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: print ( proc. get _ append ( ) )
+bleu_metric: 14.253628894439752
+----------
+NL: decode a hex string '4a4b4c' to utf-8.
+ref: bytes.fromhex('4a4b4c').decode('utf-8')
+Output: print ( str. decode ('utf8') )
+bleu_metric: 23.083362053422253
+----------
+NL: check if all elements in list `mylist` are identical
+ref: all(x == myList[0] for x in myList)
+Output: len ( set ( mylist ) ) = = 1
+bleu_metric: 9.360791398475978
+----------
+NL: format number of spaces between strings `python`, `:` and `very good` to be `20`
+ref: print('%*s : %*s' % (20, 'Python', 20, 'Very Good'))
+Output: " " " { 0 :, } " " ". format ( 123456789, 16 ) ( lambda x : x.
+bleu_metric: 7.030766106716325
+----------
+NL: how to convert a string from cp-1251 to utf-8?
+ref: d.decode('cp1251').encode('utf8')
+Output: d. decode ('cp1252'). encode ('utf8')
+bleu_metric: 82.82477531331041
+----------
+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: k for k, v in list ( dict. items ( ) ) if v is not none ]
+bleu_metric: 54.70605317141546
+----------
+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: k for k, v in list ( dict. items ( ) ) if v is not none ]
+bleu_metric: 40.55760587628575
+----------
+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: subprocess. call ( ['mystring. py ','mystring. py'] )
+bleu_metric: 11.327304007996714
+----------
+NL: concatenate a list of strings `['a', 'b', 'c']`
+ref: """""".join(['a', 'b', 'c'])
+Output: ['a ','b ','c ','d']
+bleu_metric: 50.018452828984486
+----------
+NL: find intersection data between series `s1` and series `s2`
+ref: pd.Series(list(set(s1).intersection(set(s2))))
+Output: s1. intersection ( s1, how ='left'). reset _ index ( )
+bleu_metric: 16.132580731955684
+----------
+NL: sending http headers to `client`
+ref: client.send('HTTP/1.0 200 OK\r\n')
+Output: server. serve _ forever ( )
+bleu_metric: 6.063168540538235
+----------
+NL: format a datetime string `when` to extract date only
+ref: then = datetime.datetime.strptime(when, '%Y-%m-%d').date()
+Output: datetime. datetime. strptime ('2012 - 05 - 29t19 : 30 : 03. 283z ','% y
+bleu_metric: 28.12321366714592
+----------
+NL: split a multi-line string `inputstring` into separate strings
+ref: inputString.split('\n')
+Output: from functools import reduce reduce ( lambda x, y : x + y, l. split ('' ), x. split (
+bleu_metric: 14.32450175903488
+----------
+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: re. split ('\ n \ r \ n ','a \ n b \ n')
+bleu_metric: 65.72891674091409
+----------
+NL: concatenate elements of list `b` by a colon ":"
+ref: """:""".join(str(x) for x in b)
+Output: ['a ','b']
+bleu_metric: 6.490157276489835
+----------
+NL: get the first object from a queryset in django model `entry`
+ref: Entry.objects.filter()[:1].get()
+Output: model. objects. all ( ). order _ by ('- id') [ 0 ]
+bleu_metric: 16.1692143534558
+----------
+NL: calculate sum over all rows of 2d numpy array
+ref: a.sum(axis=1)
+Output: a. sum ( axis = 1 )
+bleu_metric: 100.0
+----------
+NL: enable warnings using action 'always'
+ref: warnings.simplefilter('always')
+Output: warnings. simplefilter ('always')
+bleu_metric: 100.0
+----------
+NL: concatenate items of list `l` with a space ' '
+ref: print(' '.join(map(str, l)))
+Output: print (''. join ( [ str ( x ) for x in l ] ) )
+bleu_metric: 42.5874468039002
+----------
+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 ( ['python ','hello world ','world'] )
+bleu_metric: 48.69018522875219
+----------
+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: time. strptime ('2015 / 01 / 01 / 03. 0. 0. 0 ','% d / % m / %
+bleu_metric: 31.63951380366603
+----------
+NL: convert a string `my_string` with dot and comma into a float number `my_float`
+ref: my_float = float(my_string.replace(',', ''))
+Output: my _ string. replace ( ', ','' )
+bleu_metric: 59.23488237152091
+----------
+NL: convert a string `123,456.908` with dot and comma into a floating number
+ref: float('123,456.908'.replace(',', ''))
+Output: " " " { 0 :,. 2f } " " ". format ( 12345678. 456789, 16
+bleu_metric: 11.63901425102663
+----------
+NL: set pythonpath in python script.
+ref: sys.path.append('/path/to/whatever')
+Output: os. path. join ( sys. path. append ('/ path / to / my / root') )
+bleu_metric: 51.191407111567585
+----------
+NL: split string 'words, words, words.' using a regex '(\\w+)'
+ref: re.split('(\\W+)', 'Words, words, words.')
+Output: re. split ('\ \ w + ','\ \ 1 ', words, words )
+bleu_metric: 39.78163327099985
+----------
+NL: open a file `output.txt` in append mode
+ref: file = open('Output.txt', 'a')
+Output: file. write ('whatever')
+bleu_metric: 17.234857810563913
+----------
+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. google. com / file. gz ', '
+bleu_metric: 49.30073739073773
+----------
+NL: download a file `url` over http and save to `file_name`
+ref: u = urllib.request.urlopen(url)
+Output: response. close ( )
+bleu_metric: 14.35549295013305
+----------
+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. google. com / ','http : / /
+bleu_metric: 42.4884196166917
+----------
+NL: download a file `url` over http
+ref: r = requests.get(url)
+Output: response = requests. get ( url )
+bleu_metric: 88.01117367933934
+----------
+NL: download a file `url` over http and save to "10mb"
+ref: response = requests.get(url, stream=True)
+Output: urllib. request. urlretrieve ('http : / / example. com / file. ext ','wb '
+bleu_metric: 6.370405230161802
+----------
+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 ('- - type ', nargs ='*')
+bleu_metric: 17.102327939516414
+----------
+NL: remove key 'c' from dictionary `d`
+ref: {i: d[i] for i in d if i != 'c'}
+Output: d = { k : v for k, v in list ( d. items ( ) ) if v }
+bleu_metric: 9.637411586929977
+----------
+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. drop _ duplicates ( df. columns ['column _ name']. apply ( lambda x :'column _ name '
+bleu_metric: 8.416754937260349
+----------
+NL: split a string `s` by space with `4` splits
+ref: s.split(' ', 4)
+Output: ['x ','y ','x ','z ','x ','d ','x ', '
+bleu_metric: 5.648892151960283
+----------
+NL: read keyboard-input
+ref: input('Enter your input:')
+Output: input ('press enter to continue :')
+bleu_metric: 42.40125351805037
+----------
+NL: enable debug mode on flask application `app`
+ref: app.run(debug=True)
+Output: app. run ( debug = true )
+bleu_metric: 70.71067811865474
+----------
+NL: python save list `mylist` to file object 'save.txt'
+ref: pickle.dump(mylist, open('save.txt', 'wb'))
+Output: savefig. save ('filename. txt')
+bleu_metric: 13.988120102483691
+----------
+NL: multiply a matrix `p` with a 3d tensor `t` in scipy
+ref: scipy.tensordot(P, T, axes=[1, 1]).swapaxes(0, 1)
+Output: scipy. matrix ( [ 1, 2, 3 ], [ 1, 2, 3 ] ). t
+bleu_metric: 16.944114083600446
+----------
+NL: create 3d array of zeroes of size `(3,3,3)`
+ref: numpy.zeros((3, 3, 3))
+Output: np. zeros ( ( 3, 3 ) ). transpose ( 2, 3 )
+bleu_metric: 54.166685749344154
+----------
+NL: cut off the last word of a sentence `content`
+ref: """ """.join(content.split(' ')[:-1])
+Output: " " " " " ". join ( content. split ('' ) [ : - 1 ] )
+bleu_metric: 100.0
+----------
+NL: convert scalar `x` to array
+ref: x = np.asarray(x).reshape(1, -1)[(0), :]
+Output: x = np. zeros ( x ). reshape ( - 1, 2 )
+bleu_metric: 35.6520799192373
+----------
+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 in l ) for i in l ]
+bleu_metric: 19.29709182756385
+----------
+NL: convert hex string '470fc614' to a float number
+ref: struct.unpack('!f', '470FC614'.decode('hex'))[0]
+Output: struct. unpack ('d ', binascii. unhexlify ('4081637ef7d
+bleu_metric: 20.061882169394003
+----------
+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 list ( my _ dict. items ( ) ) )
+bleu_metric: 35.25419189217226
+----------
+NL: running bash script 'sleep.sh'
+ref: subprocess.call('sleep.sh', shell=True)
+Output: os. system ('echo $ command - flags arguments')
+bleu_metric: 14.814869013357976
+----------
+NL: join elements of list `l` with a comma `,`
+ref: """,""".join(l)
+Output: " " ", " " ". join ( l )
+bleu_metric: 100.0
+----------
+NL: make a comma-separated string from a list `mylist`
+ref: myList = ','.join(map(str, myList))
+Output: mylist = [ i. split ( ',') for i in mylist ]
+bleu_metric: 15.526320210133507
+----------
+NL: reverse the list that contains 1 to 10
+ref: list(reversed(list(range(10))))
+Output: [ ( 1, 2 ), ( 3, 4 ), ( 2, - 1 ), ( 1, 2 ), ( 3, 4
+bleu_metric: 5.834347180338517
+----------
+NL: remove substring 'bag,' from a string 'lamp, bag, mirror'
+ref: print('lamp, bag, mirror'.replace('bag,', ''))
+Output: re. sub ( ', ', ', '. join ( chr ) for ch in asking )
+bleu_metric: 22.93476949918813
+----------
+NL: reverse the order of words, delimited by `.`, in string `s`
+ref: """.""".join(s.split('.')[::-1])
+Output: re. sub ('\ \ b (. + ) (? = \ \ s + \ \ s + ) ','\ \ 1
+bleu_metric: 7.508325052934217
+----------
+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: datetime. datetime. now ( ). strftime ('% y - % m - % d % h : % m : %
+bleu_metric: 42.22477323756999
+----------
+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. now ( ). strftime ('% y - % m - % d % h : % m : %
+bleu_metric: 32.53778694644052
+----------
+NL: get the date 7 days before the current date
+ref: (datetime.datetime.now() - datetime.timedelta(days=7)).date()
+Output: ( datetime. datetime. now ( ) - datetime. timedelta ( days = 1 ) )
+bleu_metric: 69.7353047868514
+----------
+NL: sum elements at index `column` of each list in list `data`
+ref: print(sum(row[column] for row in data))
+Output: [ sum ( item ) for item in zip ( * items ) ]
+bleu_metric: 14.009047908905242
+----------
+NL: sum columns of a list `array`
+ref: [sum(row[i] for row in array) for i in range(len(array[0]))]
+Output: sum ( i )
+bleu_metric: 0.4295018804111146
+----------
+NL: encode binary string 'your string' to base64 code
+ref: base64.b64encode(bytes('your string', 'utf-8'))
+Output: base64. b64encode ('base64')
+bleu_metric: 17.22724743406155
+----------
+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: 32.12854967972961
+----------
+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 d in list ( d. items ( ) ) for k, v in list ( d. items (
+bleu_metric: 10.14524997386625
+----------
+NL: how do i get the url parameter in a flask view
+ref: request.args['myParam']
+Output: request. args ['myapp']
+bleu_metric: 58.8837385313777
+----------
+NL: identify duplicate values in list `mylist`
+ref: [k for k, v in list(Counter(mylist).items()) if v > 1]
+Output: [ k for k, v in mylist if v. startswith ('mycount') else v for k, v in mylist
+bleu_metric: 34.06227682100307
+----------
+NL: insert directory 'apps' into directory `__file__`
+ref: sys.path.insert(1, os.path.join(os.path.dirname(__file__), 'apps'))
+Output: urllib. request. urlopen ('http : / / www. example. com ','/ path')
+bleu_metric: 9.10451572772839
+----------
+NL: modify sys.path for python module `subdir`
+ref: sys.path.append(os.path.join(os.path.dirname(__file__), 'subdir'))
+Output: subprocess. call ( ['start / subdir / bin / 2014 ','inr'] )
+bleu_metric: 8.368553758184737
+----------
+NL: insert a 'none' value into a sqlite3 table.
+ref: db.execute("INSERT INTO present VALUES('test2', ?, 10)", (None,))
+Output: cursor. execute ('select * from table where column in l')
+bleu_metric: 13.62767317410906
+----------
+NL: flatten list `list_of_menuitems`
+ref: [image for menuitem in list_of_menuitems for image in menuitem]
+Output: [ item for sublist in list _ of _ lists ]
+bleu_metric: 17.194656088289214
+----------
+NL: append elements of a set `b` to a list `a`
+ref: a.extend(b)
+Output: a. extend ( b )
+bleu_metric: 100.0
+----------
+NL: append elements of a set to a list in python
+ref: a.extend(list(b))
+Output: a. extend ( b )
+bleu_metric: 46.81294641285289
+----------
+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 _ csv ('filename. txt ', header = false )
+bleu_metric: 14.209004647554735
+----------
+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 \ \ t. txt ', sep ='\ t ', index
+bleu_metric: 37.50344293811855
+----------
+NL: split a string `x` by last occurrence of character `-`
+ref: print(x.rpartition('-')[0])
+Output: x. split ('-') [ : - 1 ]
+bleu_metric: 43.353400970949984
+----------
+NL: get the last part of a string before the character '-'
+ref: print(x.rsplit('-', 1)[0])
+Output: re. sub ('(? < = - -. ) ','- ', line )
+bleu_metric: 20.776569671187676
+----------
+NL: upload file using ftp
+ref: ftp.storlines('STOR ' + filename, open(filename, 'r'))
+Output: ftp. quit ( )
+bleu_metric: 3.710176016036221
+----------
+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 ('window. scrollto ( 0, document. body. com') )
+bleu_metric: 10.807499589580493
+----------
+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. array ( [ [ 1, 2, 3 ], [ 4, 5, 6 ] ] )
+bleu_metric: 28.489318277723964
+----------
+NL: print a list `l` and move first 3 elements to the end of the list
+ref: print(l[3:] + l[:3])
+Output: print ( [ l [ 3 : ] for l in l ] )
+bleu_metric: 43.472087194499146
+----------
+NL: loop over files in directory '.'
+ref: for fn in os.listdir('.'):
+Output: for ( root, dirs, dirs ) in os. walk ( '.') : if os. path. isdir ( '
+bleu_metric: 26.167475767017738
+----------
+NL: loop over files in directory `source`
+ref: for (root, dirs, filenames) in os.walk(source):
+Output: os. listdir ('os')
+bleu_metric: 9.081302407530908
+----------
+NL: create a random list of integers
+ref: [int(1000 * random.random()) for i in range(10000)]
+Output: random. choice ( [ int ( i ) for i in range ( 10 ) ] )
+bleu_metric: 38.95592996402149
+----------
+NL: using %f with strftime() in python to get microseconds
+ref: datetime.datetime.now().strftime('%H:%M:%S.%f')
+Output: datetime. datetime. now ( ). strftime ('% h : % m : % s. % f')
+bleu_metric: 69.80972118841675
+----------
+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: c. objects. filter ( id = 1 ). order _ by ('foobar')
+bleu_metric: 10.30143863225941
+----------
+NL: filter rows in pandas starting with alphabet 'f' using regular expression.
