'], ['D', 'E', 'F'], ['G', 'H', 'I']], [])
+Output: [['A', 'B', 'C'], ['D', 'E', 'F'], ['
+bleu_metric: 53.170929354453335
+----------
+NL: select rows in a dataframe `df` column 'closing_price' between two values 99 and 101
+ref: df = df[(df['closing_price'] >= 99) & (df['closing_price'] <= 101)]
+Output: df.loc[df['cold_Price'] == 99]
+bleu_metric: 9.613222089577718
+----------
+NL: replace all occurences of newlines `\n` with `
` in dataframe `df`
+ref: df.replace({'\n': '
'}, regex=True)
+Output: df = pd.read_csv('\n').replace('\n', '
')
+bleu_metric: 33.87144423286913
+----------
+NL: replace all occurrences of a string `\n` by string `
` in a pandas data frame `df`
+ref: df.replace({'\n': '
'}, regex=True)
+Output: df = pd.read_csv('\n').replace('\n', '
')
+bleu_metric: 33.87144423286913
+----------
+NL: create a list containing each two adjacent letters in string `word` as its elements
+ref: [(x + y) for x, y in zip(word, word[1:])]
+Output: [word[i:i + 2] for i in range(0, len(word), 2)]
+bleu_metric: 15.253223052326929
+----------
+NL: Get a list of pairs from a string `word` using lambda function
+ref: list(map(lambda x, y: x + y, word[:-1], word[1:]))
+Output: ▁list(word.split(','))
+bleu_metric: 7.965309748053327
+----------
+NL: extract a url from a string `myString`
+ref: print(re.findall('(https?://[^\\s]+)', myString))
+Output: re.findall('\\w+', myString)
+bleu_metric: 25.34712979535927
+----------
+NL: extract a url from a string `myString`
+ref: print(re.search('(?Phttps?://[^\\s]+)', myString).group('url'))
+Output: re.findall('\\w+', myString)
+bleu_metric: 8.146914412736555
+----------
+NL: remove all special characters, punctuation and spaces from a string `mystring` using regex
+ref: re.sub('[^A-Za-z0-9]+', '', mystring)
+Output: """""".join(re.sub('\\w+', '', mystring))
+bleu_metric: 40.6731552331291
+----------
+NL: create a DatetimeIndex containing 13 periods of the second friday of each month starting from date '2016-01-01'
+ref: pd.date_range('2016-01-01', freq='WOM-2FRI', periods=13)
+Output: ▁datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
+bleu_metric: 8.004434939657356
+----------
+NL: Create multidimensional array `matrix` with 3 rows and 2 columns in python
+ref: matrix = [[a, b], [c, d], [e, f]]
+Output: ▁matrix = [[[0, 0], [0, 0], [0, 0], [0, 0], [0, 0]], [0,
+bleu_metric: 15.448396990132697
+----------
+NL: replace spaces with underscore
+ref: mystring.replace(' ', '_')
+Output: re.sub('(?<=[a-z])(?=[A-Z])', '\\1'
+bleu_metric: 9.500266321476488
+----------
+NL: get an absolute file path of file 'mydir/myfile.txt'
+ref: os.path.abspath('mydir/myfile.txt')
+Output: os.path.abspath('mydir/myfile.txt')
+bleu_metric: 100.0
+----------
+NL: split string `my_string` on white spaces
+ref: """ """.join(my_string.split())
+Output: """ """.join(my_string.split())
+bleu_metric: 100.0
+----------
+NL: get filename without extension from file `filename`
+ref: os.path.splitext(filename)[0]
+Output: os.path.splitext(filename)
+bleu_metric: 64.74591278836638
+----------
+NL: get a list containing the sum of each element `i` in list `l` plus the previous elements
+ref: [sum(l[:i]) for i, _ in enumerate(l)]
+Output: [sum(i) for i, j in zip(l, l)]
+bleu_metric: 41.46278264707056
+----------
+NL: split a string `Docs/src/Scripts/temp` by `/` keeping `/` in the result
+ref: """Docs/src/Scripts/temp""".replace('/', '/\x00/').split('\x00')
+Output: ▁print(docs/src/Scripts/temp.split('/'))
+bleu_metric: 23.898608686716166
+----------
+NL: shuffle columns of an numpy array 'r'
+ref: np.random.shuffle(np.transpose(r))
+Output: np.random.shuffle(np.random.randn(r))
+bleu_metric: 69.64705665515707
+----------
+NL: copy all values in a column 'B' to a new column 'D' in a pandas data frame 'df'
+ref: df['D'] = df['B']
+Output: df['B'] = df['D'].values.tolist()
+bleu_metric: 63.53091572554783
+----------
+NL: find a value within nested json 'data' where the key inside another key 'B' is unknown.
+ref: list(data['A']['B'].values())[0]['maindata'][0]['Info']
+Output: [item for item in data▁if item['B'] not in item]
+bleu_metric: 12.5687883644936
+----------
+NL: check characters of string `string` are true predication of function `predicate`
+ref: all(predicate(x) for x in string)
+Output: ▁any(pred(x) for x in▁string.split())
+bleu_metric: 33.92375549949802
+----------
+NL: determine number of files on a drive with python
+ref: os.statvfs('/').f_files - os.statvfs('/').f_ffree
+Output: ▁sum(i for i in os.listdir('filename')▁if os.path.isfile(filename))
+bleu_metric: 10.12895233093219
+----------
+NL: how to get a single result from a SQLite query in python?
+ref: cursor.fetchone()[0]
+Output: ▁cursor.execute('INSERT▁INTO▁table▁VALUES (%s, %s, %s)', (var1, var2,
+bleu_metric: 5.912329024877923
+----------
+NL: convert string `user_input` into a list of integers `user_list`
+ref: user_list = [int(number) for number in user_input.split(',')]
+Output: user_list = [int(i) for i in▁user_input.split(',')]
+bleu_metric: 66.43548861507492
+----------
+NL: Get a list of integers by splitting a string `user` with comma
+ref: [int(s) for s in user.split(',')]
+Output: [int(x) for x in▁user.split(',')]
+bleu_metric: 64.55651607183006
+----------
+NL: Sorting a Python list by two criteria
+ref: sorted(list, key=lambda x: (x[0], -x[1]))
+Output: a.sort(key=lambda x: (x[0], x[0]), reverse=True)
+bleu_metric: 49.45703137422567
+----------
+NL: sort a list of objects `ut`, based on a function `cmpfun` in descending order
+ref: ut.sort(key=cmpfun, reverse=True)
+Output: ▁sorted(ut,▁key=lambda x: func(x, -1))
+bleu_metric: 10.813005337959174
+----------
+NL: reverse list `ut` based on the `count` attribute of each object
+ref: ut.sort(key=lambda x: x.count, reverse=True)
+Output: ▁print(reversed(ut,▁key=lambda x: x.count(), reverse=True))
+bleu_metric: 45.26353562390146
+----------
+NL: sort a list of objects `ut` in reverse order by their `count` property
+ref: ut.sort(key=lambda x: x.count, reverse=True)
+Output: ▁sorted(ut,▁key=lambda x:▁count.count(x[0], x[0]))
+bleu_metric: 20.563241139732103
+----------
+NL: click a href button 'Send' with selenium
+ref: driver.find_element_by_partial_link_text('Send').click()
+Output: ▁driver.find_element_by_css_selector('send')
+bleu_metric: 20.083909447408573
+----------
+NL: click a href button having text `Send InMail` with selenium
+ref: driver.findElement(By.linkText('Send InMail')).click()