+ref: df.b.str.contains('^f')
+Output: df ['a']. str. contains ('c')
+bleu_metric: 43.85068972747104
+----------
+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: print ('\ n '. join ( map ( str, row ) ) )
+bleu_metric: 28.575895973659073
+----------
+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. drop ( df. index. map ( lambda x : x. index, axis = 1 ) )
+bleu_metric: 13.21619418680914
+----------
+NL: format the variables `self.goals` and `self.penalties` using string formatting
+ref: """({:d} goals, ${:d})""".format(self.goals, self.penalties)
+Output: " " " { } \ \ w { }. { } " " ". format ( date,'user')
+bleu_metric: 28.34142738530324
+----------
+NL: format string "({} goals, ${})" with variables `goals` and `penalties`
+ref: """({} goals, ${})""".format(self.goals, self.penalties)
+Output: " " " { } { } " " ". format ( {'foo':'bar'} )
+bleu_metric: 34.620457601133026
+----------
+NL: format string "({0.goals} goals, ${0.penalties})"
+ref: """({0.goals} goals, ${0.penalties})""".format(self)
+Output: df. set _ index ('name')
+bleu_metric: 3.9504404329436813
+----------
+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: 18.761335250609946
+----------
+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: map ( int, l )
+bleu_metric: 2.5915324112223583
+----------
+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: 11.473780780974305
+----------
+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 '. join ( map ( str, mylist ) ) )
+bleu_metric: 60.5478327168454
+----------
+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 [ x if x! ='b'] if x! ='c']
+bleu_metric: 20.792998340455547
+----------
+NL: remove duplicate words from a string `text` using regex
+ref: text = re.sub('\\b(\\w+)( \\1\\b)+', '\\1', text)
+Output: re. sub ('(. * ) \ \ 1 + ','\ \ 1 ', text )
+bleu_metric: 42.691904454956294
+----------
+NL: count non zero values in each column in pandas data frame
+ref: df.astype(bool).sum(axis=1)
+Output: df. groupby ( level = 0 ). sum ( )
+bleu_metric: 33.17838167503153
+----------
+NL: search for string that matches regular expression pattern '(?.*<' in xml string `line`
+ref: print(re.search('>.*<', line).group(0))
+Output: print ( doc. group ( 1 ). group ( 1 ) )
+bleu_metric: 22.08517793512327
+----------
+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 ('2010 - 11 - 13 10 : 33 : 54. 227806 ','% y
+bleu_metric: 23.69248624414322
+----------
+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: [ x [ 0 ] for x in l if x [ 0 ] = ='score']
+bleu_metric: 32.08521724608534
+----------
+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 ('[ ^ \ \ w. ] + ','', text ). strip ( )
+bleu_metric: 54.08690470924506
+----------
+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 ] + ','', s ). strip ( )
+bleu_metric: 38.80387619814469
+----------
+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. plot ( x, y, label ='h \ u2082o')
+bleu_metric: 72.76817202342092
+----------
+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. plot ( x, y, label ='h \ u2082o')
+bleu_metric: 71.60350546947922
+----------
+NL: loop over a list `mylist` if sublists length equals 3
+ref: [x for x in mylist if len(x) == 3]
+Output: mylist [ : : 2 ] = = mylist [ 0 ]
+bleu_metric: 12.640504935009755
+----------
+NL: initialize a list `lst` of 100 objects object()
+ref: lst = [Object() for _ in range(100)]
+Output: lst = [ [ ] for i in range ( 3 ) ]
+bleu_metric: 28.992166587078206
+----------
+NL: create list `lst` containing 100 instances of object `object`
+ref: lst = [Object() for i in range(100)]
+Output: list ( chain. from _ iterable ( l ) )
+bleu_metric: 10.649638629044373
+----------
+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: soup. find _ elements _ by _ xpath ( " / / * [ @ type ='submit'] " ). get _ attribute (
+bleu_metric: 10.389889377420038
+----------
+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, on = ['date _ time ','date _ time'] )
+bleu_metric: 21.22363344155404
+----------
+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 ( str ( a ) +'' + str ( b ) )
+bleu_metric: 6.053698751871955
+----------
+NL: split a string by a delimiter in python
+ref: [x.strip() for x in '2.MATCHES $$TEXT$$ STRING'.split('$$TEXT$$')]
+Output: s. split (';')
+bleu_metric: 2.904501569386027
+----------
+NL: check if directory `directory ` exists and create it if necessary
+ref: if (not os.path.exists(directory)):
+ os.makedirs(directory)
+Output: os. makedirs ( path )
+bleu_metric: 9.597952472216766
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: distutils.dir_util.mkpath(path)
+Output: os. makedirs ( path )
+bleu_metric: 40.661103887968814
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: distutils.dir_util.mkpath(path)
+Output: os. makedirs ( path )
+bleu_metric: 40.661103887968814
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: os.makedirs(path)
+Output: os. makedirs ( path )
+bleu_metric: 100.0
+----------
+NL: replace a separate word 'h3' by 'h1' in a string 'text'
+ref: re.sub('\\bH3\\b', 'H1', text)
+Output: re. sub ('\ \ w + ','\ \ 1 ', text )
+bleu_metric: 56.31071891586576
+----------
+NL: substitute ascii letters in string 'aas30dsa20' with empty string ''
+ref: re.sub('\\D', '', 'aas30dsa20')
+Output: re. sub ('[ a - z ] ','','abcd')
+bleu_metric: 51.003234295212685
+----------
+NL: get digits only from a string `aas30dsa20` using lambda function
+ref: """""".join([x for x in 'aas30dsa20' if x.isdigit()])
+Output: " " " " " ". join ( map ( lambda x : x * 7 ) )
+bleu_metric: 39.54603942707659
+----------
+NL: access a tag called "name" in beautifulsoup `soup`
+ref: print(soup.find('name').string)
+Output: soup. findall ('div ', {'name':'description'} )
+bleu_metric: 17.457404828304572
+----------
+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 ( ( x [ 0 ], x [ 1 ] ) for x in cursor. items ( ) )
+bleu_metric: 16.34453553468014
+----------
+NL: create new matrix object by concatenating data from matrix a and matrix b
+ref: np.concatenate((A, B))
+Output: numpy. concatenate ( [ a, b ], axis = 0 )
+bleu_metric: 18.92240568795935
+----------
+NL: concat two matrices `a` and `b` in numpy
+ref: np.vstack((A, B))
+Output: np. sqrt ( ( a. shape [ 0 ], b ) )
+bleu_metric: 19.923405658137924
+----------
+NL: get the characters count in a file `filepath`
+ref: os.stat(filepath).st_size
+Output: print ( os. path. join ( get _ list ) )
+bleu_metric: 13.834368456410944
+----------
+NL: count the occurrences of item "a" in list `l`
+ref: l.count('a')
+Output: [ x. count ('a') for x in l ]
+bleu_metric: 43.78826865860793
+----------
+NL: count the occurrences of items in list `l`
+ref: Counter(l)
+Output: [ sum ( x ) for x in l ]
+bleu_metric: 13.065113298388567
+----------
+NL: count the occurrences of items in list `l`
+ref: [[x, l.count(x)] for x in set(l)]
+Output: [ sum ( x ) for x in l ]
+bleu_metric: 16.443715448412245
+----------
+NL: count the occurrences of items in list `l`
+ref: dict(((x, l.count(x)) for x in set(l)))
+Output: [ sum ( x ) for x in l ]
+bleu_metric: 15.927159921457768
+----------
+NL: count the occurrences of item "b" in list `l`
+ref: l.count('b')
+Output: [ ( x, l. count ( x ) ) for x in l ]
+bleu_metric: 20.61477352156375
+----------
+NL: copy file `srcfile` to directory `dstdir`
+ref: shutil.copy(srcfile, dstdir)
+Output: os. chdir ('copyfile. txt ', dst )
+bleu_metric: 14.949751774990682
+----------
+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 ['x'], x ['x'] ) )
+bleu_metric: 9.186672262412122
+----------
+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: ( k for k, v in x. items ( ) if v! = 0 )
+bleu_metric: 100.0
+----------
+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 ( x, key = lambda x : ( x [ 1 ], x [ 0 ] ) )
+bleu_metric: 11.162938527547608
+----------
+NL: put the curser at beginning of the file
+ref: file.seek(0)
+Output: os. system ('with open ( filename ). read ( ) ','rb')
+bleu_metric: 8.687475782716618
+----------
+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 ['a']. apply ( lambda x : x ['b']. fillna
+bleu_metric: 43.50703089837015
+----------
+NL: remove key 'ele' from dictionary `d`
+ref: del d['ele']
+Output: d = { k : v for k, v in list ( d. items ( ) ) if v! ='none'}
+bleu_metric: 5.442133807846856
+----------
+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: mymodel. objects. filter ( date _ _ length ='% y - % m - % d')
+bleu_metric: 12.351156662075915
+----------
+NL: merge list `['it']` and list `['was']` and list `['annoying']` into one list
+ref: ['it'] + ['was'] + ['annoying']
+Output: [ dict ( zip ( * [ ['one ','two'], ['one ','two'] ] ) )
+bleu_metric: 10.105385745342533
+----------
+NL: increment a value with leading zeroes in a number `x`
+ref: str(int(x) + 1).zfill(len(x))
+Output: print ('\ xe, { 0 : x. 0 } '. format ( x, - 1 ) )
+bleu_metric: 13.471766853689124
+----------
+NL: check if a pandas dataframe `df`'s index is sorted
+ref: all(df.index[:-1] <= df.index[1:])
+Output: df. index. values. tolist ( )
+bleu_metric: 9.980212030371455
+----------
+NL: convert tuple `t` to list
+ref: list(t)
+Output: list ( t )
+bleu_metric: 100.0
+----------
+NL: convert list `t` to tuple
+ref: tuple(l)
+Output: tuple ( l )
+bleu_metric: 100.0
+----------
+NL: convert tuple `level1` to list
+ref: level1 = map(list, level1)
+Output: [ ( x + 1 ) for x in a ]
+bleu_metric: 11.35935489027116
+----------
+NL: send the output of pprint object `dataobject` to file `logfile`
+ref: pprint.pprint(dataobject, logFile)
+Output: sys. stdout. flush ( )
+bleu_metric: 18.67587389639562
+----------
+NL: get index of rows in column 'boolcol'
+ref: df.loc[df['BoolCol']]
+Output: df [ ['bool'] ]. index ('matrix2'). index
+bleu_metric: 18.710158230410627
+----------
+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. index. map ( lambda x : x [ 1 ]. isin ( x ) ) ]
+bleu_metric: 14.72821272412462
+----------
+NL: get list of indexes of rows where column 'boolcol' values match true
+ref: df[df['BoolCol'] == True].index.tolist()
+Output: [ df ['ratio'] for d in l ]
+bleu_metric: 19.62812256638541
+----------
+NL: get index of rows in dataframe `df` which column 'boolcol' matches value true
+ref: df[df['BoolCol']].index.tolist()
+Output: df ['column _ name'] = = df. index. str. get ('columnid')
+bleu_metric: 18.088201962129936
+----------
+NL: change working directory to the directory `owd`
+ref: os.chdir(owd)
+Output: os. chdir ('c : \ \ users \ \ users \ \ desktop \ \ desktop \ \ python')
+bleu_metric: 15.553986407412198
+----------
+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 ) ', ( ast, ) )
+bleu_metric: 34.29547961820059
+----------
+NL: decode string "\\x89\\n" into a normal string
+ref: """\\x89\\n""".decode('string_escape')
+Output: " " " \ \ xc3 \ \ x96 \ xc3 \ x96 \ x93 \ x93 \
+bleu_metric: 29.888108815158
+----------
+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: 35.556702356686955
+----------
+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: 30.620490882364876
+----------
+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 ('\ \ d + (? : \ \ s + ) ','a \ n b')
+bleu_metric: 12.792208459604188
+----------
+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: soup. find _ all ('div ', id ='all')
+bleu_metric: 46.28692516962493
+----------
+NL: filter rows containing key word `ball` in column `ids`
+ref: df[df['ids'].str.contains('ball')]
+Output: id. idxmax ( word ). drop ( false )
+bleu_metric: 7.630799524006417
+----------
+NL: convert index at level 0 into a column in dataframe `df`
+ref: df.reset_index(level=0, inplace=True)
+Output: df. set _ index ('level1'). unstack ( level = 0 )
+bleu_metric: 24.29335519279861
+----------
+NL: add indexes in a data frame `df` to a column `index1`
+ref: df['index1'] = df.index
+Output: df. loc [ index ] = df. index. get _ level _ values ( 0 )
+bleu_metric: 24.04315522172745
+----------
+NL: convert pandas index in a dataframe to columns
+ref: df.reset_index(level=['tick', 'obs'])
+Output: df. reset _ index ( drop = true )
+bleu_metric: 11.884004683899853
+----------
+NL: get reverse of list items from list 'b' using extended slicing
+ref: [x[::-1] for x in b]
+Output: [ [ i [ 0 ] for i in range ( len ( x ) ) ] for i in range ( 0, len ( list ( a )
+bleu_metric: 7.434360163166209
+----------
+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: b = np. where ( a = = 0, axis = 0 )
+bleu_metric: 9.027924728401977
+----------
+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. ravel ( ), b. ravel ( ) ), dtype ='i4, i4
+bleu_metric: 67.03200460356393
+----------
+NL: convert list `list_of_ints` into a comma separated string
+ref: """,""".join([str(i) for i in list_of_ints])
+Output: list ( map ( int, list _ of _ lists. split ( ',') ) )
+bleu_metric: 13.283261562346366
+----------
+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 = {'username':'username'} )
+bleu_metric: 37.18698903847989
+----------
+NL: find last occurrence of character '}' in string "abcd}def}"
+ref: 'abcd}def}'.rfind('}')
+Output: re. findall ('(? : ^ [ a - z ] * ) ','abcd0')
+bleu_metric: 12.962472880491877
+----------
+NL: iterate ove list `[1, 2, 3]` using list comprehension
+ref: print([item for item in [1, 2, 3]])
+Output: [ [ 1, 2, 3 ], [ 4, 5, 6 ], [ 7, 8, 9 ] ]
+bleu_metric: 27.073362211548464
+----------
+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'] for d in x if'y'in d ]
+bleu_metric: 37.60857585579953
+----------
+NL: get the filename without the extension from file 'hemanth.txt'
+ref: print(os.path.splitext(os.path.basename('hemanth.txt'))[0])
+Output: print ( glob. glob ('liname. txt') [ 0 ] )
+bleu_metric: 31.18776276497533
+----------
+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 ( ( k, v ) for k, v in zip ( a, b ) )
+bleu_metric: 12.152094584183624
+----------
+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'], ['a ','b ','c
+bleu_metric: 17.7080331882364
+----------
+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 ['column _ name']. isin ( some _ values ('one ','two
+bleu_metric: 10.948000684047626
+----------
+NL: replace all occurences of newlines `\n` with `
` in dataframe `df`
+ref: df.replace({'\n': '
'}, regex=True)
+Output: df. replace ('\ n ','\ n')
+bleu_metric: 27.770641036119947
+----------
+NL: replace all occurrences of a string `\n` by string `
` in a pandas data frame `df`
+ref: df.replace({'\n': '
'}, regex=True)
+Output: df. replace ( {'\ n':'< br >'}, regex = true )
+bleu_metric: 89.21616972156079
+----------
+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: [ ('a ','b'), ('b ','c'), ('c ','d')
+bleu_metric: 8.840994001530046
+----------
+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: [ str ( x ) for x in words ]
+bleu_metric: 3.8733243551839442
+----------
+NL: extract a url from a string `mystring`
+ref: print(re.findall('(https?://[^\\s]+)', myString))
+Output: print ( re. search ('( https? : / / [ ^ \ \ s ] + ) ', mystring ) )
+bleu_metric: 77.28607577370117
+----------
+NL: extract a url from a string `mystring`
+ref: print(re.search('(?Phttps?://[^\\s]+)', myString).group('url'))
+Output: print ( re. search ('( https? : / / [ ^ \ \ s ] + ) ', mystring ) )
+bleu_metric: 51.916231800616785
+----------
+NL: remove all special characters, punctuation and spaces from a string `mystring` using regex
+ref: re.sub('[^A-Za-z0-9]+', '', mystring)
+Output: re. sub ('\ \ w + (? = [ ^. ] * ) ','', mystring )
+bleu_metric: 45.91497693322866
+----------
+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: df. groupby ('date'). apply ( lambda x : x. year )
+bleu_metric: 8.453429635655233
+----------
+NL: create multidimensional array `matrix` with 3 rows and 2 columns in python
+ref: matrix = [[a, b], [c, d], [e, f]]
+Output: matrix = np. concatenate ( ( [ a, b ], [ [ 0, 3, 6 ], [ 7, 8, 9
+bleu_metric: 32.24199399071415
+----------
+NL: replace spaces with underscore
+ref: mystring.replace(' ', '_')
+Output: re. sub ('\ \ b (? = \ \ w + ) ','','this is a string ','replace
+bleu_metric: 14.117915208981241
+----------
+NL: get an absolute file path of file 'mydir/myfile.txt'
+ref: os.path.abspath('mydir/myfile.txt')
+Output: os. path. join ( os. path. join ( root, myfile ) )
+bleu_metric: 23.83041256525615
+----------
+NL: split string `my_string` on white spaces
+ref: """ """.join(my_string.split())
+Output: " " " 42, " " ". split ( )
+bleu_metric: 47.72263780018156
+----------
+NL: get filename without extension from file `filename`
+ref: os.path.splitext(filename)[0]
+Output: os. path. splitext ( filename ) [ 0 ]
+bleu_metric: 100.0
+----------
+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 ( x for x in l )
+bleu_metric: 6.924459302580939
+----------
+NL: split a string `docs/src/scripts/temp` by `/` keeping `/` in the result
+ref: """Docs/src/Scripts/temp""".replace('/', '/\x00/').split('\x00')
+Output: re. split ('\ \ s + ', string )
+bleu_metric: 7.926805970774756
+----------
+NL: shuffle columns of an numpy array 'r'
+ref: np.random.shuffle(np.transpose(r))
+Output: np. random. shuffle ( np. transpose ( r ) )
+bleu_metric: 100.0
+----------
+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 ['c']. apply ( lambda x : x. tolist ( ), axis =
+bleu_metric: 26.492685902784498
+----------
+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: print ( json. dumps ( ['b ','c'] ) )
+bleu_metric: 5.979778160388024
+----------
+NL: check characters of string `string` are true predication of function `predicate`
+ref: all(predicate(x) for x in string)
+Output: print ( any ( test _ string ) for substring in string _ string )
+bleu_metric: 19.923405658137924
+----------
+NL: determine number of files on a drive with python
+ref: os.statvfs('/').f_files - os.statvfs('/').f_ffree
+Output: os. stat ('/ myfile. txt'). os. listdir ('/')
+bleu_metric: 37.211067993181466
+----------
+NL: how to get a single result from a sqlite query in python?
+ref: cursor.fetchone()[0]
+Output: cursor. fetchone ( ) [ 0 ]
+bleu_metric: 100.0
+----------
+NL: convert string `user_input` into a list of integers `user_list`
+ref: user_list = [int(number) for number in user_input.split(',')]
+Output: [ int ( x ) for x in input ( ). split ( ',') ]
+bleu_metric: 51.23350305765596
+----------
+NL: get a list of integers by splitting a string `user` with comma
+ref: [int(s) for s in user.split(',')]
+Output: map ( int,'re. split ('( \ \ d + ) ','hello world') )
+bleu_metric: 20.80315522738391
+----------
+NL: sorting a python list by two criteria
+ref: sorted(list, key=lambda x: (x[0], -x[1]))
+Output: sorted ( list ( range ( 0, len ( s ) ), key = lambda x : ( x [ 0 ], x [ 0 ] )
+bleu_metric: 48.440807796000044
+----------
+NL: sort a list of objects `ut`, based on a function `cmpfun` in descending order
+ref: ut.sort(key=cmpfun, reverse=True)
+Output: sorted ( l, key = lambda x : x. modified, reverse = true )
+bleu_metric: 18.107197870881738
+----------
+NL: reverse list `ut` based on the `count` attribute of each object
+ref: ut.sort(key=lambda x: x.count, reverse=True)
+Output: ['city ','same2 ','type']
+bleu_metric: 7.453337029678027
+----------
+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 ( l, key = lambda x : x. count ('c') )
+bleu_metric: 45.34010611888324
+----------
+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 inmail'). click ( )
+bleu_metric: 35.13874939965221
+----------
+NL: click a href button having text `send inmail` with selenium
+ref: driver.findElement(By.linkText('Send InMail')).click()
+Output: driver. find _ element _ by _ link _ text ('send inmail')
+bleu_metric: 15.175278640994005
+----------
+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 _ link _ text ('send')
+bleu_metric: 15.27678022374703
+----------
+NL: cast an int `i` to a string and concat to string 'me'
+ref: 'ME' + str(i)
+Output: str ( i )
+bleu_metric: 36.787944117144235
+----------
+NL: sorting data in dataframe pandas
+ref: df.sort_values(['System_num', 'Dis'])
+Output: df. sort _ values ( ['system _ id ','dis'] )
+bleu_metric: 31.671615012203976
+----------
+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 ('test. txt ','w'). write ('\ n')
+bleu_metric: 36.176855814312944
+----------
+NL: sort a list `l` by length of value in tuple
+ref: l.sort(key=lambda t: len(t[1]), reverse=True)
+Output: l. sort ( key = lambda x : ( len ( x ), x ) )
+bleu_metric: 37.875123684341865
+----------
+NL: split string `s` by words that ends with 'd'
+ref: re.findall('\\b(\\w+)d\\b', s)
+Output: re. split ('( \ \ d + ) ', s )
+bleu_metric: 29.23155541924467
+----------
+NL: return `true` if string `foobarrrr` contains regex `ba[rzd]`
+ref: bool(re.search('ba[rzd]', 'foobarrrr'))
+Output: pattern = re. compile ('\ \ b [ a - za - z0 - 9 - ]')
+bleu_metric: 12.384901282810546
+----------
+NL: removing duplicates in list `t`
+ref: list(set(t))
+Output: list ( set ( t ) )
+bleu_metric: 100.0
+----------
+NL: removing duplicates in list `source_list`
+ref: list(set(source_list))
+Output: list ( set ( t ) )
+bleu_metric: 59.69491792019645
+----------
+NL: removing duplicates in list `abracadabra`
+ref: list(OrderedDict.fromkeys('abracadabra'))
+Output: [ ('a ','b'), ('b ','c'), ('a ','d')
+bleu_metric: 7.555283720961675
+----------
+NL: convert array `a` into a list
+ref: numpy.array(a).reshape(-1).tolist()
+Output: list ( a )
+bleu_metric: 5.237603475197421
+----------
+NL: convert the first row of numpy matrix `a` to a list
+ref: numpy.array(a)[0].tolist()
+Output: numpy. apply _ along _ axis ( numpy. array ( [ 0, 1, 2 ] ), a )
+bleu_metric: 20.97823227594146
+----------
+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 _ id ('< id >'). text
+bleu_metric: 12.033574792347215
+----------
+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 ( x ) for x in l ]
+bleu_metric: 17.41466588478686
+----------
+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: " " " " " ". join ( [ x for x in l if x [ 0 ] in l ] )
+bleu_metric: 36.45674601639763
+----------
+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 : / / www. google. com')
+bleu_metric: 19.209534151258666
+----------
+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: [ i for i in teststr if re. search ('\ \ d + [ ^ \ \ ] * \ \ ] ', i )
+bleu_metric: 69.89029944247109
+----------
+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 [ ( df ['a'] > 1 ) & ( df ['b'] < = 5 ) ]
+bleu_metric: 39.042104382208294
+----------
+NL: sort dictionary `o` in ascending order based on its keys and items
+ref: sorted(o.items())
+Output: sorted ( list ( dictionary. items ( ) ), key = lambda x : x [ 1 ] )
+bleu_metric: 24.42195895839763
+----------
+NL: get sorted list of keys of dict `d`
+ref: sorted(d)
+Output: sorted ( d, key = d. get )
+bleu_metric: 25.21193618434983
+----------
+NL: how to sort dictionaries by keys in python
+ref: sorted(d.items())
+Output: sorted ( d. items ( ) )
+bleu_metric: 100.0
+----------
+NL: convert string "1" into integer
+ref: int('1')
+Output: int ('1')
+bleu_metric: 100.0
+----------
+NL: function to convert strings into integers
+ref: int()
+Output: map ( int,'1')
+bleu_metric: 21.105340631872636
+----------
+NL: convert items in `t1` to integers
+ref: T2 = [map(int, x) for x in T1]
+Output: int ('\ n '. join ( str ( i ) for i in str ( n ) ) )
+bleu_metric: 8.607692533178168
+----------
+NL: call a shell script `./test.sh` using subprocess
+ref: subprocess.call(['./test.sh'])
+Output: subprocess. call ( ['test. sh ','/ test. sh otherfunc'] )
+bleu_metric: 53.66411241731205
+----------
+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 in l2 for y in zip ( l1, l2 ) ]
+bleu_metric: 38.146366463982886
+----------
+NL: encode string 'data to be encoded'
+ref: encoded = base64.b64encode('data to be encoded')
+Output: data. encode ('utf8')
+bleu_metric: 19.260282250359822
+----------
+NL: encode a string `data to be encoded` to `ascii` encoding
+ref: encoded = 'data to be encoded'.encode('ascii')
+Output: data. encode ('ascii')
+bleu_metric: 41.51380951089673
+----------
+NL: parse tab-delimited csv file 'text.txt' into a list
+ref: lol = list(csv.reader(open('text.txt', 'rb'), delimiter='\t'))
+Output: txt = open ('myfile. txt ','r'). read ( )
+bleu_metric: 20.692050735338164
+----------
+NL: get attribute `my_str` of object `my_object`
+ref: getattr(my_object, my_str)
+Output: getattr ( my _ obj,'id')
+bleu_metric: 19.72940627795883
+----------
+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: values = [ dict ( k, v ) for k, v in list ( d. items ( ) ) ]
+bleu_metric: 17.057966110550765
+----------
+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 ( [ item [ 0 ] for item in tuple ] )
+bleu_metric: 41.37441202051881
+----------
+NL: convert unicode string u"{'code1':1,'code2':1}" into dictionary
+ref: d = ast.literal_eval("{'code1':1,'code2':1}")
+Output: print ('\ u0420 \ u043e \ u0441 \ u0438 \ u0438 \ u
+bleu_metric: 6.351019285070924
+----------
+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 ('\ \ b [ ^ \ \ w \ \ ] ', mystring )
+bleu_metric: 9.850775379403402
+----------
+NL: remove any url within string `text`
+ref: text = re.sub('^https?:\\/\\/.*[\\r\\n]*', '', text, flags=re.MULTILINE)
+Output: re. sub ('\ \. \ \ blog _ addr \ \ r \ \ n ','', text )
+bleu_metric: 32.17551945847391
+----------
+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. in1d ( a, b ) ) [ 0 ]
+bleu_metric: 22.828499613963487
+----------
+NL: calculate mean across dimension in a 2d array `a`
+ref: np.mean(a, axis=1)
+Output: np. mean ( a, axis = 1 )
+bleu_metric: 100.0
+----------
+NL: running r script '/pathto/myrscript.r' from python
+ref: subprocess.call(['/usr/bin/Rscript', '--vanilla', '/pathto/MyrScript.r'])
+Output: sys. path. append ('/ path / to / my / my / mypath')
+bleu_metric: 7.04483155666154
+----------
+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. path. join ( root, '. / main _ folder')
+bleu_metric: 6.726052538373431
+----------
+NL: add a header to a csv file
+ref: writer.writeheader()
+Output: writer. writeheader ( )
+bleu_metric: 100.0
+----------
+NL: replacing nan in the dataframe `df` with row average
+ref: df.fillna(df.mean(axis=1), axis=1)
+Output: df. fillna ( df. mean ( axis = 1 ), axis = 1 )
+bleu_metric: 100.0
+----------
+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: datetime. strptime ('21 / 11 / 06 16 : 30 ','% d / % m / % y')
+bleu_metric: 10.247261229308657
+----------
+NL: call a base class's class method `do` from derived class `derived`
+ref: super(Derived, cls).do(a)
+Output: super ( childclass, self ). _ _ init _ _ ( * args, * * kwargs )
+bleu_metric: 11.96655750514248
+----------
+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. arange ( 3 ), ( 0, 1, 0 ) ]
+bleu_metric: 13.581342454223687
+----------
+NL: separate words delimited by one or more spaces into a list
+ref: re.split(' +', 'hello world sample text')
+Output: ['one ','two ','one ','two']
+bleu_metric: 13.78592993183041
+----------
+NL: length of longest element in list `words`
+ref: len(max(words, key=len))
+Output: len ( max ( words, key = len ) )
+bleu_metric: 100.0
+----------
+NL: get the value associated with unicode key 'from_user' of first dictionary in list `result`
+ref: result[0]['from_user']
+Output: ['key1 ','key2 ','value']
+bleu_metric: 18.20705281109213
+----------
+NL: retrieve each line from a file 'file.txt' as a list
+ref: [line.split() for line in open('File.txt')]
+Output: print ( line. rstrip ('\ n') )
+bleu_metric: 11.789701057610694
+----------
+NL: swap keys with values in a dictionary `a`
+ref: res = dict((v, k) for k, v in a.items())
+Output: a [ key ] = dict ( ( v ) for k, v in a. items ( ) )
+bleu_metric: 73.78351342269067
+----------
+NL: open a file `path/to/file_name.ext` in write mode
+ref: new_file = open('path/to/FILE_NAME.ext', 'w')
+Output: file _ name = os. path. join ( os. path. join ( dir, '.') )
+bleu_metric: 10.713701843513144
+----------
+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'] ) ['col3']. apply ( lambda x :
+bleu_metric: 72.38800804180103
+----------
+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 ( isinstance ( x, int ( dict. values ( ) ) ) for x in dict _ string in input. lower (
+bleu_metric: 8.839374326825924
+----------
+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: [ dict ( zip ( k, v ), v ) for k, v in d. items ( ) if v. startswith ( '
+bleu_metric: 24.14138031982595
+----------
+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: pd. concat ( [ df1, df2 ], axis = 1 )
+bleu_metric: 8.12657367724027
+----------
+NL: print elements of list `list` seperated by tabs `\t`
+ref: print('\t'.join(map(str, list)))
+Output: print ( [ i. partition ('\ t') for i in list _ of _ lists ] )
+bleu_metric: 25.58058053378779
+----------
+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 ('\ xc5 \ xa4 \ xd0 \ x80 \ xd0 \ xb0 \ xd0
+bleu_metric: 29.472590166651962
+----------
+NL: encode a latin character in string `sopet\xc3\xb3n` properly
+ref: 'Sopet\xc3\xb3n'.encode('latin-1').decode('utf-8')
+Output: " " " \ xc3 \ xa3 \ x9f " " ". encode ('utf - 8')
+bleu_metric: 46.255815148030386
+----------
+NL: resized image `image` to width, height of `(x, y)` with filter of `antialias`
+ref: image = image.resize((x, y), Image.ANTIALIAS)
+Output: img = np. array ( [ 0, 1, 0, 1, 2, - 1, 0, 1, 2 ] )
+bleu_metric: 7.278748444764005
+----------
+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 ('(? < = [ ^ \ \ s ] ) (? = [ ^ \ \ s ] + ) ',
+bleu_metric: 44.74679573461579
+----------
+NL: display the float `1/3*100` as a percentage
+ref: print('{0:.0f}%'.format(1.0 / 3 * 100))
+Output: print ( ('0. 1 ', 1 ) )
+bleu_metric: 11.406542828786534
+----------
+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: l. sort ( key = lambda x : x ['title'] )
+bleu_metric: 100.0
+----------
+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 ( l, key = lambda x : x ['name'] )
+bleu_metric: 17.362697834284052
+----------
+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: [ ( x, y ) for x, y in zip ( l1, l2 ) ]
+bleu_metric: 5.935298261592072
+----------
+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 ( text ='div ', style ='width = 300px ;')
+bleu_metric: 11.970571135993998
+----------
+NL: write records in dataframe `df` to table 'test' in schema 'a_schema'
+ref: df.to_sql('test', engine, schema='a_schema')
+Output: df. plot ( row ='test ','test _ txt')
+bleu_metric: 32.62478546610937
+----------
+NL: extract brackets from string `s`
+ref: brackets = re.sub('[^(){}[\\]]', '', s)
+Output: re. sub ('\ \ [ ( \ \ w + ) \ \ s + \ \ ] ','', s )
+bleu_metric: 47.1113978575501
+----------
+NL: remove duplicate elements from list 'l'
+ref: list(dict((x[0], x) for x in L).values())
+Output: [ k for k, g in l if'l'in k ]
+bleu_metric: 6.942420857748657
+----------
+NL: read a file `file` without newlines
+ref: [line.rstrip('\n') for line in file]
+Output: file. write ('\ n')
+bleu_metric: 30.98188793972842
+----------
+NL: get the position of item 1 in `testlist`
+ref: [i for (i, x) in enumerate(testlist) if (x == 1)]
+Output: print ( [ item for item in mylist if item. isdigit ( ) ] )
+bleu_metric: 9.243620302376062
+----------
+NL: get the position of item 1 in `testlist`
+ref: [i for (i, x) in enumerate(testlist) if (x == 1)]
+Output: print ( [ item for item in mylist if item. isdigit ( ) ] )
+bleu_metric: 9.243620302376062
+----------
+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 ( [ item for item in mylist if item. isdigit ( ) ] )
+bleu_metric: 7.565780264883953
+----------
+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 ( [ item for item in mylist if item. isdigit ( ) ] )
+bleu_metric: 6.0168435849456285
+----------
+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 ( [ item for item in mylist if item. isdigit ( ) ] )
+bleu_metric: 5.043459282125431
+----------
+NL: get the position of item `element` in list `testlist`
+ref: print(testlist.index(element))
+Output: print ( testlist. index ( element ) )
+bleu_metric: 100.0
+----------
+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: 49.01732980393223
+----------
+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 ( x [ 0 ] for x in rows )
+bleu_metric: 14.988854080083206
+----------
+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 ( x, key = lambda x : x [ 1 ] ) [ 0 ]
+bleu_metric: 35.27295712700594
+----------
+NL: make a delay of 1 second
+ref: time.sleep(1)
+Output: time. sleep ( 0. 1 )
+bleu_metric: 52.53819788848316
+----------
+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: obj. save ('new _ name ','default _ number')
+bleu_metric: 10.318679011582763
+----------
+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 ( lst, key = lambda x : x [ 1 ] )
+bleu_metric: 19.143591202259493
+----------
+NL: how do i perform secondary sorting in python?