+Output: ▁driver.find_element_by_xpath('send inMail').click()
+bleu_metric: 33.83154336027537
+----------
+NL: click a href button with text 'Send InMail' with selenium
+ref: driver.find_element_by_link_text('Send InMail').click()
+Output: ▁driver.find_element_by_css_selector('Send InMail').click()
+bleu_metric: 73.51460991014883
+----------
+NL: cast an int `i` to a string and concat to string 'ME'
+ref: 'ME' + str(i)
+Output: """ME""".encode('utf-8')
+bleu_metric: 17.124730448946572
+----------
+NL: Sorting data in DataFrame Pandas
+ref: df.sort_values(['System_num', 'Dis'])
+Output: df.sort_values(['a', 'b'],▁ascending=[True, False],
+bleu_metric: 31.011575752288344
+----------
+NL: prepend the line '#test firstline\n' to the contents of file 'infile' and save as the file 'outfile'
+ref: open('outfile', 'w').write('#test firstline\n' + open('infile').read())
+Output: open('infile', '#test firstline\n')
+bleu_metric: 23.71937737244433
+----------
+NL: sort a list `l` by length of value in tuple
+ref: l.sort(key=lambda t: len(t[1]), reverse=True)
+Output: ▁sorted(l,▁key=len)
+bleu_metric: 8.655647308002333
+----------
+NL: split string `s` by words that ends with 'd'
+ref: re.findall('\\b(\\w+)d\\b', s)
+Output: re.split('\\s+', s)
+bleu_metric: 28.449457729768916
+----------
+NL: return `True` if string `foobarrrr` contains regex `ba[rzd]`
+ref: bool(re.search('ba[rzd]', 'foobarrrr'))
+Output: """""".join(re.findall('[^a-zd]*', 'foobarrarrrr
+bleu_metric: 15.35259783865636
+----------
+NL: Removing duplicates in list `t`
+ref: list(set(t))
+Output: ▁list(set(t))
+bleu_metric: 78.60753021519787
+----------
+NL: Removing duplicates in list `source_list`
+ref: list(set(source_list))
+Output: ▁print([item for item in▁list(source_list)▁if item.endswith('item')])
+bleu_metric: 11.295714543953919
+----------
+NL: Removing duplicates in list `abracadabra`
+ref: list(OrderedDict.fromkeys('abracadabra'))
+Output: [dict((x, x) for x, y in▁itertools.groupby(lambda x: x.group(0))
+bleu_metric: 8.635800047213174
+----------
+NL: Convert array `a` into a list
+ref: numpy.array(a).reshape(-1).tolist()
+Output: ▁list(array.from_iterable(a))
+bleu_metric: 16.778292801990535
+----------
+NL: Convert the first row of numpy matrix `a` to a list
+ref: numpy.array(a)[0].tolist()
+Output: a = np.array(a)
+bleu_metric: 32.2645258949984
+----------
+NL: In `soup`, get the content of the sibling of the `td` tag with text content `Address:`
+ref: print(soup.find(text='Address:').findNext('td').contents[0])
+Output: soup.find_element_by_xpath("//p[@id, 'id']/text()").
+bleu_metric: 12.825436069560483
+----------
+NL: convert elements of each tuple in list `l` into a string separated by character `@`
+ref: """ """.join([('%d@%d' % t) for t in l])
+Output: """""".join(l[0] for l in zip(l[::2], l[1::2]))
+bleu_metric: 32.31269177549822
+----------
+NL: convert each tuple in list `l` to a string with '@' separating the tuples' elements
+ref: """ """.join([('%d@%d' % (t[0], t[1])) for t in l])
+Output: [(x, y) for x, y in zip(l, l[::2], l[1::2])]
+bleu_metric: 8.274984205902886
+----------
+NL: get the html from the current web page of a Selenium driver
+ref: driver.execute_script('return document.documentElement.outerHTML;')
+Output: ▁driver.get('http://stackoverflow.com')
+bleu_metric: 18.710158230410627
+----------
+NL: Get all matches with regex pattern `\\d+[xX]` in list of string `teststr`
+ref: [i for i in teststr if re.search('\\d+[xX]', i)]
+Output: re.findall('\\d+[xX]', teststr)
+bleu_metric: 43.99022083366858
+----------
+NL: select values from column 'A' for which corresponding values in column 'B' will be greater than 50, and in column 'C' - equal 900 in dataframe `df`
+ref: df['A'][(df['B'] > 50) & (df['C'] == 900)]
+Output: df['B'] = df['C'].str.lower()
+bleu_metric: 32.3967345710975
+----------
+NL: Sort dictionary `o` in ascending order based on its keys and items
+ref: sorted(o.items())
+Output: ▁sorted(list(o.items()),▁key=lambda x: x[1])
+bleu_metric: 29.452884374019472
+----------
+NL: get sorted list of keys of dict `d`
+ref: sorted(d)
+Output: ▁sorted(list(d.items()),▁key=lambda x: x[1])
+bleu_metric: 9.043565779655339
+----------
+NL: How to sort dictionaries by keys in Python
+ref: sorted(d.items())
+Output: ▁sorted(list(d.items()),▁key=lambda x: x[1])
+bleu_metric: 29.452884374019472
+----------
+NL: convert string "1" into integer
+ref: int('1')
+Output: int('1', 2)
+bleu_metric: 58.143073696821936
+----------
+NL: function to convert strings into integers
+ref: int()
+Output: int(''.join(map(str, s)))
+bleu_metric: 12.729922658368398
+----------
+NL: convert items in `T1` to integers
+ref: T2 = [map(int, x) for x in T1]
+Output: int(''.join(map(str,▁row)))
+bleu_metric: 11.75029694362029
+----------
+NL: call a shell script `./test.sh` using subprocess
+ref: subprocess.call(['./test.sh'])
+Output: subprocess.call(['./test.sh'])
+bleu_metric: 100.0
+----------
+NL: call a shell script `notepad` using subprocess
+ref: subprocess.call(['notepad'])
+Output: subprocess.call(['notepad'])
+bleu_metric: 100.0
+----------
+NL: combine lists `l1` and `l2` by alternating their elements
+ref: [val for pair in zip(l1, l2) for val in pair]
+Output: [(x + y) for x, y in zip(l1, l2)]
+bleu_metric: 40.59264050928986
+----------
+NL: encode string 'data to be encoded'
+ref: encoded = base64.b64encode('data to be encoded')
+Output: 'data to be encoded'.encode('data to be encoded')
+bleu_metric: 50.495676290868666
+----------
+NL: encode a string `data to be encoded` to `ascii` encoding
+ref: encoded = 'data to be encoded'.encode('ascii')
+Output: ▁print(ascii.encode('data to be encoded'))
+bleu_metric: 51.09375879171261
+----------
+NL: parse tab-delimited CSV file 'text.txt' into a list
+ref: lol = list(csv.reader(open('text.txt', 'rb'), delimiter='\t'))
+Output: ▁writer.writerows(['text.txt', 'rb').split(','))
+bleu_metric: 43.42718513715163
+----------
+NL: Get attribute `my_str` of object `my_object`
+ref: getattr(my_object, my_str)
+Output: ▁getattr(my_str, my_object)
+bleu_metric: 26.26909894424158
+----------
+NL: group a list of dicts `LD` into one dict by key
+ref: print(dict(zip(LD[0], zip(*[list(d.values()) for d in LD]))))
+Output: mydict = {k: v for k, v in▁list(LD.items())}
+bleu_metric: 9.34970953032271
+----------
+NL: How do I sum the first value in each tuple in a list of tuples in Python?
+ref: sum([pair[0] for pair in list_of_pairs])
+Output: ▁sum(x[0] for x in▁rows)
+bleu_metric: 30.130404892785684
+----------
+NL: convert unicode string u"{'code1':1,'code2':1}" into dictionary
+ref: d = ast.literal_eval("{'code1':1,'code2':1}")
+Output: ▁dict(('{'code1':1,'code2':1}'.encode('unicode_escape')
+bleu_metric: 61.32878268086118
+----------
+NL: find all words in a string `mystring` that start with the `$` sign
+ref: [word for word in mystring.split() if word.startswith('$')]
+Output: re.findall('\\w+', mystring)
+bleu_metric: 8.395097551291249
+----------
+NL: remove any url within string `text`
+ref: text = re.sub('^https?:\\/\\/.*[\\r\\n]*', '', text, flags=re.MULTILINE)
+Output: ▁url = re.sub('▁?\\s+', '', text)
+bleu_metric: 21.464898124935406
+----------
+NL: replace all elements in array `A` that are not present in array `[1, 3, 4]` with zeros
+ref: np.where(np.in1d(A, [1, 3, 4]).reshape(A.shape), A, 0)
+Output: np.where(np.isnan(a), [1, 3, 4])
+bleu_metric: 40.9574484442826
+----------
+NL: calculate mean across dimension in a 2d array `a`
+ref: np.mean(a, axis=1)
+Output: np.mean(a,▁axis=1)
+bleu_metric: 76.11606003349893
+----------
+NL: running r script '/pathto/MyrScript.r' from python
+ref: subprocess.call(['/usr/bin/Rscript', '--vanilla', '/pathto/MyrScript.r'])
+Output: os.system('/pathto/MyrScript.r')
+bleu_metric: 20.95597855867423
+----------
+NL: run r script '/usr/bin/Rscript --vanilla /pathto/MyrScript.r'
+ref: subprocess.call('/usr/bin/Rscript --vanilla /pathto/MyrScript.r', shell=True)
+Output: os.system('/usr/bin/rscript --vanilla /pathto/MyrScript.r')
+bleu_metric: 60.11385699767446
+----------
+NL: add a header to a csv file
+ref: writer.writeheader()
+Output: ▁writer.writeheader()
+bleu_metric: 72.59795291154771
+----------
+NL: replacing nan in the dataframe `df` with row average
+ref: df.fillna(df.mean(axis=1), axis=1)
+Output: df.fillna(lambda x: x.fillna(x.mean(),▁axis=1))
+bleu_metric: 31.449494592022948
+----------
+NL: Convert unix timestamp '1347517370' to formatted string '%Y-%m-%d %H:%M:%S'
+ref: time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(1347517370))
+Output: time.strftime('.%Y-%m-%d %H:%M:%S')
+bleu_metric: 66.6355295714607
+----------
+NL: Call a base class's class method `do` from derived class `Derived`
+ref: super(Derived, cls).do(a)
+Output: super(Do,▁self).__init__(*args,▁**kwargs)
+bleu_metric: 13.26475916741222
+----------
+NL: selecting rows in Numpy ndarray 'a', where the value in the first column is 0 and value in the second column is 1
+ref: a[np.where((a[:, (0)] == 0) * (a[:, (1)] == 1))]
+Output: ▁numpy.array([[0, 0, 1, 0], [0, 1, 0], [0, 1, 0], [
+bleu_metric: 5.636775411569542
+----------
+NL: separate words delimited by one or more spaces into a list
+ref: re.split(' +', 'hello world sample text')
+Output: [' '.join(l) for l in re.split('\\s+','', s)]
+bleu_metric: 23.977362037775197
+----------
+NL: length of longest element in list `words`
+ref: len(max(words, key=len))
+Output: len(words) - len(words)
+bleu_metric: 21.258637840736228
+----------
+NL: get the value associated with unicode key 'from_user' of first dictionary in list `result`
+ref: result[0]['from_user']
+Output: ▁print([x['from_user'] for x in▁list(result.items())])
+bleu_metric: 19.427446513842177
+----------
+NL: Retrieve each line from a file 'File.txt' as a list
+ref: [line.split() for line in open('File.txt')]
+Output: ▁file.write('File.txt')
+bleu_metric: 32.500555971011345
+----------
+NL: swap keys with values in a dictionary `a`
+ref: res = dict((v, k) for k, v in a.items())
+Output: ▁dict((v, k) for k, v in a.items())
+bleu_metric: 90.21315316859231
+----------
+NL: Open a file `path/to/FILE_NAME.ext` in write mode
+ref: new_file = open('path/to/FILE_NAME.ext', 'w')
+Output: os.system('path/to/FILE_NAME.ext')
+bleu_metric: 55.9978208842951
+----------
+NL: How to count distinct values in a column of a pandas group by object?
+ref: df.groupby(['col1', 'col2'])['col3'].nunique().reset_index()
+Output: df.groupby(['col1', 'col2']).apply(lambda x: x.groupby('
+bleu_metric: 54.116466213090966
+----------
+NL: Check if any key in the dictionary `dict1` starts with the string `EMP$$`
+ref: any(key.startswith('EMP$$') for key in dict1)
+Output: ▁any(k for k, v in▁dict1.items()▁if 'EMP$$$' in k)
+bleu_metric: 19.91148803162055
+----------
+NL: create list of values from dictionary `dict1` that have a key that starts with 'EMP$$'
+ref: [value for key, value in list(dict1.items()) if key.startswith('EMP$$')]
+Output: [d['EMP$'] for d in▁dict1.values()]
+bleu_metric: 14.693064660889684
+----------
+NL: convert a pandas series `sf` into a pandas dataframe `df` with columns `email` and `list`
+ref: pd.DataFrame({'email': sf.index, 'list': sf.values})
+Output: df = pd.DataFrame(sf,▁columns=['email', 'list'])
+bleu_metric: 35.69768415530451
+----------
+NL: print elements of list `list` seperated by tabs `\t`
+ref: print('\t'.join(map(str, list)))
+Output: ▁print('\t'.join(map(str,▁list)))
+bleu_metric: 77.39321540474097
+----------
+NL: print unicode string '\xd0\xbf\xd1\x80\xd0\xb8' with utf-8
+ref: print('\xd0\xbf\xd1\x80\xd0\xb8'.encode('raw_unicode_escape'))
+Output: ▁print('\xd0\xbf\xd1\x80\xd0\xb8'.encode(
+bleu_metric: 75.96292058519337
+----------
+NL: Encode a latin character in string `Sopet\xc3\xb3n` properly
+ref: 'Sopet\xc3\xb3n'.encode('latin-1').decode('utf-8')
+Output: sopet\xc3\xb3n.encode('latin')
+bleu_metric: 21.252609690759098
+----------
+NL: resized image `image` to width, height of `(x, y)` with filter of `ANTIALIAS`
+ref: image = image.resize((x, y), Image.ANTIALIAS)
+Output: image =▁Image.reset_image(image, (x, y))
+bleu_metric: 37.432615061973365
+----------
+NL: regex, find "n"s only in the middle of string `s`
+ref: re.findall('n(?<=[^n]n)n+(?=[^n])(?i)', s)
+Output: re.findall('\\d+', s)
+bleu_metric: 13.004547755725982
+----------
+NL: display the float `1/3*100` as a percentage
+ref: print('{0:.0f}%'.format(1.0 / 3 * 100))
+Output: ▁print('{0:.2f}'.format(1/3*100))
+bleu_metric: 65.24084871752925
+----------
+NL: sort a list of dictionary `mylist` by the key `title`
+ref: mylist.sort(key=lambda x: x['title'])
+Output: mylist.sort(key=lambda x: x['title'])
+bleu_metric: 100.0
+----------
+NL: sort a list `l` of dicts by dict value 'title'
+ref: l.sort(key=lambda x: x['title'])
+Output: ▁sorted(l,▁key=itemgetter('title', 'title'))
+bleu_metric: 15.448759309616733
+----------
+NL: sort a list of dictionaries by the value of keys 'title', 'title_url', 'id' in ascending order.