+ref: sorted(list5, key=lambda vertex: (degree(vertex), vertex))
+Output: sorted ( list5, key = lambda x : ( sum ( x [ 1 ] ), x ) )
+bleu_metric: 41.26152034907945
+----------
+NL: convert a list into a generator object
+ref: (n for n in [1, 2, 3, 5])
+Output: list ( itertools. combinations ( * ( l ) ) )
+bleu_metric: 9.930751066504135
+----------
+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 set ( split _ list [ i ] for i in range ( len ( x ) ) if i not in index ]
+bleu_metric: 18.860270984795903
+----------
+NL: open a file `yourfile.txt` in write mode
+ref: f = open('yourfile.txt', 'w')
+Output: os. system ('your string. txt')
+bleu_metric: 23.571525044211057
+----------
+NL: get attribute 'attr' from object `obj`
+ref: getattr(obj, 'attr')
+Output: getattr ( obj,'attr')
+bleu_metric: 100.0
+----------
+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 ( ['a ','b ','c ','d'] )
+bleu_metric: 6.101077050680879
+----------
+NL: convert tuple of tuples `(('aa',), ('bb',), ('cc',))` to list in one line
+ref: map(lambda a: a[0], (('aa',), ('bb',), ('cc',)))
+Output: map ( lambda s : ( d ['a'], d ['b'], d ['c'] ), d [
+bleu_metric: 11.200083731986627
+----------
+NL: python pandas: how to replace a characters in a column of a dataframe?
+ref: df['range'].replace(',', '-', inplace=True)
+Output: df ['brandname'] = df ['brandname']. replace ('- ', ','in')
+bleu_metric: 45.97006398721532
+----------
+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 ','b ','c ','d'] ) )
+bleu_metric: 36.501102157810664
+----------
+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 ','b ','c ','d'] ) )
+bleu_metric: 36.501102157810664
+----------
+NL: unzip list `original`
+ref: result = ([a for (a, b) in original], [b for (a, b) in original])
+Output: zip ( * list ( range ( 10 ) ) )
+bleu_metric: 4.606184881898074
+----------
+NL: unzip list `original` and return a generator
+ref: result = ((a for (a, b) in original), (b for (a, b) in original))
+Output: [ i for i in range ( 1, len ( l ) ) if i! ='+ 1 ]
+bleu_metric: 7.62604991346693
+----------
+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 ','b ','c ','d ','e'] ) )
+bleu_metric: 34.87995424158042
+----------
+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 ','b'), ('b ','c ','d'), ( '
+bleu_metric: 35.430131828411845
+----------
+NL: encode `decimal('3.9')` to a json string
+ref: json.dumps(Decimal('3.9'))
+Output: json. dumps ( ['1 ','2 ','3'] )
+bleu_metric: 23.58844810653421
+----------
+NL: add key "mynewkey" to dictionary `d` with value "mynewvalue"
+ref: d['mynewkey'] = 'mynewvalue'
+Output: df. setdefault ('name ', [ ] ). append ( value )
+bleu_metric: 9.50724412002623
+----------
+NL: add key 'a' to dictionary `data` with value 1
+ref: data.update({'a': 1, })
+Output: a ['a']. append ( {'a': 1,'b': 2 } )
+bleu_metric: 37.863495563972855
+----------
+NL: add key 'a' to dictionary `data` with value 1
+ref: data.update(dict(a=1))
+Output: a ['a']. append ( {'a': 1,'b': 2 } )
+bleu_metric: 8.24055698798382
+----------
+NL: add key 'a' to dictionary `data` with value 1
+ref: data.update(a=1)
+Output: a ['a']. append ( {'a': 1,'b': 2 } )
+bleu_metric: 8.24055698798382
+----------
+NL: find maximal value in matrix `matrix`
+ref: max([max(i) for i in matrix])
+Output: max ( matrix, key = lambda x : max ( x ) )
+bleu_metric: 16.943571815930884
+----------
+NL: round number `answer` to 2 precision after the decimal point
+ref: answer = str(round(answer, 2))
+Output: round ( 1. 0005, 3 )
+bleu_metric: 17.51643270174889
+----------
+NL: extract ip address from an html string
+ref: ip = re.findall('[0-9]+(?:\\.[0-9]+){3}', s)
+Output: print ( re. findall ('[ ^ a - za - z0 - 9 - ] + (? : \ \. [ a -
+bleu_metric: 40.70867848415918
+----------
+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 ( df. filter ( lambda x : x. any ( axis = 1 ) ) )
+bleu_metric: 34.640201116587235
+----------
+NL: append each line in file `myfile` into a list
+ref: [x for x in myfile.splitlines() if x != '']
+Output: myfile. write ('\ n '. join ( [ str ( i ) for i in myfile ] ) )
+bleu_metric: 10.435185733665413
+----------
+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 lst ]
+bleu_metric: 7.020674481218203
+----------
+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. groupby ('text'). apply ( lambda x : x. sum ( ) ). most _ common ( 1 )
+bleu_metric: 21.52643286782715
+----------
+NL: python split a string using regex
+ref: re.findall('(.+?):(.+?)\\b ?', text)
+Output: re. split ('(? < = \ \ d ) \ \ d + ', text )
+bleu_metric: 29.769819080461428
+----------
+NL: generate all 2-element subsets of tuple `(1, 2, 3)`
+ref: list(itertools.combinations((1, 2, 3), 2))
+Output: list ( itertools. product ( [ 1, 2, 3 ], [ 4, 5, 6 ] ) )
+bleu_metric: 34.29547961820059
+----------
+NL: get a value of datetime.today() in the utc time zone
+ref: datetime.now(pytz.utc)
+Output: pytz. utcnow ( )
+bleu_metric: 23.206041459353084
+----------
+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 [ : ] = = [ ] ]
+bleu_metric: 69.3395566222006
+----------
+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: list ( zip ( a, b ) )
+bleu_metric: 10.320893749383378
+----------
+NL: django response with json `data`
+ref: return HttpResponse(data, mimetype='application/json')
+Output: return httpresponse ( json. dumps ( response _ data ), content _ type ='application / json')
+bleu_metric: 36.17640392425988
+----------
+NL: get all text that is not enclosed within square brackets in string `example_str`
+ref: re.findall('(.*?)\\[.*?\\]', example_str)
+Output: str1. split ('\ \ ) [ ^ \ \ ] ', str. strip ( )
+bleu_metric: 26.91730507369109
+----------
+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: re. findall ('\ \ [ (? : \ \ w + ) \ \ ] ', str1 )
+bleu_metric: 35.15147667690756
+----------
+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. sub ('\ \ ( [ ^ \ \ w ] ) ','', filename )
+bleu_metric: 27.664786097762494
+----------
+NL: match regex '\\((.*?)\\)|(\\w)' with string '(zyx)bc'
+ref: re.findall('\\((.*?)\\)|(\\w)', '(zyx)bc')
+Output: re. sub ('(? < = \ \ w ) ( \ \ w + ) ( \ \ w + ) ','\ \
+bleu_metric: 30.955387518700363
+----------
+NL: match multiple regex patterns with the alternation operator `|` in a string `(zyx)bc`
+ref: re.findall('\\(.*?\\)|\\w', '(zyx)bc')
+Output: re. findall ('(? < = \ \ w ) ( [ ^ \ \ w \ \ w ] | \ \ w + )
+bleu_metric: 29.552612305356018
+----------
+NL: formate each string cin list `elements` into pattern '%{0}%'
+ref: elements = ['%{0}%'.format(element) for element in elements]
+Output: " " " { 0 :. 2f } " " ". format ('foo ','bar')
+bleu_metric: 17.414675099971177
+----------
+NL: open a background process 'background-process' with arguments 'arguments'
+ref: subprocess.Popen(['background-process', 'arguments'])
+Output: os. system ('some _ command - some _ file')
+bleu_metric: 12.107470158860224
+----------
+NL: get list of values from dictionary 'mydict' w.r.t. list of keys 'mykeys'
+ref: [mydict[x] for x in mykeys]
+Output: [ key for item in my _ list if item. get ('name ','age') ]
+bleu_metric: 9.119675426861836
+----------
+NL: convert list `[('name', 'joe'), ('age', 22)]` into a dictionary
+ref: dict([('Name', 'Joe'), ('Age', 22)])
+Output: dict ( ( ('class ', 2 ), ('class ', 3 ) ) )
+bleu_metric: 24.080231808676874
+----------
+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. arange ( len ( df. columns ) ), axis = 1 )
+bleu_metric: 18.410768031391818
+----------
+NL: double backslash escape all double quotes in string `s`
+ref: print(s.encode('unicode-escape').replace('"', '\\"'))
+Output: print ( s. split ('\ \') )
+bleu_metric: 21.765088513075124
+----------
+NL: split a string into a list of words and whitespace
+ref: re.split('(\\W+)', s)
+Output: ['this is a string : ','this is a string ','this is a string ','this is a string ',
+bleu_metric: 5.648892151960283
+----------
+NL: plotting stacked barplots on a panda data frame
+ref: df.plot(kind='barh', stacked=True)
+Output: df. plot ( kind ='barh ', stacked = true )
+bleu_metric: 83.7117009877792
+----------
+NL: reverse the keys and values in a dictionary `mydictionary`
+ref: {i[1]: i[0] for i in list(myDictionary.items())}
+Output: dict ( ( k, v ) for k, v in list ( d. items ( ) ) )
+bleu_metric: 23.690464140247087
+----------
+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: ['x ','y ','red ','var ','var']
+bleu_metric: 6.180827527485682
+----------
+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 ( s, str )
+bleu_metric: 59.46035575013605
+----------
+NL: check if object `o` is a string
+ref: (type(o) is str)
+Output: isinstance ( s, str )
+bleu_metric: 23.813694985189848
+----------
+NL: check if object `o` is a string
+ref: isinstance(o, str)
+Output: isinstance ( s, str )
+bleu_metric: 59.46035575013605
+----------
+NL: check if `obj_to_test` is a string
+ref: isinstance(obj_to_test, str)
+Output: print ( obj. isfile ('attr name') )
+bleu_metric: 13.508625657351411
+----------
+NL: append list `list1` to `list2`
+ref: list2.extend(list1)
+Output: [ list1. append ( list1 ) for a in range ( 3 ) ]
+bleu_metric: 14.879641171245488
+----------
+NL: append list `mylog` to `list1`
+ref: list1.extend(mylog)
+Output: list3. append ( list1 )
+bleu_metric: 29.071536848410968
+----------
+NL: append list `a` to `c`
+ref: c.extend(a)
+Output: a. append ( a )
+bleu_metric: 54.10822690539396
+----------
+NL: append items in list `mylog` to `list1`
+ref: for line in mylog:
+ list1.append(line)
+Output: list3. append ( list1 )
+bleu_metric: 20.917479021833486
+----------
+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: a = [ [ ] for i in range ( 0, len ( a ), [ 3, 4 ] ] )
+bleu_metric: 11.556647985416685
+----------
+NL: initialize `secret_key` in flask config with `your_secret_string `
+ref: app.config['SECRET_KEY'] = 'Your_secret_string'
+Output: app. config ['your _ id'] ='your _ config '
+bleu_metric: 43.343660127583234
+----------
+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. values. tolist ( ), columns = ['a ','b'], axis = 1
+bleu_metric: 35.31494509780309
+----------
+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: print ( list ( enumerate ( my _ list ) ) )
+bleu_metric: 5.9394573438219
+----------
+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 ( ), rotation = 45 )
+bleu_metric: 29.88309778054086
+----------
+NL: remove symbols from a string `s`
+ref: re.sub('[^\\w]', ' ', s)
+Output: re. sub ('[ ^ a - z ] ','', s )
+bleu_metric: 74.83293841345244
+----------
+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. abspath ( _ _ file _ _ ) )
+bleu_metric: 48.431464491050775
+----------
+NL: find octal characters matches from a string `str` using regex
+ref: print(re.findall("'\\\\[0-7]{1,3}'", str))
+Output: re. findall ('\ \ [ (? < = \ \ d ) \ \ ) ','- ', str )
+bleu_metric: 34.98979283499355
+----------
+NL: split string `input` based on occurrences of regex pattern '[ ](?=[a-z]+\\b)'
+ref: re.split('[ ](?=[A-Z]+\\b)', input)
+Output: re. findall ('(? < = [ a - z ] ) (? = [ a - z ] ) (? = [ a
+bleu_metric: 16.927318970546587
+----------
+NL: split string `input` at every space followed by an upper-case letter
+ref: re.split('[ ](?=[A-Z])', input)
+Output: re. split ('\ \ b ', text )
+bleu_metric: 23.369549958807866
+----------
+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: r = requests. get ( url )
+bleu_metric: 13.273006566054132
+----------
+NL: write bytes `bytes_` to a file `filename` in python 3
+ref: open('filename', 'wb').write(bytes_)
+Output: f. write ( struct. pack ('> i ', filename ) )
+bleu_metric: 21.042990347620457
+----------
+NL: get a list from a list `lst` with values mapped into a dictionary `dct`
+ref: [dct[k] for k in lst]
+Output: [ list ( lst [ i ] ) for i in lst ]
+bleu_metric: 28.787878181011266
+----------
+NL: find duplicate names in column 'name' of the dataframe `x`
+ref: x.set_index('name').index.get_duplicates()
+Output: print ( list ( df. index. get _ duplicates ( ) ) )
+bleu_metric: 19.923405658137924
+----------
+NL: truncate float 1.923328437452 to 3 decimal places
+ref: round(1.923328437452, 3)
+Output: round ( 1. 005, 2 )
+bleu_metric: 46.71379777282001
+----------
+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 = operator. itemgetter ( 1 ), reverse = true )
+bleu_metric: 19.26332463060877
+----------
+NL: place the radial ticks in plot `ax` at 135 degrees
+ref: ax.set_rlabel_position(135)
+Output: ax. set _ xticklabels ( [ ] )
+bleu_metric: 21.834177214239062
+----------
+NL: check if path `my_path` is an absolute path
+ref: os.path.isabs(my_path)
+Output: os. path. exists ('my _ path')
+bleu_metric: 31.61487584488943
+----------
+NL: get number of keys in dictionary `yourdict`
+ref: len(list(yourdict.keys()))
+Output: len ( mydict )
+bleu_metric: 11.141275535087015
+----------
+NL: count the number of keys in dictionary `yourdictfile`
+ref: len(set(open(yourdictfile).read().split()))
+Output: sum ( len ( v ) for v in yourdict. values ( ) )
+bleu_metric: 18.767030063813223
+----------
+NL: pandas dataframe get first row of each group by 'id'
+ref: df.groupby('id').first()
+Output: df. groupby ('id'). first ( ). first ( )
+bleu_metric: 75.62722459724604
+----------
+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 [ ['c ','df'] ]. apply ( lambda x : x [ 1 : - 1 ]. split (
+bleu_metric: 19.037861963633805
+----------
+NL: extract attributes 'src="js/([^"]*\\bjquery\\b[^"]*)"' from string `data`
+ref: re.findall('src="js/([^"]*\\bjquery\\b[^"]*)"', data)
+Output: " " " \ \ b " " "
+bleu_metric: 3.6253194206622026
+----------
+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 ( [ 1, 2, 3, 4, 5 ] )
+bleu_metric: 3.255351208100268
+----------