+ref: l.sort(key=lambda x: (x['title'], x['title_url'], x['id']))
+Output: ▁sorted(list_of_dicts,▁key=lambda x: x['title']['title'], reverse=Tru
+bleu_metric: 34.214468780181804
+----------
+NL: find 10 largest differences between each respective elements of list `l1` and list `l2`
+ref: heapq.nlargest(10, range(len(l1)), key=lambda i: abs(l1[i] - l2[i]))
+Output: max(l1,▁key=lambda x: (x[1], x[0]), reverse=True)
+bleu_metric: 11.878881948379775
+----------
+NL: BeautifulSoup find all 'span' elements in HTML string `soup` with class of 'starGryB sp'
+ref: soup.find_all('span', {'class': 'starGryB sp'})
+Output: soup.findAll('starGryB sp')
+bleu_metric: 27.40227985814397
+----------
+NL: write records in dataframe `df` to table 'test' in schema 'a_schema'
+ref: df.to_sql('test', engine, schema='a_schema')
+Output: df.to_csv('test')
+bleu_metric: 23.118543024585154
+----------
+NL: Extract brackets from string `s`
+ref: brackets = re.sub('[^(){}[\\]]', '', s)
+Output: re.findall('\\[(\\w+)\\]', s)
+bleu_metric: 28.446938709683227
+----------
+NL: remove duplicate elements from list 'L'
+ref: list(dict((x[0], x) for x in L).values())
+Output: ▁list(set(l))
+bleu_metric: 7.902834459544201
+----------
+NL: read a file `file` without newlines
+ref: [line.rstrip('\n') for line in file]
+Output: open('file', 'w').readline()
+bleu_metric: 13.399118318121793
+----------
+NL: get the position of item 1 in `testlist`
+ref: [i for (i, x) in enumerate(testlist) if (x == 1)]
+Output: ▁print(testlist.position())
+bleu_metric: 6.435022756527756
+----------
+NL: get the position of item 1 in `testlist`
+ref: [i for (i, x) in enumerate(testlist) if (x == 1)]
+Output: ▁print(testlist.position())
+bleu_metric: 6.435022756527756
+----------
+NL: get the position of item 1 in `testlist`
+ref: for i in [i for (i, x) in enumerate(testlist) if (x == 1)]:
+ pass
+Output: ▁print(testlist.position())
+bleu_metric: 4.559019536899008
+----------
+NL: get the position of item 1 in `testlist`
+ref: for i in (i for (i, x) in enumerate(testlist) if (x == 1)):
+ pass
+Output: ▁print(testlist.position())
+bleu_metric: 5.045384493046664
+----------
+NL: get the position of item 1 in `testlist`
+ref: gen = (i for (i, x) in enumerate(testlist) if (x == 1))
+Output: ▁print(testlist.position())
+bleu_metric: 8.036848499107174
+----------
+NL: get the position of item `element` in list `testlist`
+ref: print(testlist.index(element))
+Output: ▁print(testlist.index(element))
+bleu_metric: 82.42367502646054
+----------
+NL: get the position of item `element` in list `testlist`
+ref: gen = (i for (i, x) in enumerate(testlist) if (x == 1))
+Output: ▁print(testlist.index(element))
+bleu_metric: 8.792719200156416
+----------
+NL: find the first element of the tuple with the maximum second element in a list of tuples `lis`
+ref: max(lis, key=lambda item: item[1])[0]
+Output: max(lis,▁key=lambda x: x[1])[0]
+bleu_metric: 60.39435155169263
+----------
+NL: get the item at index 0 from the tuple that has maximum value at index 1 in list `lis`
+ref: max(lis, key=itemgetter(1))[0]
+Output: max(enumerate(a),▁key=lambda x: x[1])[0]
+bleu_metric: 27.274191069381914
+----------
+NL: Make a delay of 1 second
+ref: time.sleep(1)
+Output: time.sleep(1)
+bleu_metric: 100.0
+----------
+NL: convert list of tuples `L` to a string
+ref: """, """.join('(' + ', '.join(i) + ')' for i in L)
+Output: """""".join(l)
+bleu_metric: 14.164334116374599
+----------
+NL: Django set default value of field `b` equal to '0000000'
+ref: b = models.CharField(max_length=7, default='0000000', editable=False)
+Output: super(B,▁self).save(*args,▁**kwargs)
+bleu_metric: 8.945053454772305
+----------
+NL: Sort lis `list5` in ascending order based on the degrees value of its elements
+ref: sorted(list5, lambda x: (degree(x), x))
+Output: ▁sorted(list5,▁key=lambda x: (int(x)[0], x))
+bleu_metric: 40.802623528764286
+----------
+NL: How do I perform secondary sorting in python?
+ref: sorted(list5, key=lambda vertex: (degree(vertex), vertex))
+Output: ▁sorted(list(range(10)),▁key=lambda x: x[1])
+bleu_metric: 15.94000950514835
+----------
+NL: convert a list into a generator object
+ref: (n for n in [1, 2, 3, 5])
+Output: ▁list(itertools.product(list(range(10)),▁repeat=4))
+bleu_metric: 8.282282660969601
+----------
+NL: remove elements from list `oldlist` that have an index number mentioned in list `removelist`
+ref: newlist = [v for i, v in enumerate(oldlist) if i not in removelist]
+Output: [x for x in a▁if x not in b]
+bleu_metric: 11.298101090980193
+----------
+NL: Open a file `yourfile.txt` in write mode
+ref: f = open('yourfile.txt', 'w')
+Output: open('yourfile.txt', 'r')
+bleu_metric: 67.09548263862051
+----------
+NL: get attribute 'attr' from object `obj`
+ref: getattr(obj, 'attr')
+Output: ▁getattr(obj, 'attr')
+bleu_metric: 80.70557274927982
+----------
+NL: convert tuple of tuples `(('aa',), ('bb',), ('cc',))` to tuple
+ref: from functools import reduce
+reduce(lambda a, b: a + b, (('aa',), ('bb',), ('cc',)))
+Output: tuple(('aa',), ('bb',), ('cc',))
+bleu_metric: 50.621900464488824
+----------
+NL: convert tuple of tuples `(('aa',), ('bb',), ('cc',))` to list in one line
+ref: map(lambda a: a[0], (('aa',), ('bb',), ('cc',)))
+Output: ▁list(('aa',), ('bb',), ('cc',))
+bleu_metric: 64.0614517827111
+----------
+NL: Python Pandas: How to replace a characters in a column of a dataframe?
+ref: df['range'].replace(',', '-', inplace=True)
+Output: df['c'] = df['d'].apply(lambda x: x.replace('-',
+bleu_metric: 28.80991254706902
+----------
+NL: unzip the list `[('a', 1), ('b', 2), ('c', 3), ('d', 4)]`
+ref: zip(*[('a', 1), ('b', 2), ('c', 3), ('d', 4)])
+Output: zip([('a', 1), ('b', 2), ('c', 3), ('d', 4)]
+bleu_metric: 89.10617687473993
+----------
+NL: unzip the list `[('a', 1), ('b', 2), ('c', 3), ('d', 4)]`
+ref: zip(*[('a', 1), ('b', 2), ('c', 3), ('d', 4)])
+Output: zip([('a', 1), ('b', 2), ('c', 3), ('d', 4)]
+bleu_metric: 89.10617687473993
+----------
+NL: unzip list `original`
+ref: result = ([a for (a, b) in original], [b for (a, b) in original])
+Output: zip(*Ordered)
+bleu_metric: 0.7985269443272172
+----------
+NL: unzip list `original` and return a generator
+ref: result = ((a for (a, b) in original), (b for (a, b) in original))
+Output: ▁return (i for i in range(10)▁if i < 0 for i in range(10))
+bleu_metric: 10.587752134048651
+----------
+NL: unzip list `[('a', 1), ('b', 2), ('c', 3), ('d', 4), ('e', )]`
+ref: zip(*[('a', 1), ('b', 2), ('c', 3), ('d', 4), ('e',)])
+Output: zip([('a', 1), ('b', 2), ('c', 3), ('d', 4),
+bleu_metric: 73.36039503547302
+----------
+NL: unzip list `[('a', 1), ('b', 2), ('c', 3), ('d', 4), ('e', )]` and fill empty results with None
+ref: map(None, *[('a', 1), ('b', 2), ('c', 3), ('d', 4), ('e',)])