+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 ( ['shutdown ','/ r ','/ tmp / test. py'] )
+bleu_metric: 16.665685159230854
+----------
+NL: reverse a priority queue `q` in python without using classes
+ref: q.put((-n, n))
+Output: t. start ( )
+bleu_metric: 12.151662434083677
+----------
+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: df. groupby ('o') ['value']. first ( )
+bleu_metric: 3.5795004593328636
+----------
+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 - z ] ) ( [ a - z ] ) ','abcd _ string
+bleu_metric: 33.336490646495264
+----------
+NL: get the length of list `my_list`
+ref: len(my_list)
+Output: len ( my _ list )
+bleu_metric: 30.739407647563223
+----------
+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 _ list )
+bleu_metric: 30.739407647563223
+----------
+NL: getting the length of `my_string`
+ref: len(my_string)
+Output: len ( my _ string )
+bleu_metric: 30.739407647563223
+----------
+NL: remove escape character from string "\\a"
+ref: """\\a""".decode('string_escape')
+Output: print ('\ n '. join ('[ _ name _ ]') )
+bleu_metric: 15.071676257541073
+----------
+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: a = re. sub ('[ a - za - z0 - 9 ] ','', line )
+bleu_metric: 6.256610657452671
+----------
+NL: remove directory tree '/folder_name'
+ref: shutil.rmtree('/folder_name')
+Output: os. path. dirname ( os. path. join ( dirpath, os. path. dirname ( _ _ file _ _ ) )
+bleu_metric: 6.4718242450883325
+----------
+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: df. apply ( lambda x : x. to _ frame ('new _ column'), axis = 1 )
+bleu_metric: 34.619282106807496
+----------
+NL: reverse sort counter `x` by values
+ref: sorted(x, key=x.get, reverse=True)
+Output: sorted ( x, key = itemgetter ( 1 ) )
+bleu_metric: 40.540555470323305
+----------
+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: 15.53218858512746
+----------
+NL: append a numpy array 'b' to a numpy array 'a'
+ref: np.vstack((a, b))
+Output: np. concatenate ( ( a, b ) )
+bleu_metric: 76.91605673134586
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: print(concatenate((a, b), axis=0))
+Output: print ( concatenate ( ( a, b ), axis = 1 ) )
+bleu_metric: 82.82477531331041
+----------
+NL: numpy concatenate two arrays `a` and `b` along the second axis
+ref: print(concatenate((a, b), axis=1))
+Output: print ( concatenate ( ( a, b ), axis = 1 ) )
+bleu_metric: 100.0
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: c = np.r_[(a[None, :], b[None, :])]
+Output: print ( concatenate ( ( a, b ), axis = 1 ) )
+bleu_metric: 9.520068110441361
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: np.array((a, b))
+Output: print ( concatenate ( ( a, b ), axis = 1 ) )
+bleu_metric: 41.70507291634131
+----------
+NL: fetch address information for host 'google.com' ion port 80
+ref: print(socket.getaddrinfo('google.com', 80))
+Output: os. environ ['localhost : \ \ python27 \ \ python']
+bleu_metric: 9.672649511413093
+----------
+NL: add a column 'day' with value 'sat' to dataframe `df`
+ref: df.xs('sat', level='day', drop_level=False)
+Output: df ['value'] = df ['value']. apply ( lambda x : x ['value'] )
+bleu_metric: 8.179879050970094
+----------
+NL: return a 401 unauthorized in django
+ref: return HttpResponse('Unauthorized', status=401)
+Output: return httpresponse ( status = 204 )
+bleu_metric: 15.451666492113134
+----------
+NL: flask set folder 'wherever' as the default template folder
+ref: Flask(__name__, template_folder='wherever')
+Output: app. run ( debug = true )
+bleu_metric: 16.89983564524028
+----------
+NL: how do i insert into t1 (select * from t2) in sqlalchemy?
+ref: session.execute('INSERT INTO t1 (SELECT * FROM t2)')
+Output: session. execute ('select * from from t1 ;')
+bleu_metric: 34.9740896964685
+----------
+NL: sort a list of lists 'c2' such that third row comes first
+ref: c2.sort(key=lambda row: row[2])
+Output: sorted ( l, key = lambda x : ( - x [ 1 ], x [ 2 ] ) )
+bleu_metric: 22.169194554222997
+----------
+NL: sorting a list of lists in python
+ref: c2.sort(key=lambda row: (row[2], row[1], row[0]))
+Output: [ ( row [ i ], row [ i + 2 ] ) for i in range ( 0, len ( row ), row [ 2 ]
+bleu_metric: 26.189293164591476
+----------
+NL: sorting a list of lists in python
+ref: c2.sort(key=lambda row: (row[2], row[1]))
+Output: [ ( row [ i ], row [ i + 2 ] ) for i in range ( 0, len ( row ), row [ 2 ]
+bleu_metric: 22.303080489498637
+----------
+NL: set font `arial` to display non-ascii characters in matplotlib
+ref: matplotlib.rc('font', **{'sans-serif': 'Arial', 'family': 'sans-serif'})
+Output: ax. set _ ylabel ('y - r ', title ='\ u2022')
+bleu_metric: 6.906418747057865
+----------
+NL: convert datetime column 'date' of pandas dataframe 'df' to ordinal
+ref: df['date'].apply(lambda x: x.toordinal())
+Output: df ['date'] = df ['date']. map ( lambda x : x. date ( ) )
+bleu_metric: 51.29497107827519
+----------
+NL: get html source of selenium webelement `element`
+ref: element.get_attribute('innerHTML')
+Output: print ( element. get _ attribute ('innerhtml') )
+bleu_metric: 18.69300079996002
+----------
+NL: get the integer location of a key `bob` in a pandas data frame
+ref: df.index.get_loc('bob')
+Output: df. reset _ index ( )
+bleu_metric: 19.840332506055578
+----------
+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 ('echo $ filename. exe')
+bleu_metric: 12.044064401053362
+----------
+NL: add an item with key 'third_key' and value 1 to an dictionary `my_dict`
+ref: my_dict.update({'third_key': 1})
+Output: my _ dict ['my _ dict'] = var
+bleu_metric: 10.579369505074823
+----------
+NL: declare an array
+ref: my_list = []
+Output: variable = [ ]
+bleu_metric: 75.98356856515926
+----------
+NL: insert item `12` to a list `my_list`
+ref: my_list.append(12)
+Output: my _ list. append ('{ }, { } '. format ( c. c ) )
+bleu_metric: 12.259149958656423
+----------
+NL: add an entry 'wuggah' at the beginning of list `mylist`
+ref: myList.insert(0, 'wuggah')
+Output: mylist = [ 'an ','i ','j']
+bleu_metric: 11.502783619900047
+----------
+NL: convert a hex-string representation to actual bytes
+ref: """\\xF3\\xBE\\x80\\x80""".replace('\\x', '').decode('hex')
+Output: " " " \ \ xc3 \ \ x85あ " " ". encode ('hex')
+bleu_metric: 24.712086993335753
+----------
+NL: select the last column of dataframe `df`
+ref: df[df.columns[-1]]
+Output: df [ df. columns [ - 1 ] ]
+bleu_metric: 100.0
+----------
+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 ['value'] = df. groupby ('value') ['value']. apply ( lambda x : x
+bleu_metric: 17.966302274629616
+----------
+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: 35.61693754184905
+----------
+NL: get the type of `i`
+ref: type(i)
+Output: print ( len ( a ) )
+bleu_metric: 22.679164443904003
+----------
+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 ( ) ['a']
+bleu_metric: 16.784459625186194
+----------
+NL: get the 5th item of a generator
+ref: next(itertools.islice(range(10), 5, 5 + 1))
+Output: next ( ( iter ( range ( n ) ) ) for i in range ( n ) )
+bleu_metric: 18.81478574691708
+----------
+NL: print a string `word` with string format
+ref: print('"{}"'.format(word))
+Output: print ('\ n '. join ('% s = % s'% ( value, ) ) )
+bleu_metric: 16.331948281960493
+----------
+NL: join a list of strings `list` using a space ' '
+ref: """ """.join(list)
+Output: list (''. join ( map ( str, list ) ) )
+bleu_metric: 33.918247054808944
+----------
+NL: create list `y` containing two empty lists
+ref: y = [[] for n in range(2)]
+Output: y = [ ( x, y ) for x in zip ( l1, l2 ) ]
+bleu_metric: 18.951629567590743
+----------
+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: new _ list = [ line. rstrip ('\ t') for line in open ('filename. txt') ]
+bleu_metric: 37.34651115614923
+----------
+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: 19.720950806796015
+----------
+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: " " " a b : b. c '! ='a')
+bleu_metric: 23.234178119893244
+----------
+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: " " " " " ". join ( [ char for char in'it is icy'if char! ='i'] )
+bleu_metric: 100.0
+----------
+NL: drop rows of pandas dataframe `df` having nan in column at index "1"
+ref: df.dropna(subset=[1])
+Output: df. drop ( df. index. duplicated ( )
+bleu_metric: 17.996531271765896
+----------
+NL: get elements from list `mylist`, that have a field `n` value 30
+ref: [x for x in myList if x.n == 30]
+Output: mylist = [ i for i in range ( 10 ) if i % 2 = = 0 ]
+bleu_metric: 11.709061630923735
+----------
+NL: converting list of strings `intstringlist` to list of integer `nums`
+ref: nums = [int(x) for x in intstringlist]
+Output: nums = [ int ( x ) for x in intstringlist ]
+bleu_metric: 100.0
+----------
+NL: convert list of string numbers into list of integers
+ref: map(int, eval(input('Enter the unfriendly numbers: ')))
+Output: [ int ( s ) for s in example _ string. split ( ',') ]
+bleu_metric: 11.433361115787454
+----------
+NL: print "." without newline
+ref: sys.stdout.write('.')
+Output: sys. stdout. write ( '.')
+bleu_metric: 100.0
+----------
+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 ( 1123. 456789, 0 ) )
+bleu_metric: 39.832871551569504
+----------
+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') :
+bleu_metric: 25.1696695878184
+----------
+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') :
+bleu_metric: 95.90965597935381
+----------
+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') :
+bleu_metric: 50.25808484829805
+----------
+NL: plot dataframe `df` without a legend
+ref: df.plot(legend=False)
+Output: df. legend ( ). plot ( kind ='bar')
+bleu_metric: 21.305413619585085
+----------
+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: 192. setsockopt ('192. 168. 1. 1 ','192. 168. 1. 255')
+bleu_metric: 13.252382555750852
+----------
+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: 192. setsockopt ('192. 168. 1. 1 ','192. 168. 1. 255')
+bleu_metric: 5.334065332387164
+----------
+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: 192. setsockopt ('192. 168. 1. 1 ','192. 168. 1. 255')
+bleu_metric: 61.58601992013119
+----------
+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: float ( [ x for x in list ( d. values ( ) ) ] )
+bleu_metric: 11.156857921595945
+----------
+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: result = [ line. split ( ) for line in open ('filename. txt ','r') ]
+bleu_metric: 7.848329424931549
+----------
+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: [ a for b in c for a in list _ of _ b ]
+bleu_metric: 8.372912227572577
+----------
+NL: print new line character as `\n` in a string `foo\nbar`
+ref: print('foo\nbar'.encode('string_escape'))
+Output: print ('\ n '. join ( chr ( i ) for i in text ) )
+bleu_metric: 19.268479640608692
+----------
+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, y )
+bleu_metric: 8.67666492910357
+----------
+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: np. array ( [ [ 2, 3, 4 ], [ 2, 5, 6 ] ] )
+bleu_metric: 9.410489957951636
+----------
+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: [ k for k, v in mylist if v. startswith ('mycount') else v for k, v in mylist
+bleu_metric: 34.06227682100307
+----------
+NL: Insert directory 'apps' into directory `__file__`
+ref: sys.path.insert(1, os.path.join(os.path.dirname(__file__), 'apps'))
+Output: urllib. request. urlopen ('http : / / www. example. com ','/ path')
+bleu_metric: 9.10451572772839
+----------
+NL: modify sys.path for python module `subdir`
+ref: sys.path.append(os.path.join(os.path.dirname(__file__), 'subdir'))
+Output: subprocess. call ( ['start / subdir / bin / 2014 ','inr'] )
+bleu_metric: 8.368553758184737
+----------
+NL: Insert a 'None' value into a SQLite3 table.
+ref: db.execute("INSERT INTO present VALUES('test2', ?, 10)", (None,))
+Output: cursor. execute ('select * from table where column in l')
+bleu_metric: 13.62767317410906
+----------
+NL: flatten list `list_of_menuitems`
+ref: [image for menuitem in list_of_menuitems for image in menuitem]
+Output: [ item for sublist in list _ of _ lists ]
+bleu_metric: 17.194656088289214
+----------
+NL: append elements of a set `b` to a list `a`
+ref: a.extend(b)
+Output: a. extend ( b )
+bleu_metric: 100.0
+----------
+NL: Append elements of a set to a list in Python
+ref: a.extend(list(b))
+Output: a. extend ( b )
+bleu_metric: 46.81294641285289
+----------
+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 _ csv ('filename. txt ', header = false )
+bleu_metric: 14.209004647554735
+----------
+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 \ \ t. txt ', sep ='\ t ', index
+bleu_metric: 37.50344293811855
+----------
+NL: Split a string `x` by last occurrence of character `-`
+ref: print(x.rpartition('-')[0])
+Output: x. split ('-') [ : - 1 ]
+bleu_metric: 43.353400970949984
+----------
+NL: get the last part of a string before the character '-'
+ref: print(x.rsplit('-', 1)[0])
+Output: re. sub ('(? < = - -. ) ','- ', line )
+bleu_metric: 20.776569671187676
+----------
+NL: upload file using FTP
+ref: ftp.storlines('STOR ' + filename, open(filename, 'r'))
+Output: ftp. quit ( )
+bleu_metric: 3.710176016036221
+----------
+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 ('window. scrollto ( 0, document. body. com') )
+bleu_metric: 10.807499589580493
+----------
+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. array ( [ [ 1, 2, 3 ], [ 4, 5, 6 ] ] )
+bleu_metric: 28.489318277723964
+----------
+NL: print a list `l` and move first 3 elements to the end of the list
+ref: print(l[3:] + l[:3])
+Output: print ( [ l [ 3 : ] for l in l ] )
+bleu_metric: 43.472087194499146
+----------
+NL: loop over files in directory '.'