+Output: zip(*[('a', 1), ('b', 2), ('c', 3), ('d', 4
+bleu_metric: 64.62811194838824
+----------
+NL: encode `Decimal('3.9')` to a JSON string
+ref: json.dumps(Decimal('3.9'))
+Output: ▁print(json.dumps({'Temperature': 'Temperature', 'Temperature': 'Temper
+bleu_metric: 17.059573701616802
+----------
+NL: Add key "mynewkey" to dictionary `d` with value "mynewvalue"
+ref: d['mynewkey'] = 'mynewvalue'
+Output: d = {'mynewkey':'mynewvalue','mynewkey':'mynewvalue'}
+bleu_metric: 17.976451428761386
+----------
+NL: Add key 'a' to dictionary `data` with value 1
+ref: data.update({'a': 1, })
+Output: data.update({'a': 1, 'b': 1})
+bleu_metric: 63.83240325919925
+----------
+NL: Add key 'a' to dictionary `data` with value 1
+ref: data.update(dict(a=1))
+Output: data.update({'a': 1, 'b': 1})
+bleu_metric: 21.951524426618455
+----------
+NL: Add key 'a' to dictionary `data` with value 1
+ref: data.update(a=1)
+Output: data.update({'a': 1, 'b': 1})
+bleu_metric: 21.951524426618455
+----------
+NL: find maximal value in matrix `matrix`
+ref: max([max(i) for i in matrix])
+Output: np.max(matrix,▁axis=0)
+bleu_metric: 15.233734832972273
+----------
+NL: Round number `answer` to 2 precision after the decimal point
+ref: answer = str(round(answer, 2))
+Output: ▁round(answer, 2)
+bleu_metric: 49.196255036686594
+----------
+NL: extract ip address from an html string
+ref: ip = re.findall('[0-9]+(?:\\.[0-9]+){3}', s)
+Output: re.findall('[a-zA-Z0-9-]+', text)
+bleu_metric: 20.302505814938357
+----------
+NL: filter dataframe `df` by values in column `A` that appear more than once
+ref: df.groupby('A').filter(lambda x: len(x) > 1)
+Output: df['A'] = df['B'].filter(df['A'])
+bleu_metric: 15.604242268653643
+----------
+NL: append each line in file `myfile` into a list
+ref: [x for x in myfile.splitlines() if x != '']
+Output: with open('myfile', 'w').append(line)
+bleu_metric: 9.082300379267782
+----------
+NL: Get a list of integers `lst` from a file `filename.txt`
+ref: lst = map(int, open('filename.txt').readlines())
+Output: [int(x) for x in open('filename.txt', 'lst')]
+bleu_metric: 34.46913316729032
+----------
+NL: add color bar with image `mappable` to plot `plt`
+ref: plt.colorbar(mappable=mappable, cax=ax3)
+Output: plt.show()
+bleu_metric: 11.914378950211951
+----------
+NL: count most frequent 100 words in column 'text' of dataframe `df`
+ref: Counter(' '.join(df['text']).split()).most_common(100)
+Output: df['text'].str.contains(100)
+bleu_metric: 27.101674478567933
+----------
+NL: Python split a string using regex
+ref: re.findall('(.+?):(.+?)\\b ?', text)
+Output: re.split('\\s+', s)
+bleu_metric: 9.868832370006283
+----------
+NL: generate all 2-element subsets of tuple `(1, 2, 3)`
+ref: list(itertools.combinations((1, 2, 3), 2))
+Output: ▁list(itertools.product(list(range(1, 2, 3)),▁repeat=4))
+bleu_metric: 39.47641631338841
+----------
+NL: get a value of datetime.today() in the UTC time zone
+ref: datetime.now(pytz.utc)
+Output: ▁datetime.utcnow().strftime('%Y-%m-%d %H:%M')
+bleu_metric: 8.085298080223222
+----------
+NL: Get a new list `list2`by removing empty list from a list of lists `list1`
+ref: list2 = [x for x in list1 if x != []]
+Output: ▁list2 = [x for x in▁list1▁if x is not▁None]
+bleu_metric: 39.5373839333436
+----------
+NL: Create `list2` to contain the lists from list `list1` excluding the empty lists from `list1`
+ref: list2 = [x for x in list1 if x]
+Output: ▁list2 = [[] for i in▁list1, 0]
+bleu_metric: 18.842393723950337
+----------
+NL: Django response with JSON `data`
+ref: return HttpResponse(data, mimetype='application/json')
+Output: ▁return HttpResponse(json.dumps(data))
+bleu_metric: 29.361568020583483
+----------
+NL: get all text that is not enclosed within square brackets in string `example_str`
+ref: re.findall('(.*?)\\[.*?\\]', example_str)
+Output: ▁example_str.split('\\[(\\w+)\\]', '\\[\\]]')
+bleu_metric: 23.62618715634458
+----------
+NL: Use a regex to get all text in a string `example_str` that is not surrounded by square brackets
+ref: re.findall('(.*?)(?:\\[.*?\\]|$)', example_str)
+Output: ▁print(re.findall('\\[(\\w+)\\]',▁example_str))
+bleu_metric: 27.808268092146015
+----------
+NL: get whatever is between parentheses as a single match, and any char outside as an individual match in string '(zyx)bc'
+ref: re.findall('\\(.+?\\)|\\w', '(zyx)bc')
+Output: re.search('(zyx)bc', '(zyx)bc')
+bleu_metric: 37.12324840533176
+----------
+NL: match regex '\\((.*?)\\)|(\\w)' with string '(zyx)bc'
+ref: re.findall('\\((.*?)\\)|(\\w)', '(zyx)bc')
+Output: re.findall('\\((.*?)\\)(\\w)', '(zyx)
+bleu_metric: 79.99881739549791
+----------
+NL: match multiple regex patterns with the alternation operator `|` in a string `(zyx)bc`
+ref: re.findall('\\(.*?\\)|\\w', '(zyx)bc')
+Output: re.findall('(?<=[a-z])(?=[A-Z])', '(zy
+bleu_metric: 25.2840740779747
+----------
+NL: formate each string cin list `elements` into pattern '%{0}%'
+ref: elements = ['%{0}%'.format(element) for element in elements]
+Output: re.sub('[\\W_]+', '', '\\1', '\\1', '\
+bleu_metric: 5.517784889816492
+----------
+NL: Open a background process 'background-process' with arguments 'arguments'
+ref: subprocess.Popen(['background-process', 'arguments'])
+Output: subprocess.call(['background-process', 'arguments'])
+bleu_metric: 85.78928092681434
+----------
+NL: get list of values from dictionary 'mydict' w.r.t. list of keys 'mykeys'
+ref: [mydict[x] for x in mykeys]
+Output: [d['mykey'] for d in mydict]
+bleu_metric: 18.69300079996002
+----------
+NL: convert list `[('Name', 'Joe'), ('Age', 22)]` into a dictionary
+ref: dict([('Name', 'Joe'), ('Age', 22)])
+Output: ▁dict([('Name', 'Joe', 'Age', 22)])
+bleu_metric: 80.40514736345938
+----------
+NL: average each two columns of array `data`
+ref: data.reshape(-1, j).mean(axis=1).reshape(data.shape[0], -1)
+Output: np.mean(np.mean(a.shape[0],▁axis=1))
+bleu_metric: 31.214272516254898
+----------
+NL: double backslash escape all double quotes in string `s`
+ref: print(s.encode('unicode-escape').replace('"', '\\"'))
+Output: ▁print(s.replace('"', ''))
+bleu_metric: 42.801426144397794
+----------
+NL: split a string into a list of words and whitespace
+ref: re.split('(\\W+)', s)
+Output: [s.split() for s in re.findall('\\S+[^\\S+]+',
+bleu_metric: 12.913533075470381
+----------
+NL: plotting stacked barplots on a panda data frame
+ref: df.plot(kind='barh', stacked=True)
+Output: df.plot(kind='bar',▁stacked=True)
+bleu_metric: 66.75075987129311
+----------
+NL: reverse the keys and values in a dictionary `myDictionary`
+ref: {i[1]: i[0] for i in list(myDictionary.items())}
+Output: ▁dict((v, k) for k, v in myDict.items())
+bleu_metric: 21.05044094649095
+----------
+NL: finding the index of elements containing substring 'how' and 'what' in a list of strings 'myList'.