+ref: for fn in os.listdir('.'):
+ if os.path.isfile(fn):
+ pass
+Output: for ( root, dirs, dirs ) in os. walk ( '.') : if os. path. isdir ( '
+bleu_metric: 46.596572764176955
+----------
+NL: loop over files in directory `source`
+ref: for (root, dirs, filenames) in os.walk(source):
+ for f in filenames:
+ pass
+Output: os. listdir ('os')
+bleu_metric: 4.9782661836656885
+----------
+NL: create a random list of integers
+ref: [int(1000 * random.random()) for i in range(10000)]
+Output: random. choice ( [ int ( i ) for i in range ( 10 ) ] )
+bleu_metric: 38.95592996402149
+----------
+NL: Using %f with strftime() in Python to get microseconds
+ref: datetime.datetime.now().strftime('%H:%M:%S.%f')
+Output: datetime. datetime. now ( ). strftime ('% h : % m : % s. % f')
+bleu_metric: 69.80972118841675
+----------
+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: c. objects. filter ( id = 1 ). order _ by ('foobar')
+bleu_metric: 10.30143863225941
+----------
+NL: filter rows in pandas starting with alphabet 'f' using regular expression.
+ref: df.b.str.contains('^f')
+Output: df ['a']. str. contains ('c')
+bleu_metric: 43.85068972747104
+----------
+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: print ('\ n '. join ( map ( str, row ) ) )
+bleu_metric: 28.575895973659073
+----------
+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. drop ( df. index. map ( lambda x : x. index, axis = 1 ) )
+bleu_metric: 13.21619418680914
+----------
+NL: format the variables `self.goals` and `self.penalties` using string formatting
+ref: """({:d} goals, ${:d})""".format(self.goals, self.penalties)
+Output: " " " { } \ \ w { }. { } " " ". format ( date,'user')
+bleu_metric: 28.34142738530324
+----------
+NL: format string "({} goals, ${})" with variables `goals` and `penalties`
+ref: """({} goals, ${})""".format(self.goals, self.penalties)
+Output: " " " { } { } " " ". format ( {'foo':'bar'} )
+bleu_metric: 34.620457601133026
+----------
+NL: format string "({0.goals} goals, ${0.penalties})"
+ref: """({0.goals} goals, ${0.penalties})""".format(self)
+Output: df. set _ index ('name')
+bleu_metric: 3.9504404329436813
+----------
+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: 18.761335250609946
+----------
+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: map ( int, l )
+bleu_metric: 2.5915324112223583
+----------
+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: 11.473780780974305
+----------
+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 '. join ( map ( str, mylist ) ) )
+bleu_metric: 60.5478327168454
+----------
+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 [ x if x! ='b'] if x! ='c']
+bleu_metric: 20.792998340455547
+----------
+NL: Remove duplicate words from a string `text` using regex
+ref: text = re.sub('\\b(\\w+)( \\1\\b)+', '\\1', text)
+Output: re. sub ('(. * ) \ \ 1 + ','\ \ 1 ', text )
+bleu_metric: 42.691904454956294
+----------
+NL: count non zero values in each column in pandas data frame
+ref: df.astype(bool).sum(axis=1)
+Output: df. groupby ( level = 0 ). sum ( )
+bleu_metric: 33.17838167503153
+----------
+NL: search for string that matches regular expression pattern '(?.*<' in xml string `line`
+ref: print(re.search('>.*<', line).group(0))
+Output: print ( doc. group ( 1 ). group ( 1 ) )
+bleu_metric: 22.08517793512327
+----------
+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 ('2010 - 11 - 13 10 : 33 : 54. 227806 ','% y
+bleu_metric: 23.69248624414322
+----------
+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: [ x [ 0 ] for x in l if x [ 0 ] = ='score']
+bleu_metric: 32.08521724608534
+----------
+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 ('[ ^ \ \ w. ] + ','', text ). strip ( )
+bleu_metric: 54.08690470924506
+----------
+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 ] + ','', s ). strip ( )
+bleu_metric: 38.80387619814469
+----------
+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. plot ( x, y, label ='h \ u2082o')
+bleu_metric: 72.76817202342092
+----------
+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. plot ( x, y, label ='h \ u2082o')
+bleu_metric: 71.60350546947922
+----------
+NL: loop over a list `mylist` if sublists length equals 3
+ref: [x for x in mylist if len(x) == 3]
+Output: mylist [ : : 2 ] = = mylist [ 0 ]
+bleu_metric: 12.640504935009755
+----------
+NL: initialize a list `lst` of 100 objects Object()
+ref: lst = [Object() for _ in range(100)]
+Output: lst = [ [ ] for i in range ( 3 ) ]
+bleu_metric: 28.992166587078206
+----------
+NL: create list `lst` containing 100 instances of object `Object`
+ref: lst = [Object() for i in range(100)]
+Output: list ( chain. from _ iterable ( l ) )
+bleu_metric: 10.649638629044373
+----------
+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: soup. find _ elements _ by _ xpath ( " / / * [ @ type ='submit'] " ). get _ attribute (
+bleu_metric: 10.389889377420038
+----------
+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, on = ['date _ time ','date _ time'] )
+bleu_metric: 21.22363344155404
+----------
+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 ( str ( a ) +'' + str ( b ) )
+bleu_metric: 6.053698751871955
+----------
+NL: Split a string by a delimiter in python
+ref: [x.strip() for x in '2.MATCHES $$TEXT$$ STRING'.split('$$TEXT$$')]
+Output: s. split (';')
+bleu_metric: 2.904501569386027
+----------
+NL: check if directory `directory ` exists and create it if necessary
+ref: if (not os.path.exists(directory)):
+ os.makedirs(directory)
+Output: os. makedirs ( path )
+bleu_metric: 9.597952472216766
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: os.makedirs(path)
+Output: os. makedirs ( path )
+bleu_metric: 100.0
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: distutils.dir_util.mkpath(path)
+Output: os. makedirs ( path )
+bleu_metric: 40.661103887968814
+----------
+NL: check if a directory `path` exists and create it if necessary
+ref: os.makedirs(path)
+Output: os. makedirs ( path )
+bleu_metric: 100.0
+----------
+NL: Replace a separate word 'H3' by 'H1' in a string 'text'
+ref: re.sub('\\bH3\\b', 'H1', text)
+Output: re. sub ('\ \ w + ','\ \ 1 ', text )
+bleu_metric: 56.31071891586576
+----------
+NL: substitute ASCII letters in string 'aas30dsa20' with empty string ''
+ref: re.sub('\\D', '', 'aas30dsa20')
+Output: re. sub ('[ a - z ] ','','abcd')
+bleu_metric: 51.003234295212685
+----------
+NL: get digits only from a string `aas30dsa20` using lambda function
+ref: """""".join([x for x in 'aas30dsa20' if x.isdigit()])
+Output: " " " " " ". join ( map ( lambda x : x * 7 ) )
+bleu_metric: 39.54603942707659
+----------
+NL: access a tag called "name" in beautifulsoup `soup`
+ref: print(soup.find('name').string)
+Output: soup. findall ('div ', {'name':'description'} )
+bleu_metric: 17.457404828304572
+----------
+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 ( ( x [ 0 ], x [ 1 ] ) for x in cursor. items ( ) )
+bleu_metric: 16.34453553468014
+----------
+NL: Create new matrix object by concatenating data from matrix A and matrix B
+ref: np.concatenate((A, B))
+Output: numpy. concatenate ( [ a, b ], axis = 0 )
+bleu_metric: 18.92240568795935
+----------
+NL: concat two matrices `A` and `B` in numpy
+ref: np.vstack((A, B))
+Output: np. sqrt ( ( a. shape [ 0 ], b ) )
+bleu_metric: 19.923405658137924
+----------
+NL: Get the characters count in a file `filepath`
+ref: os.stat(filepath).st_size
+Output: print ( os. path. join ( get _ list ) )
+bleu_metric: 13.834368456410944
+----------
+NL: count the occurrences of item "a" in list `l`
+ref: l.count('a')
+Output: [ x. count ('a') for x in l ]
+bleu_metric: 43.78826865860793
+----------
+NL: count the occurrences of items in list `l`
+ref: Counter(l)
+Output: [ sum ( x ) for x in l ]
+bleu_metric: 13.065113298388567
+----------
+NL: count the occurrences of items in list `l`
+ref: [[x, l.count(x)] for x in set(l)]
+Output: [ sum ( x ) for x in l ]
+bleu_metric: 16.443715448412245
+----------
+NL: count the occurrences of items in list `l`
+ref: dict(((x, l.count(x)) for x in set(l)))
+Output: [ sum ( x ) for x in l ]
+bleu_metric: 15.927159921457768
+----------
+NL: count the occurrences of item "b" in list `l`
+ref: l.count('b')
+Output: [ ( x, l. count ( x ) ) for x in l ]
+bleu_metric: 20.61477352156375
+----------
+NL: copy file `srcfile` to directory `dstdir`
+ref: shutil.copy(srcfile, dstdir)
+Output: os. chdir ('copyfile. txt ', dst )
+bleu_metric: 14.949751774990682
+----------
+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 ['x'], x ['x'] ) )
+bleu_metric: 9.186672262412122
+----------
+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: ( k for k, v in x. items ( ) if v! = 0 )
+bleu_metric: 100.0
+----------
+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 ( x, key = lambda x : ( x [ 1 ], x [ 0 ] ) )
+bleu_metric: 11.162938527547608
+----------
+NL: Put the curser at beginning of the file
+ref: file.seek(0)
+Output: os. system ('with open ( filename ). read ( ) ','rb')
+bleu_metric: 8.687475782716618
+----------
+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 ['a']. apply ( lambda x : x ['b']. fillna
+bleu_metric: 43.50703089837015
+----------
+NL: remove key 'ele' from dictionary `d`
+ref: del d['ele']
+Output: d = { k : v for k, v in list ( d. items ( ) ) if v! ='none'}
+bleu_metric: 5.442133807846856
+----------
+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: mymodel. objects. filter ( date _ _ length ='% y - % m - % d')
+bleu_metric: 12.351156662075915
+----------
+NL: merge list `['it']` and list `['was']` and list `['annoying']` into one list
+ref: ['it'] + ['was'] + ['annoying']
+Output: [ dict ( zip ( * [ ['one ','two'], ['one ','two'] ] ) )
+bleu_metric: 10.105385745342533
+----------
+NL: increment a value with leading zeroes in a number `x`
+ref: str(int(x) + 1).zfill(len(x))
+Output: print ('\ xe, { 0 : x. 0 } '. format ( x, - 1 ) )
+bleu_metric: 13.471766853689124
+----------
+NL: check if a pandas dataframe `df`'s index is sorted
+ref: all(df.index[:-1] <= df.index[1:])
+Output: df. index. values. tolist ( )
+bleu_metric: 9.980212030371455
+----------
+NL: Convert tuple `t` to list
+ref: list(t)
+Output: list ( t )
+bleu_metric: 100.0
+----------
+NL: Convert list `t` to tuple
+ref: tuple(l)
+Output: tuple ( l )
+bleu_metric: 100.0
+----------
+NL: Convert tuple `level1` to list
+ref: level1 = map(list, level1)
+Output: [ ( x + 1 ) for x in a ]
+bleu_metric: 11.35935489027116
+----------
+NL: send the output of pprint object `dataobject` to file `logFile`
+ref: pprint.pprint(dataobject, logFile)
+Output: sys. stdout. flush ( )
+bleu_metric: 18.67587389639562
+----------
+NL: get index of rows in column 'BoolCol'
+ref: df.loc[df['BoolCol']]
+Output: df [ ['bool'] ]. index ('matrix2'). index
+bleu_metric: 18.710158230410627
+----------
+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. index. map ( lambda x : x [ 1 ]. isin ( x ) ) ]
+bleu_metric: 14.72821272412462
+----------
+NL: get list of indexes of rows where column 'BoolCol' values match True
+ref: df[df['BoolCol'] == True].index.tolist()
+Output: [ df ['ratio'] for d in l ]
+bleu_metric: 19.62812256638541
+----------
+NL: get index of rows in dataframe `df` which column 'BoolCol' matches value True
+ref: df[df['BoolCol']].index.tolist()
+Output: df ['column _ name'] = = df. index. str. get ('columnid')
+bleu_metric: 18.088201962129936
+----------
+NL: change working directory to the directory `owd`
+ref: os.chdir(owd)
+Output: os. chdir ('c : \ \ users \ \ users \ \ desktop \ \ desktop \ \ python')
+bleu_metric: 15.553986407412198
+----------
+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 ) ', ( ast, ) )
+bleu_metric: 34.29547961820059
+----------
+NL: decode string "\\x89\\n" into a normal string
+ref: """\\x89\\n""".decode('string_escape')
+Output: " " " \ \ xc3 \ \ x96 \ xc3 \ x96 \ x93 \ x93 \
+bleu_metric: 29.888108815158
+----------
+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: 35.556702356686955
+----------
+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: 30.620490882364876
+----------
+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 ('\ \ d + (? : \ \ s + ) ','a \ n b')
+bleu_metric: 12.792208459604188
+----------
+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: soup. find _ all ('div ', id ='all')
+bleu_metric: 46.28692516962493
+----------
+NL: filter rows containing key word `ball` in column `ids`
+ref: df[df['ids'].str.contains('ball')]
+Output: id. idxmax ( word ). drop ( false )
+bleu_metric: 7.630799524006417
+----------
+NL: convert index at level 0 into a column in dataframe `df`
+ref: df.reset_index(level=0, inplace=True)
+Output: df. set _ index ('level1'). unstack ( level = 0 )
+bleu_metric: 24.29335519279861
+----------
+NL: Add indexes in a data frame `df` to a column `index1`
+ref: df['index1'] = df.index
+Output: df. loc [ index ] = df. index. get _ level _ values ( 0 )
+bleu_metric: 24.04315522172745
+----------
+NL: convert pandas index in a dataframe to columns
+ref: df.reset_index(level=['tick', 'obs'])
+Output: df. reset _ index ( drop = true )
+bleu_metric: 11.884004683899853
+----------
+NL: Get reverse of list items from list 'b' using extended slicing
+ref: [x[::-1] for x in b]
+Output: [ [ i [ 0 ] for i in range ( len ( x ) ) ] for i in range ( 0, len ( list ( a )
+bleu_metric: 7.434360163166209
+----------
+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: b = np. where ( a = = 0, axis = 0 )
+bleu_metric: 9.027924728401977
+----------
+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. ravel ( ), b. ravel ( ) ), dtype ='i4, i4
+bleu_metric: 67.03200460356393
+----------
+NL: convert list `list_of_ints` into a comma separated string
+ref: """,""".join([str(i) for i in list_of_ints])
+Output: list ( map ( int, list _ of _ lists. split ( ',') ) )
+bleu_metric: 13.283261562346366
+----------
+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 = {'username':'username'} )
+bleu_metric: 37.18698903847989
+----------
+NL: Find last occurrence of character '}' in string "abcd}def}"
+ref: 'abcd}def}'.rfind('}')
+Output: re. findall ('(? : ^ [ a - z ] * ) ','abcd0')
+bleu_metric: 12.962472880491877
+----------
+NL: Iterate ove list `[1, 2, 3]` using list comprehension
+ref: print([item for item in [1, 2, 3]])
+Output: [ [ 1, 2, 3 ], [ 4, 5, 6 ], [ 7, 8, 9 ] ]
+bleu_metric: 27.073362211548464
+----------
+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'] for d in x if'y'in d ]
+bleu_metric: 37.60857585579953
+----------
+NL: get the filename without the extension from file 'hemanth.txt'
+ref: print(os.path.splitext(os.path.basename('hemanth.txt'))[0])
+Output: print ( glob. glob ('liname. txt') [ 0 ] )
+bleu_metric: 31.18776276497533
+----------
+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 ( ( k, v ) for k, v in zip ( a, b ) )
+bleu_metric: 12.152094584183624
+----------
+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'], ['a ','b ','c
+bleu_metric: 17.7080331882364
+----------
+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 ['column _ name']. isin ( some _ values ('one ','two
+bleu_metric: 10.948000684047626
+----------
+NL: replace all occurences of newlines `\n` with `
` in dataframe `df`
+ref: df.replace({'\n': '
'}, regex=True)
+Output: df. replace ('\ n ','\ n')
+bleu_metric: 27.770641036119947
+----------
+NL: replace all occurrences of a string `\n` by string `
` in a pandas data frame `df`
+ref: df.replace({'\n': '
'}, regex=True)
+Output: df. replace ( {'\ n':'< br >'}, regex = true )
+bleu_metric: 89.21616972156079
+----------
+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: [ ('a ','b'), ('b ','c'), ('c ','d')
+bleu_metric: 8.840994001530046
+----------
+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: [ str ( x ) for x in words ]
+bleu_metric: 3.8733243551839442
+----------
+NL: extract a url from a string `myString`
+ref: print(re.findall('(https?://[^\\s]+)', myString))
+Output: print ( re. search ('( https? : / / [ ^ \ \ s ] + ) ', mystring ) )
+bleu_metric: 77.28607577370117
+----------
+NL: extract a url from a string `myString`
+ref: print(re.search('(?Phttps?://[^\\s]+)', myString).group('url'))
+Output: print ( re. search ('( https? : / / [ ^ \ \ s ] + ) ', mystring ) )
+bleu_metric: 51.916231800616785
+----------
+NL: remove all special characters, punctuation and spaces from a string `mystring` using regex
+ref: re.sub('[^A-Za-z0-9]+', '', mystring)
+Output: re. sub ('\ \ w + (? = [ ^. ] * ) ','', mystring )
+bleu_metric: 45.91497693322866
+----------
+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: df. groupby ('date'). apply ( lambda x : x. year )
+bleu_metric: 8.453429635655233
+----------
+NL: Create multidimensional array `matrix` with 3 rows and 2 columns in python
+ref: matrix = [[a, b], [c, d], [e, f]]
+Output: matrix = np. concatenate ( ( [ a, b ], [ [ 0, 3, 6 ], [ 7, 8, 9
+bleu_metric: 32.24199399071415
+----------
+NL: replace spaces with underscore
+ref: mystring.replace(' ', '_')
+Output: re. sub ('\ \ b (? = \ \ w + ) ','','this is a string ','replace
+bleu_metric: 14.117915208981241
+----------
+NL: get an absolute file path of file 'mydir/myfile.txt'
+ref: os.path.abspath('mydir/myfile.txt')
+Output: os. path. join ( os. path. join ( root, myfile ) )
+bleu_metric: 23.83041256525615
+----------
+NL: split string `my_string` on white spaces
+ref: """ """.join(my_string.split())
+Output: " " " 42, " " ". split ( )
+bleu_metric: 47.72263780018156
+----------
+NL: get filename without extension from file `filename`
+ref: os.path.splitext(filename)[0]
+Output: os. path. splitext ( filename ) [ 0 ]
+bleu_metric: 100.0
+----------
+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 ( x for x in l )
+bleu_metric: 6.924459302580939
+----------
+NL: split a string `Docs/src/Scripts/temp` by `/` keeping `/` in the result
+ref: """Docs/src/Scripts/temp""".replace('/', '/\x00/').split('\x00')
+Output: re. split ('\ \ s + ', string )
+bleu_metric: 7.926805970774756
+----------
+NL: shuffle columns of an numpy array 'r'
+ref: np.random.shuffle(np.transpose(r))
+Output: np. random. shuffle ( np. transpose ( r ) )
+bleu_metric: 100.0
+----------
+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 ['c']. apply ( lambda x : x. tolist ( ), axis =
+bleu_metric: 26.492685902784498
+----------
+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: print ( json. dumps ( ['b ','c'] ) )
+bleu_metric: 5.979778160388024
+----------
+NL: check characters of string `string` are true predication of function `predicate`
+ref: all(predicate(x) for x in string)
+Output: print ( any ( test _ string ) for substring in string _ string )
+bleu_metric: 19.923405658137924
+----------
+NL: determine number of files on a drive with python
+ref: os.statvfs('/').f_files - os.statvfs('/').f_ffree
+Output: os. stat ('/ myfile. txt'). os. listdir ('/')
+bleu_metric: 37.211067993181466
+----------
+NL: how to get a single result from a SQLite query in python?
+ref: cursor.fetchone()[0]
+Output: cursor. fetchone ( ) [ 0 ]
+bleu_metric: 100.0
+----------
+NL: convert string `user_input` into a list of integers `user_list`
+ref: user_list = [int(number) for number in user_input.split(',')]
+Output: [ int ( x ) for x in input ( ). split ( ',') ]
+bleu_metric: 51.23350305765596
+----------
+NL: Get a list of integers by splitting a string `user` with comma
+ref: [int(s) for s in user.split(',')]
+Output: map ( int,'re. split ('( \ \ d + ) ','hello world') )
+bleu_metric: 20.80315522738391
+----------
+NL: Sorting a Python list by two criteria
+ref: sorted(list, key=lambda x: (x[0], -x[1]))
+Output: sorted ( list ( range ( 0, len ( s ) ), key = lambda x : ( x [ 0 ], x [ 0 ] )
+bleu_metric: 48.440807796000044
+----------
+NL: sort a list of objects `ut`, based on a function `cmpfun` in descending order
+ref: ut.sort(key=cmpfun, reverse=True)
+Output: sorted ( l, key = lambda x : x. modified, reverse = true )
+bleu_metric: 18.107197870881738
+----------
+NL: reverse list `ut` based on the `count` attribute of each object
+ref: ut.sort(key=lambda x: x.count, reverse=True)
+Output: ['city ','same2 ','type']
+bleu_metric: 7.453337029678027
+----------
+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 ( l, key = lambda x : x. count ('c') )
+bleu_metric: 45.34010611888324
+----------
+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 inmail'). click ( )
+bleu_metric: 35.13874939965221
+----------
+NL: click a href button having text `Send InMail` with selenium
+ref: driver.findElement(By.linkText('Send InMail')).click()
+Output: driver. find _ element _ by _ link _ text ('send inmail')
+bleu_metric: 15.175278640994005
+----------
+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 _ link _ text ('send')
+bleu_metric: 15.27678022374703
+----------
+NL: cast an int `i` to a string and concat to string 'ME'
+ref: 'ME' + str(i)
+Output: str ( i )
+bleu_metric: 36.787944117144235
+----------
+NL: Sorting data in DataFrame Pandas
+ref: df.sort_values(['System_num', 'Dis'])
+Output: df. sort _ values ( ['system _ id ','dis'] )
+bleu_metric: 31.671615012203976
+----------
+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 ('test. txt ','w'). write ('\ n')
+bleu_metric: 36.176855814312944
+----------
+NL: sort a list `l` by length of value in tuple
+ref: l.sort(key=lambda t: len(t[1]), reverse=True)
+Output: l. sort ( key = lambda x : ( len ( x ), x ) )
+bleu_metric: 37.875123684341865
+----------
+NL: split string `s` by words that ends with 'd'
+ref: re.findall('\\b(\\w+)d\\b', s)
+Output: re. split ('( \ \ d + ) ', s )
+bleu_metric: 29.23155541924467
+----------
+NL: return `True` if string `foobarrrr` contains regex `ba[rzd]`
+ref: bool(re.search('ba[rzd]', 'foobarrrr'))
+Output: pattern = re. compile ('\ \ b [ a - za - z0 - 9 - ]')
+bleu_metric: 12.384901282810546
+----------
+NL: Removing duplicates in list `t`
+ref: list(set(t))
+Output: list ( set ( t ) )
+bleu_metric: 100.0
+----------
+NL: Removing duplicates in list `source_list`
+ref: list(set(source_list))
+Output: list ( set ( t ) )
+bleu_metric: 59.69491792019645
+----------
+NL: Removing duplicates in list `abracadabra`
+ref: list(OrderedDict.fromkeys('abracadabra'))
+Output: [ ('a ','b'), ('b ','c'), ('a ','d')
+bleu_metric: 7.555283720961675
+----------
+NL: Convert array `a` into a list
+ref: numpy.array(a).reshape(-1).tolist()
+Output: list ( a )
+bleu_metric: 5.237603475197421
+----------
+NL: Convert the first row of numpy matrix `a` to a list
+ref: numpy.array(a)[0].tolist()
+Output: numpy. apply _ along _ axis ( numpy. array ( [ 0, 1, 2 ] ), a )
+bleu_metric: 20.97823227594146
+----------
+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 _ id ('< id >'). text
+bleu_metric: 12.033574792347215
+----------
+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 ( x ) for x in l ]
+bleu_metric: 17.41466588478686
+----------
+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: " " " " " ". join ( [ x for x in l if x [ 0 ] in l ] )
+bleu_metric: 36.45674601639763
+----------
+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 : / / www. google. com')
+bleu_metric: 19.209534151258666
+----------
+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: [ i for i in teststr if re. search ('\ \ d + [ ^ \ \ ] * \ \ ] ', i )
+bleu_metric: 69.89029944247109
+----------
+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 [ ( df ['a'] > 1 ) & ( df ['b'] < = 5 ) ]
+bleu_metric: 39.042104382208294
+----------
+NL: Sort dictionary `o` in ascending order based on its keys and items
+ref: sorted(o.items())
+Output: sorted ( list ( dictionary. items ( ) ), key = lambda x : x [ 1 ] )
+bleu_metric: 24.42195895839763
+----------
+NL: get sorted list of keys of dict `d`
+ref: sorted(d)
+Output: sorted ( d, key = d. get )
+bleu_metric: 25.21193618434983
+----------
+NL: How to sort dictionaries by keys in Python
+ref: sorted(d.items())
+Output: sorted ( d. items ( ) )
+bleu_metric: 100.0
+----------
+NL: convert string "1" into integer
+ref: int('1')
+Output: int ('1')
+bleu_metric: 100.0
+----------
+NL: function to convert strings into integers
+ref: int()
+Output: map ( int,'1')
+bleu_metric: 21.105340631872636
+----------
+NL: convert items in `T1` to integers
+ref: T2 = [map(int, x) for x in T1]
+Output: int ('\ n '. join ( str ( i ) for i in str ( n ) ) )
+bleu_metric: 8.607692533178168
+----------
+NL: call a shell script `./test.sh` using subprocess
+ref: subprocess.call(['./test.sh'])
+Output: subprocess. call ( ['test. sh ','/ test. sh otherfunc'] )
+bleu_metric: 53.66411241731205
+----------
+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 in l2 for y in zip ( l1, l2 ) ]
+bleu_metric: 38.146366463982886
+----------
+NL: encode string 'data to be encoded'
+ref: encoded = base64.b64encode('data to be encoded')
+Output: data. encode ('utf8')
+bleu_metric: 19.260282250359822
+----------
+NL: encode a string `data to be encoded` to `ascii` encoding
+ref: encoded = 'data to be encoded'.encode('ascii')
+Output: data. encode ('ascii')
+bleu_metric: 41.51380951089673
+----------
+NL: parse tab-delimited CSV file 'text.txt' into a list
+ref: lol = list(csv.reader(open('text.txt', 'rb'), delimiter='\t'))
+Output: txt = open ('myfile. txt ','r'). read ( )
+bleu_metric: 20.692050735338164
+----------
+NL: Get attribute `my_str` of object `my_object`
+ref: getattr(my_object, my_str)
+Output: getattr ( my _ obj,'id')
+bleu_metric: 19.72940627795883
+----------
+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: values = [ dict ( k, v ) for k, v in list ( d. items ( ) ) ]
+bleu_metric: 17.057966110550765
+----------
+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 ( [ item [ 0 ] for item in tuple ] )
+bleu_metric: 41.37441202051881
+----------
+NL: convert unicode string u"{'code1':1,'code2':1}" into dictionary
+ref: d = ast.literal_eval("{'code1':1,'code2':1}")
+Output: print ('\ u0420 \ u043e \ u0441 \ u0438 \ u0438 \ u
+bleu_metric: 6.351019285070924
+----------
+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 ('\ \ b [ ^ \ \ w \ \ ] ', mystring )
+bleu_metric: 9.850775379403402
+----------
+NL: remove any url within string `text`
+ref: text = re.sub('^https?:\\/\\/.*[\\r\\n]*', '', text, flags=re.MULTILINE)
+Output: re. sub ('\ \. \ \ blog _ addr \ \ r \ \ n ','', text )
+bleu_metric: 32.17551945847391
+----------
+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. in1d ( a, b ) ) [ 0 ]
+bleu_metric: 22.828499613963487
+----------
+NL: calculate mean across dimension in a 2d array `a`
+ref: np.mean(a, axis=1)
+Output: np. mean ( a, axis = 1 )
+bleu_metric: 100.0
+----------
+NL: running r script '/pathto/MyrScript.r' from python
+ref: subprocess.call(['/usr/bin/Rscript', '--vanilla', '/pathto/MyrScript.r'])
+Output: sys. path. append ('/ path / to / my / my / mypath')
+bleu_metric: 7.04483155666154
+----------
+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. path. join ( root, '. / main _ folder')
+bleu_metric: 6.726052538373431
+----------
+NL: add a header to a csv file
+ref: writer.writeheader()
+Output: writer. writeheader ( )
+bleu_metric: 100.0
+----------
+NL: replacing nan in the dataframe `df` with row average
+ref: df.fillna(df.mean(axis=1), axis=1)
+Output: df. fillna ( df. mean ( axis = 1 ), axis = 1 )
+bleu_metric: 100.0
+----------
+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: datetime. strptime ('21 / 11 / 06 16 : 30 ','% d / % m / % y')
+bleu_metric: 10.247261229308657
+----------
+NL: Call a base class's class method `do` from derived class `Derived`
+ref: super(Derived, cls).do(a)
+Output: super ( childclass, self ). _ _ init _ _ ( * args, * * kwargs )
+bleu_metric: 11.96655750514248
+----------
+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. arange ( 3 ), ( 0, 1, 0 ) ]
+bleu_metric: 13.581342454223687
+----------
+NL: separate words delimited by one or more spaces into a list
+ref: re.split(' +', 'hello world sample text')
+Output: ['one ','two ','one ','two']
+bleu_metric: 13.78592993183041
+----------
+NL: length of longest element in list `words`
+ref: len(max(words, key=len))
+Output: len ( max ( words, key = len ) )
+bleu_metric: 100.0
+----------
+NL: get the value associated with unicode key 'from_user' of first dictionary in list `result`
+ref: result[0]['from_user']
+Output: ['key1 ','key2 ','value']
+bleu_metric: 18.20705281109213
+----------
+NL: Retrieve each line from a file 'File.txt' as a list
+ref: [line.split() for line in open('File.txt')]
+Output: print ( line. rstrip ('\ n') )
+bleu_metric: 11.789701057610694
+----------
+NL: swap keys with values in a dictionary `a`
+ref: res = dict((v, k) for k, v in a.items())
+Output: a [ key ] = dict ( ( v ) for k, v in a. items ( ) )
+bleu_metric: 73.78351342269067
+----------
+NL: Open a file `path/to/FILE_NAME.ext` in write mode
+ref: new_file = open('path/to/FILE_NAME.ext', 'w')
+Output: file _ name = os. path. join ( os. path. join ( dir, '.') )
+bleu_metric: 10.713701843513144
+----------
+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'] ) ['col3']. apply ( lambda x :
+bleu_metric: 72.38800804180103
+----------
+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 ( isinstance ( x, int ( dict. values ( ) ) ) for x in dict _ string in input. lower (
+bleu_metric: 8.839374326825924
+----------
+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: [ dict ( zip ( k, v ), v ) for k, v in d. items ( ) if v. startswith ( '
+bleu_metric: 24.14138031982595
+----------
+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: pd. concat ( [ df1, df2 ], axis = 1 )
+bleu_metric: 8.12657367724027
+----------
+NL: print elements of list `list` seperated by tabs `\t`
+ref: print('\t'.join(map(str, list)))
+Output: print ( [ i. partition ('\ t') for i in list _ of _ lists ] )
+bleu_metric: 25.58058053378779
+----------
+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 ('\ xc5 \ xa4 \ xd0 \ x80 \ xd0 \ xb0 \ xd0
+bleu_metric: 29.472590166651962
+----------
+NL: Encode a latin character in string `Sopet\xc3\xb3n` properly
+ref: 'Sopet\xc3\xb3n'.encode('latin-1').decode('utf-8')
+Output: " " " \ xc3 \ xa3 \ x9f " " ". encode ('utf - 8')
+bleu_metric: 46.255815148030386
+----------
+NL: resized image `image` to width, height of `(x, y)` with filter of `ANTIALIAS`
+ref: image = image.resize((x, y), Image.ANTIALIAS)
+Output: img = np. array ( [ 0, 1, 0, 1, 2, - 1, 0, 1, 2 ] )
+bleu_metric: 7.278748444764005
+----------
+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 ('(? < = [ ^ \ \ s ] ) (? = [ ^ \ \ s ] + ) ',
+bleu_metric: 44.74679573461579
+----------
+NL: display the float `1/3*100` as a percentage
+ref: print('{0:.0f}%'.format(1.0 / 3 * 100))
+Output: print ( ('0. 1 ', 1 ) )
+bleu_metric: 11.406542828786534
+----------
+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: l. sort ( key = lambda x : x ['title'] )
+bleu_metric: 100.0
+----------
+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 ( l, key = lambda x : x ['name'] )
+bleu_metric: 17.362697834284052
+----------
+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: [ ( x, y ) for x, y in zip ( l1, l2 ) ]
+bleu_metric: 5.935298261592072
+----------
+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 ( text ='div ', style ='width = 300px ;')
+bleu_metric: 11.970571135993998
+----------
+NL: write records in dataframe `df` to table 'test' in schema 'a_schema'
+ref: df.to_sql('test', engine, schema='a_schema')
+Output: df. plot ( row ='test ','test _ txt')
+bleu_metric: 32.62478546610937
+----------
+NL: Extract brackets from string `s`
+ref: brackets = re.sub('[^(){}[\\]]', '', s)
+Output: re. sub ('\ \ [ ( \ \ w + ) \ \ s + \ \ ] ','', s )
+bleu_metric: 47.1113978575501
+----------
+NL: remove duplicate elements from list 'L'
+ref: list(dict((x[0], x) for x in L).values())
+Output: [ k for k, g in l if'l'in k ]
+bleu_metric: 6.942420857748657
+----------
+NL: read a file `file` without newlines
+ref: [line.rstrip('\n') for line in file]
+Output: file. write ('\ n')
+bleu_metric: 30.98188793972842
+----------
+NL: get the position of item 1 in `testlist`
+ref: [i for (i, x) in enumerate(testlist) if (x == 1)]
+Output: print ( [ item for item in mylist if item. isdigit ( ) ] )
+bleu_metric: 9.243620302376062
+----------
+NL: get the position of item 1 in `testlist`
+ref: [i for (i, x) in enumerate(testlist) if (x == 1)]
+Output: print ( [ item for item in mylist if item. isdigit ( ) ] )
+bleu_metric: 9.243620302376062
+----------
+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 ( [ item for item in mylist if item. isdigit ( ) ] )
+bleu_metric: 7.565780264883953
+----------
+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 ( [ item for item in mylist if item. isdigit ( ) ] )
+bleu_metric: 6.0168435849456285
+----------
+NL: get the position of item 1 in `testlist`
+ref: gen = (i for (i, x) in enumerate(testlist) if (x == 1))
+Output: print ( [ item for item in mylist if item. isdigit ( ) ] )
+bleu_metric: 8.28910840961223
+----------
+NL: get the position of item `element` in list `testlist`
+ref: print(testlist.index(element))
+Output: print ( testlist. index ( element ) )
+bleu_metric: 100.0
+----------
+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: 7.150780327539586
+----------
+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 ( x [ 0 ] for x in rows )
+bleu_metric: 14.988854080083206
+----------
+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 ( x, key = lambda x : x [ 1 ] ) [ 0 ]
+bleu_metric: 35.27295712700594
+----------
+NL: Make a delay of 1 second
+ref: time.sleep(1)
+Output: time. sleep ( 0. 1 )
+bleu_metric: 52.53819788848316
+----------
+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: obj. save ('new _ name ','default _ number')
+bleu_metric: 10.318679011582763
+----------
+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 ( lst, key = lambda x : x [ 1 ] )
+bleu_metric: 19.143591202259493
+----------
+NL: How do I perform secondary sorting in python?