+ref: [i for i, j in enumerate(myList) if 'how' in j.lower() or 'what' in j.lower()]
+Output: [item for item in myList▁if 'how' in item]
+bleu_metric: 12.757789554746044
+----------
+NL: check if object `obj` is a string
+ref: isinstance(obj, str)
+Output: isinstance(obj, str)
+bleu_metric: 100.0
+----------
+NL: check if object `o` is a string
+ref: isinstance(o, str)
+Output: isinstance(o, str)
+bleu_metric: 100.0
+----------
+NL: check if object `o` is a string
+ref: (type(o) is str)
+Output: isinstance(o, str)
+bleu_metric: 26.78284959130086
+----------
+NL: check if object `o` is a string
+ref: isinstance(o, str)
+Output: isinstance(o, str)
+bleu_metric: 100.0
+----------
+NL: check if `obj_to_test` is a string
+ref: isinstance(obj_to_test, str)
+Output: isinstance(obj_to_test, str)
+bleu_metric: 100.0
+----------
+NL: append list `list1` to `list2`
+ref: list2.extend(list1)
+Output: ▁list2 = [[] for i in range(1)]
+bleu_metric: 9.672649511413093
+----------
+NL: append list `mylog` to `list1`
+ref: list1.extend(mylog)
+Output: ▁list1.append(mylog)
+bleu_metric: 45.18010018049224
+----------
+NL: append list `a` to `c`
+ref: c.extend(a)
+Output: a.append(c)
+bleu_metric: 30.21375397356768
+----------
+NL: append items in list `mylog` to `list1`
+ref: for line in mylog:
+ list1.append(line)
+Output: ▁list1.append({'mylog': [1, 2, 3, 4], 'list1': [1, 2, 3, 4]})
+bleu_metric: 11.110969989860408
+----------
+NL: append a tuple of elements from list `a` with indexes '[0][0] [0][2]' to list `b`
+ref: b.append((a[0][0], a[0][2]))
+Output: [(i, j) for i, j in zip(a, b)]
+bleu_metric: 9.575085238714179
+----------
+NL: Initialize `SECRET_KEY` in flask config with `Your_secret_string `
+ref: app.config['SECRET_KEY'] = 'Your_secret_string'
+Output: {'SECRET_KEY': 'Your_SECURITY_string'}
+bleu_metric: 20.33038389239877
+----------
+NL: unpack a series of tuples in pandas into a DataFrame with column names 'out-1' and 'out-2'
+ref: pd.DataFrame(out.tolist(), columns=['out-1', 'out-2'], index=out.index)
+Output: pd.concat([df1, df2],▁axis=1)
+bleu_metric: 5.527540935225947
+----------
+NL: find the index of an element 'MSFT' in a list `stocks_list`
+ref: [x for x in range(len(stocks_list)) if stocks_list[x] == 'MSFT']
+Output: from functools import▁reduce▁reduce(lambda s: s.index(s.encode('latin1'))▁if s.index
+bleu_metric: 7.022952541634667
+----------
+NL: rotate the xtick labels of matplotlib plot `ax` by `45` degrees to make long labels readable
+ref: ax.set_xticklabels(labels, rotation=45)
+Output: ax.set_xticklabels(ax.xaxis.get_majorticklabels(45))
+bleu_metric: 30.620490882364876
+----------
+NL: remove symbols from a string `s`
+ref: re.sub('[^\\w]', ' ', s)
+Output: re.sub('(?<=[a-z])(?=[A-Z])', '', s)
+bleu_metric: 35.3639760061202
+----------
+NL: Get the current directory of a script
+ref: os.path.basename(os.path.dirname(os.path.realpath(__file__)))
+Output: os.path.dirname(os.path.realpath(__file__))
+bleu_metric: 63.37967287681905
+----------
+NL: Find octal characters matches from a string `str` using regex
+ref: print(re.findall("'\\\\[0-7]{1,3}'", str))
+Output: re.findall(pattern,▁string).findall(pattern, re.UNICODE)
+bleu_metric: 13.635660702165275
+----------
+NL: split string `input` based on occurrences of regex pattern '[ ](?=[A-Z]+\\b)'
+ref: re.split('[ ](?=[A-Z]+\\b)', input)
+Output: input.split('[ ](?=[A-Z]+\b)')
+bleu_metric: 75.20601490106428
+----------
+NL: Split string `input` at every space followed by an upper-case letter
+ref: re.split('[ ](?=[A-Z])', input)
+Output: input.split('.')[1]
+bleu_metric: 16.636072310752336
+----------
+NL: send multipart encoded file `files` to url `url` with headers `headers` and metadata `data`
+ref: r = requests.post(url, files=files, headers=headers, data=data)
+Output: ▁urllib.request.urlopen(url,▁headers=headers,▁headers=data)
+bleu_metric: 33.09468095382839
+----------
+NL: write bytes `bytes_` to a file `filename` in python 3
+ref: open('filename', 'wb').write(bytes_)
+Output: open('filename', 'rb').write(bytes(bytes_))
+bleu_metric: 60.59827460443867
+----------
+NL: get a list from a list `lst` with values mapped into a dictionary `dct`
+ref: [dct[k] for k in lst]
+Output: ▁list(itertools.chain.from_iterable(lst))
+bleu_metric: 10.511846841633776
+----------
+NL: find duplicate names in column 'name' of the dataframe `x`
+ref: x.set_index('name').index.get_duplicates()
+Output: df.groupby('name')['x'].cumsum()
+bleu_metric: 34.11488281065382
+----------
+NL: truncate float 1.923328437452 to 3 decimal places
+ref: round(1.923328437452, 3)
+Output: ▁float('{0:.2f}'.format(env))
+bleu_metric: 9.629943614188138
+----------
+NL: sort list `li` in descending order based on the date value in second element of each list in list `li`
+ref: sorted(li, key=lambda x: datetime.strptime(x[1], '%d/%m/%Y'), reverse=True)
+Output: ▁sorted(li,▁key=lambda x: (x[1], x[0]))
+bleu_metric: 31.993666822129963
+----------
+NL: place the radial ticks in plot `ax` at 135 degrees
+ref: ax.set_rlabel_position(135)
+Output: ax.set_visible(False)
+bleu_metric: 32.17294420803808
+----------
+NL: check if path `my_path` is an absolute path
+ref: os.path.isabs(my_path)
+Output: ▁print(os.path.abspath(__file__))
+bleu_metric: 29.256127307315065
+----------
+NL: get number of keys in dictionary `yourdict`
+ref: len(list(yourdict.keys()))
+Output: len(yourdict.keys())
+bleu_metric: 63.70964381207871
+----------
+NL: count the number of keys in dictionary `yourdictfile`
+ref: len(set(open(yourdictfile).read().split()))
+Output: len(yourdictfile.values())
+bleu_metric: 13.938681520780069
+----------
+NL: pandas dataframe get first row of each group by 'id'
+ref: df.groupby('id').first()
+Output: df['id'] = df['id'].apply(lambda x: x.tolist())
+bleu_metric: 12.991916506579942
+----------
+NL: split a list in first column into multiple columns keeping other columns as well in pandas data frame
+ref: pd.concat([df[0].apply(pd.Series), df[1]], axis=1)
+Output: df['A'].str.split(',', 1)['B'].str.split(',')
+bleu_metric: 9.64535051675695
+----------
+NL: extract attributes 'src="js/([^"]*\\bjquery\\b[^"]*)"' from string `data`
+ref: re.findall('src="js/([^"]*\\bjquery\\b[^"]*)"', data)
+Output: re.findall('\\bjquery\\b[^"]*)', data)
+bleu_metric: 51.296752626995115
+----------
+NL: Sum integers contained in strings in list `['', '3.4', '', '', '1.0']`
+ref: sum(int(float(item)) for item in [_f for _f in ['', '3.4', '', '', '1.0'] if _f])
+Output: ▁sum(map(int, ['', '3.4', '', '1.0']))
+bleu_metric: 45.29937105346175
+----------
+NL: Call a subprocess with arguments `c:\\Program Files\\VMware\\VMware Server\\vmware-cmd.bat` that may contain spaces
+ref: subprocess.Popen(['c:\\Program Files\\VMware\\VMware Server\\vmware-cmd.bat'])
+Output: subprocess.call(['c:\\Programs\\VMware\\VMware',▁shell=T
+bleu_metric: 37.84850418718211
+----------
+NL: reverse a priority queue `q` in python without using classes
+ref: q.put((-n, n))
+Output: time.sleep(1)
+bleu_metric: 13.597602315271134
+----------
+NL: make a barplot of data in column `group` of dataframe `df` colour-coded according to list `color`
+ref: df['group'].plot(kind='bar', color=['r', 'g', 'b', 'r', 'g', 'b', 'r'])
+Output: plt.show()
+bleu_metric: 0.030120959708599737
+----------
+NL: find all matches of regex pattern '([a-fA-F\\d]{32})' in string `data`
+ref: re.findall('([a-fA-F\\d]{32})', data)
+Output: re.findall('([a-fA-F\\d]{32})', data)
+bleu_metric: 100.0
+----------
+NL: Get the length of list `my_list`
+ref: len(my_list)
+Output: len(my_list)
+bleu_metric: 100.0
+----------
+NL: Getting the length of array `l`
+ref: len(l)
+Output: len(my_array)
+bleu_metric: 45.180100180492246
+----------
+NL: Getting the length of array `s`
+ref: len(s)
+Output: len(s)
+bleu_metric: 100.0
+----------
+NL: Getting the length of `my_tuple`
+ref: len(my_tuple)
+Output: len(my_tuple)
+bleu_metric: 100.0
+----------
+NL: Getting the length of `my_string`
+ref: len(my_string)
+Output: len(my_string)
+bleu_metric: 100.0
+----------
+NL: remove escape character from string "\\a"
+ref: """\\a""".decode('string_escape')
+Output: """\\a""".decode('string_escape')
+bleu_metric: 100.0
+----------
+NL: replace each 'a' with 'b' and each 'b' with 'a' in the string 'obama' in a single pass.
+ref: """obama""".replace('a', '%temp%').replace('b', 'a').replace('%temp%', 'b')
+Output: """""".join(['a', 'b', 'a'])
+bleu_metric: 28.029141246256657
+----------
+NL: remove directory tree '/folder_name'
+ref: shutil.rmtree('/folder_name')
+Output: os.path.dirname('/folder_name')
+bleu_metric: 57.608442016038964
+----------
+NL: create a new column `weekday` in pandas data frame `data` based on the values in column `my_dt`
+ref: data['weekday'] = data['my_dt'].apply(lambda x: x.weekday())
+Output: data.apply(lambda x: x['weekday'].str.contains(x['my_dt'
+bleu_metric: 51.68669917653163
+----------
+NL: reverse sort Counter `x` by values
+ref: sorted(x, key=x.get, reverse=True)
+Output: ▁sorted(x,▁key=lambda x: x[::-1])
+bleu_metric: 21.27988282044161
+----------
+NL: reverse sort counter `x` by value
+ref: sorted(list(x.items()), key=lambda pair: pair[1], reverse=True)
+Output: ▁sorted(x,▁key=itemgetter(1), reverse=True)
+bleu_metric: 28.177860103739523
+----------
+NL: append a numpy array 'b' to a numpy array 'a'
+ref: np.vstack((a, b))
+Output: ▁numpy.append(a, b)
+bleu_metric: 46.92470064105599
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: print(concatenate((a, b), axis=0))
+Output: np.concatenate((a, b),▁axis=0)
+bleu_metric: 65.60930058975354
+----------
+NL: numpy concatenate two arrays `a` and `b` along the second axis
+ref: print(concatenate((a, b), axis=1))
+Output: np.concatenate((a, b),▁axis=0)
+bleu_metric: 55.60336961016132
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: c = np.r_[(a[None, :], b[None, :])]
+Output: np.concatenate((a, b),▁axis=0)
+bleu_metric: 10.756284573187752
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: np.array((a, b))
+Output: np.concatenate((a, b),▁axis=0)
+bleu_metric: 41.896825433302276
+----------
+NL: fetch address information for host 'google.com' ion port 80
+ref: print(socket.getaddrinfo('google.com', 80))
+Output: db.request.urlretrieve('google.com')
+bleu_metric: 39.00756812161481
+----------
+NL: add a column 'day' with value 'sat' to dataframe `df`
+ref: df.xs('sat', level='day', drop_level=False)
+Output: df['Date'] = df['Date'].apply(lambda x: x.to_date(
+bleu_metric: 7.575731225158963
+----------
+NL: return a 401 unauthorized in django
+ref: return HttpResponse('Unauthorized', status=401)
+Output: MyModel.objects.filter(author__id=1)
+bleu_metric: 14.294845713017917
+----------
+NL: Flask set folder 'wherever' as the default template folder
+ref: Flask(__name__, template_folder='wherever')
+Output: ▁app.run(debug=True)
+bleu_metric: 17.03318603763928
+----------
+NL: How do I INSERT INTO t1 (SELECT * FROM t2) in SQLAlchemy?
+ref: session.execute('INSERT INTO t1 (SELECT * FROM t2)')
+Output: ▁session.execute('INSERT▁INTO t1 (SELECT *▁FROM t2)')
+bleu_metric: 63.83964846132485
+----------
+NL: sort a list of lists 'c2' such that third row comes first
+ref: c2.sort(key=lambda row: row[2])
+Output: c2.sort(key=lambda x: x[0])
+bleu_metric: 55.33409598501608
+----------
+NL: Sorting a list of lists in Python
+ref: c2.sort(key=lambda row: (row[2], row[1], row[0]))
+Output: ▁sorted(a,▁key=lambda x: x.index(x))
+bleu_metric: 13.258555426461873
+----------
+NL: Sorting a list of lists in Python
+ref: c2.sort(key=lambda row: (row[2], row[1]))
+Output: ▁sorted(a,▁key=lambda x: x.index(x))
+bleu_metric: 17.024322156140222
+----------
+NL: set font `Arial` to display non-ascii characters in matplotlib
+ref: matplotlib.rc('font', **{'sans-serif': 'Arial', 'family': 'sans-serif'})
+Output: plt.show()
+bleu_metric: 0.3669484097406089
+----------
+NL: Convert DateTime column 'date' of pandas dataframe 'df' to ordinal
+ref: df['date'].apply(lambda x: x.toordinal())
+Output: df['date'] = df['date'].dt.to_datetime(df['date']
+bleu_metric: 28.326302347195153
+----------
+NL: get html source of Selenium WebElement `element`
+ref: element.get_attribute('innerHTML')
+Output: ▁driver.find_element_by_xpath("//a[@id='a']//a[@id
+bleu_metric: 7.9290265068413746
+----------
+NL: Get the integer location of a key `bob` in a pandas data frame
+ref: df.index.get_loc('bob')
+Output: df.loc[df['bob'].index]
+bleu_metric: 22.894156860669913
+----------
+NL: open a 'gnome' terminal from python script and run 'sudo apt-get update' command.
+ref: os.system('gnome-terminal -e \'bash -c "sudo apt-get update; exec bash"\'')
+Output: os.system('sudo apt-get update')
+bleu_metric: 21.20024487612511
+----------
+NL: add an item with key 'third_key' and value 1 to an dictionary `my_dict`
+ref: my_dict.update({'third_key': 1})
+Output: ▁dict((k, v) for k, v in my_dict.items()▁if v[0] == 1)
+bleu_metric: 7.261813302549417
+----------
+NL: declare an array
+ref: my_list = []
+Output: intarray = []
+bleu_metric: 75.98356856515926
+----------
+NL: Insert item `12` to a list `my_list`
+ref: my_list.append(12)
+Output: my_list.insert(0, 12)
+bleu_metric: 32.46679154750989
+----------
+NL: add an entry 'wuggah' at the beginning of list `myList`
+ref: myList.insert(0, 'wuggah')
+Output: ['wuggah', 'wuggah', 'wuggah']
+bleu_metric: 23.961829057131983
+----------
+NL: convert a hex-string representation to actual bytes
+ref: """\\xF3\\xBE\\x80\\x80""".replace('\\x', '').decode('hex')
+Output: ▁struct.unpack('I', 16))[0]
+bleu_metric: 6.258020733367848
+----------
+NL: select the last column of dataframe `df`
+ref: df[df.columns[-1]]
+Output: df.iloc[:, (-1)]
+bleu_metric: 20.78060434846712
+----------
+NL: get the first value from dataframe `df` where column 'Letters' is equal to 'C'
+ref: df.loc[df['Letters'] == 'C', 'Letters'].values[0]
+Output: df['Letters'].str.contains('C')
+bleu_metric: 27.69697733631702
+----------
+NL: converting two lists `[1, 2, 3]` and `[4, 5, 6]` into a matrix
+ref: np.column_stack(([1, 2, 3], [4, 5, 6]))
+Output: ▁matrix([[1, 2, 3], [4, 5, 6]))
+bleu_metric: 79.78897056262763
+----------
+NL: get the type of `i`
+ref: type(i)
+Output: type(i)
+bleu_metric: 100.0
+----------
+NL: determine the type of variable `v`
+ref: type(v)
+Output: type(v)
+bleu_metric: 100.0
+----------
+NL: determine the type of variable `v`
+ref: type(v)
+Output: type(v)
+bleu_metric: 100.0
+----------
+NL: determine the type of variable `v`
+ref: type(v)
+Output: type(v)
+bleu_metric: 100.0
+----------
+NL: determine the type of variable `v`
+ref: type(v)
+Output: type(v)
+bleu_metric: 100.0
+----------
+NL: get the type of variable `variable_name`
+ref: print(type(variable_name))
+Output: ▁globals()['factor']
+bleu_metric: 13.54599427337814
+----------
+NL: get the 5th item of a generator
+ref: next(itertools.islice(range(10), 5, 5 + 1))
+Output: ▁list(itertools.islice(it, 5))
+bleu_metric: 31.347458595184502
+----------
+NL: Print a string `word` with string format
+ref: print('"{}"'.format(word))
+Output: ▁print(word.replace(',', ''))
+bleu_metric: 22.31180367999676
+----------
+NL: join a list of strings `list` using a space ' '
+ref: """ """.join(list)
+Output: """ """.join(map(str,▁list))
+bleu_metric: 56.79161104357995
+----------
+NL: create list `y` containing two empty lists
+ref: y = [[] for n in range(2)]
+Output: y = [[] for i in range(2)]
+bleu_metric: 80.03203203844998
+----------
+NL: read a file 'C:/name/MyDocuments/numbers' into a list `data`
+ref: data = [line.strip() for line in open('C:/name/MyDocuments/numbers', 'r')]
+Output: data = [line.read() for▁line in open('C:/name/MyDocuments/numbers')]
+bleu_metric: 69.7712347922743
+----------
+NL: delete all occurrences of character 'i' in string 'it is icy'
+ref: """""".join([char for char in 'it is icy' if char != 'i'])
+Output: re.sub('i', 'it is icy')
+bleu_metric: 15.99248714366856
+----------
+NL: delete all instances of a character 'i' in a string 'it is icy'
+ref: re.sub('i', '', 'it is icy')
+Output: re.sub('i', '', 'it is icy')
+bleu_metric: 100.0
+----------
+NL: delete all characters "i" in string "it is icy"
+ref: """it is icy""".replace('i', '')
+Output: """it is icy""".replace('i', '')
+bleu_metric: 100.0
+----------
+NL: How to delete all instances of a character in a string in python?
+ref: """""".join([char for char in 'it is icy' if char != 'i'])
+Output: re.sub('[^a-zA-Z]*', '', text)
+bleu_metric: 7.866954533441649
+----------
+NL: Drop rows of pandas dataframe `df` having NaN in column at index "1"
+ref: df.dropna(subset=[1])
+Output: df.dropna(df.loc[:, (None)],▁axis=1)
+bleu_metric: 20.82198320914845
+----------
+NL: get elements from list `myList`, that have a field `n` value 30
+ref: [x for x in myList if x.n == 30]
+Output: [i for i in myList▁if n == 30]
+bleu_metric: 45.06143230242354
+----------
+NL: converting list of strings `intstringlist` to list of integer `nums`
+ref: nums = [int(x) for x in intstringlist]
+Output: [int(x) for x in intstringlist]
+bleu_metric: 83.9234989944319
+----------
+NL: convert list of string numbers into list of integers
+ref: map(int, eval(input('Enter the unfriendly numbers: ')))
+Output: [int(x) for x in lst]
+bleu_metric: 7.290792403968443
+----------
+NL: print "." without newline
+ref: sys.stdout.write('.')
+Output: ▁sys.stdout.write('.')
+bleu_metric: 83.85766789076257
+----------
+NL: round off the float that is the product of `2.52 * 100` and convert it to an int
+ref: int(round(2.51 * 100))
+Output: int(round(1.0 *▁100))
+bleu_metric: 46.82568791024402
+----------
+NL: Find all files in directory "/mydir" with extension ".txt"
+ref: os.chdir('/mydir')
+for file in glob.glob('*.txt'):
+ pass
+Output: for▁file in os.listdir('/mydir'):▁if▁file.endswith('.txt'):▁pass
+bleu_metric: 37.56995795800578
+----------
+NL: Find all files in directory "/mydir" with extension ".txt"
+ref: for file in os.listdir('/mydir'):
+ if file.endswith('.txt'):
+ pass
+Output: for▁file in os.listdir('/mydir'):▁if▁file.endswith('.txt'):▁pass
+bleu_metric: 71.95956498085113
+----------
+NL: Find all files in directory "/mydir" with extension ".txt"
+ref: for (root, dirs, files) in os.walk('/mydir'):
+ for file in files:
+ if file.endswith('.txt'):
+ pass
+Output: for▁file in os.listdir('/mydir'):▁if▁file.endswith('.txt'):▁pass
+bleu_metric: 47.1684455439837
+----------
+NL: plot dataframe `df` without a legend
+ref: df.plot(legend=False)
+Output: plt.show()
+bleu_metric: 20.034704329441453
+----------
+NL: loop through the IP address range "192.168.x.x"
+ref: for i in range(256):
+ for j in range(256):
+ ip = ('192.168.%d.%d' % (i, j))
+ print(ip)
+Output: for i in range('192.168.x.x'):▁pass
+bleu_metric: 14.238794877552813
+----------
+NL: loop through the IP address range "192.168.x.x"
+ref: for (i, j) in product(list(range(256)), list(range(256))):
+ pass
+Output: for i in range('192.168.x.x'):▁pass
+bleu_metric: 7.916713848850043
+----------
+NL: loop through the IP address range "192.168.x.x"
+ref: generator = iter_iprange('192.168.1.1', '192.168.255.255', step=1)
+Output: for i in range('192.168.x.x'):▁pass
+bleu_metric: 18.291291695991816
+----------
+NL: Sum the corresponding decimal values for binary values of each boolean element in list `x`
+ref: sum(1 << i for i, b in enumerate(x) if b)
+Output: ▁sum([int(x) for x in▁list(bool(x))])
+bleu_metric: 14.071964689317513
+----------
+NL: write multiple strings `line1`, `line2` and `line3` in one line in a file `target`
+ref: target.write('%r\n%r\n%r\n' % (line1, line2, line3))
+Output: f.write(line1,▁line2 +▁line3.split(','))
+bleu_metric: 16.116608915070366
+----------
+NL: Convert list of lists `data` into a flat list
+ref: [y for x in data for y in (x if isinstance(x, list) else [x])]
+Output: [j for i in data for j in i]
+bleu_metric: 9.344244386017191
+----------
+NL: Print new line character as `\n` in a string `foo\nbar`
+ref: print('foo\nbar'.encode('string_escape'))
+Output: """foo\nbar""".replace('\n')
+bleu_metric: 32.49189152106538
+----------
+NL: remove last comma character ',' in string `s`
+ref: """""".join(s.rsplit(',', 1))
+Output: s = s.replace(',', '')
+bleu_metric: 24.4545047029932
+----------
+NL: calculate the mean of each element in array `x` with the element previous to it
+ref: (x[1:] + x[:-1]) / 2
+Output: np.mean(x)
+bleu_metric: 6.656592803413297
+----------
+NL: get an array of the mean of each two consecutive values in numpy array `x`
+ref: x[:-1] + (x[1:] - x[:-1]) / 2
+Output: ▁numpy.array([[1, 2], [2, 4]])
+bleu_metric: 10.64536882179483
+----------
+NL: load data containing `utf-8` from file `new.txt` into numpy array `arr`
+ref: arr = numpy.fromiter(codecs.open('new.txt', encoding='utf-8'), dtype='