+ref: sorted(list5, key=lambda vertex: (degree(vertex), vertex))
+Output: sorted ( list5, key = lambda x : ( sum ( x [ 1 ] ), x ) )
+bleu_metric: 41.26152034907945
+----------
+NL: convert a list into a generator object
+ref: (n for n in [1, 2, 3, 5])
+Output: list ( itertools. combinations ( * ( l ) ) )
+bleu_metric: 9.930751066504135
+----------
+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 set ( split _ list [ i ] for i in range ( len ( x ) ) if i not in index ]
+bleu_metric: 18.860270984795903
+----------
+NL: Open a file `yourfile.txt` in write mode
+ref: f = open('yourfile.txt', 'w')
+Output: os. system ('your string. txt')
+bleu_metric: 23.571525044211057
+----------
+NL: get attribute 'attr' from object `obj`
+ref: getattr(obj, 'attr')
+Output: getattr ( obj,'attr')
+bleu_metric: 100.0
+----------
+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 ( ['a ','b ','c ','d'] )
+bleu_metric: 6.101077050680879
+----------
+NL: convert tuple of tuples `(('aa',), ('bb',), ('cc',))` to list in one line
+ref: map(lambda a: a[0], (('aa',), ('bb',), ('cc',)))
+Output: map ( lambda s : ( d ['a'], d ['b'], d ['c'] ), d [
+bleu_metric: 11.200083731986627
+----------
+NL: Python Pandas: How to replace a characters in a column of a dataframe?
+ref: df['range'].replace(',', '-', inplace=True)
+Output: df ['brandname'] = df ['brandname']. replace ('- ', ','in')
+bleu_metric: 45.97006398721532
+----------
+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 ','b ','c ','d'] ) )
+bleu_metric: 36.501102157810664
+----------
+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 ','b ','c ','d'] ) )
+bleu_metric: 36.501102157810664
+----------
+NL: unzip list `original`
+ref: result = ([a for (a, b) in original], [b for (a, b) in original])
+Output: zip ( * list ( range ( 10 ) ) )
+bleu_metric: 4.606184881898074
+----------
+NL: unzip list `original` and return a generator
+ref: result = ((a for (a, b) in original), (b for (a, b) in original))
+Output: [ i for i in range ( 1, len ( l ) ) if i! ='+ 1 ]
+bleu_metric: 7.62604991346693
+----------
+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 ','b ','c ','d ','e'] ) )
+bleu_metric: 34.87995424158042
+----------
+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 ','b'), ('b ','c ','d'), ( '
+bleu_metric: 35.430131828411845
+----------
+NL: encode `Decimal('3.9')` to a JSON string
+ref: json.dumps(Decimal('3.9'))
+Output: json. dumps ( ['1 ','2 ','3'] )
+bleu_metric: 23.58844810653421
+----------
+NL: Add key "mynewkey" to dictionary `d` with value "mynewvalue"
+ref: d['mynewkey'] = 'mynewvalue'
+Output: df. setdefault ('name ', [ ] ). append ( value )
+bleu_metric: 9.50724412002623
+----------
+NL: Add key 'a' to dictionary `data` with value 1
+ref: data.update({'a': 1, })
+Output: a ['a']. append ( {'a': 1,'b': 2 } )
+bleu_metric: 37.863495563972855
+----------
+NL: Add key 'a' to dictionary `data` with value 1
+ref: data.update(dict(a=1))
+Output: a ['a']. append ( {'a': 1,'b': 2 } )
+bleu_metric: 8.24055698798382
+----------
+NL: Add key 'a' to dictionary `data` with value 1
+ref: data.update(a=1)
+Output: a ['a']. append ( {'a': 1,'b': 2 } )
+bleu_metric: 8.24055698798382
+----------
+NL: find maximal value in matrix `matrix`
+ref: max([max(i) for i in matrix])
+Output: max ( matrix, key = lambda x : max ( x ) )
+bleu_metric: 16.943571815930884
+----------
+NL: Round number `answer` to 2 precision after the decimal point
+ref: answer = str(round(answer, 2))
+Output: round ( 1. 0005, 3 )
+bleu_metric: 17.51643270174889
+----------
+NL: extract ip address from an html string
+ref: ip = re.findall('[0-9]+(?:\\.[0-9]+){3}', s)
+Output: print ( re. findall ('[ ^ a - za - z0 - 9 - ] + (? : \ \. [ a -
+bleu_metric: 40.70867848415918
+----------
+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 ( df. filter ( lambda x : x. any ( axis = 1 ) ) )
+bleu_metric: 34.640201116587235
+----------
+NL: append each line in file `myfile` into a list
+ref: [x for x in myfile.splitlines() if x != '']
+Output: myfile. write ('\ n '. join ( [ str ( i ) for i in myfile ] ) )
+bleu_metric: 10.435185733665413
+----------
+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 lst ]
+bleu_metric: 7.020674481218203
+----------
+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. groupby ('text'). apply ( lambda x : x. sum ( ) ). most _ common ( 1 )
+bleu_metric: 21.52643286782715
+----------
+NL: Python split a string using regex
+ref: re.findall('(.+?):(.+?)\\b ?', text)
+Output: re. split ('(? < = \ \ d ) \ \ d + ', text )
+bleu_metric: 29.769819080461428
+----------
+NL: generate all 2-element subsets of tuple `(1, 2, 3)`
+ref: list(itertools.combinations((1, 2, 3), 2))
+Output: list ( itertools. product ( [ 1, 2, 3 ], [ 4, 5, 6 ] ) )
+bleu_metric: 34.29547961820059
+----------
+NL: get a value of datetime.today() in the UTC time zone
+ref: datetime.now(pytz.utc)
+Output: pytz. utcnow ( )
+bleu_metric: 23.206041459353084
+----------
+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 [ : ] = = [ ] ]
+bleu_metric: 69.3395566222006
+----------
+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: list ( zip ( a, b ) )
+bleu_metric: 10.320893749383378
+----------
+NL: Django response with JSON `data`
+ref: return HttpResponse(data, mimetype='application/json')
+Output: return httpresponse ( json. dumps ( response _ data ), content _ type ='application / json')
+bleu_metric: 36.17640392425988
+----------
+NL: get all text that is not enclosed within square brackets in string `example_str`
+ref: re.findall('(.*?)\\[.*?\\]', example_str)
+Output: str1. split ('\ \ ) [ ^ \ \ ] ', str. strip ( )
+bleu_metric: 26.91730507369109
+----------
+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: re. findall ('\ \ [ (? : \ \ w + ) \ \ ] ', str1 )
+bleu_metric: 35.15147667690756
+----------
+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. sub ('\ \ ( [ ^ \ \ w ] ) ','', filename )
+bleu_metric: 27.664786097762494
+----------
+NL: match regex '\\((.*?)\\)|(\\w)' with string '(zyx)bc'
+ref: re.findall('\\((.*?)\\)|(\\w)', '(zyx)bc')
+Output: re. sub ('(? < = \ \ w ) ( \ \ w + ) ( \ \ w + ) ','\ \
+bleu_metric: 30.955387518700363
+----------
+NL: match multiple regex patterns with the alternation operator `|` in a string `(zyx)bc`
+ref: re.findall('\\(.*?\\)|\\w', '(zyx)bc')
+Output: re. findall ('(? < = \ \ w ) ( [ ^ \ \ w \ \ w ] | \ \ w + )
+bleu_metric: 29.552612305356018
+----------
+NL: formate each string cin list `elements` into pattern '%{0}%'
+ref: elements = ['%{0}%'.format(element) for element in elements]
+Output: " " " { 0 :. 2f } " " ". format ('foo ','bar')
+bleu_metric: 17.414675099971177
+----------
+NL: Open a background process 'background-process' with arguments 'arguments'
+ref: subprocess.Popen(['background-process', 'arguments'])
+Output: os. system ('some _ command - some _ file')
+bleu_metric: 12.107470158860224
+----------
+NL: get list of values from dictionary 'mydict' w.r.t. list of keys 'mykeys'
+ref: [mydict[x] for x in mykeys]
+Output: [ key for item in my _ list if item. get ('name ','age') ]
+bleu_metric: 9.119675426861836
+----------
+NL: convert list `[('Name', 'Joe'), ('Age', 22)]` into a dictionary
+ref: dict([('Name', 'Joe'), ('Age', 22)])
+Output: dict ( ( ('class ', 2 ), ('class ', 3 ) ) )
+bleu_metric: 24.080231808676874
+----------
+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. arange ( len ( df. columns ) ), axis = 1 )
+bleu_metric: 18.410768031391818
+----------
+NL: double backslash escape all double quotes in string `s`
+ref: print(s.encode('unicode-escape').replace('"', '\\"'))
+Output: print ( s. split ('\ \') )
+bleu_metric: 21.765088513075124
+----------
+NL: split a string into a list of words and whitespace
+ref: re.split('(\\W+)', s)
+Output: ['this is a string : ','this is a string ','this is a string ','this is a string ',
+bleu_metric: 5.648892151960283
+----------
+NL: plotting stacked barplots on a panda data frame
+ref: df.plot(kind='barh', stacked=True)
+Output: df. plot ( kind ='barh ', stacked = true )
+bleu_metric: 83.7117009877792
+----------
+NL: reverse the keys and values in a dictionary `myDictionary`
+ref: {i[1]: i[0] for i in list(myDictionary.items())}
+Output: dict ( ( k, v ) for k, v in list ( d. items ( ) ) )
+bleu_metric: 23.690464140247087
+----------
+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: ['x ','y ','red ','var ','var']
+bleu_metric: 6.180827527485682
+----------
+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 ( s, str )
+bleu_metric: 59.46035575013605
+----------
+NL: check if object `o` is a string
+ref: (type(o) is str)
+Output: isinstance ( s, str )
+bleu_metric: 23.813694985189848
+----------
+NL: check if object `o` is a string
+ref: isinstance(o, str)
+Output: isinstance ( s, str )
+bleu_metric: 59.46035575013605
+----------
+NL: check if `obj_to_test` is a string
+ref: isinstance(obj_to_test, str)
+Output: print ( obj. isfile ('attr name') )
+bleu_metric: 13.508625657351411
+----------
+NL: append list `list1` to `list2`
+ref: list2.extend(list1)
+Output: [ list1. append ( list1 ) for a in range ( 3 ) ]
+bleu_metric: 14.879641171245488
+----------
+NL: append list `mylog` to `list1`
+ref: list1.extend(mylog)
+Output: list3. append ( list1 )
+bleu_metric: 29.071536848410968
+----------
+NL: append list `a` to `c`
+ref: c.extend(a)
+Output: a. append ( a )
+bleu_metric: 54.10822690539396
+----------
+NL: append items in list `mylog` to `list1`
+ref: for line in mylog:
+ list1.append(line)
+Output: list3. append ( list1 )
+bleu_metric: 20.917479021833486
+----------
+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: a = [ [ ] for i in range ( 0, len ( a ), [ 3, 4 ] ] )
+bleu_metric: 11.556647985416685
+----------
+NL: Initialize `SECRET_KEY` in flask config with `Your_secret_string `
+ref: app.config['SECRET_KEY'] = 'Your_secret_string'
+Output: app. config ['your _ id'] ='your _ config '
+bleu_metric: 43.343660127583234
+----------
+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. values. tolist ( ), columns = ['a ','b'], axis = 1
+bleu_metric: 35.31494509780309
+----------
+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: print ( list ( enumerate ( my _ list ) ) )
+bleu_metric: 5.9394573438219
+----------
+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 ( ), rotation = 45 )
+bleu_metric: 29.88309778054086
+----------
+NL: remove symbols from a string `s`
+ref: re.sub('[^\\w]', ' ', s)
+Output: re. sub ('[ ^ a - z ] ','', s )
+bleu_metric: 74.83293841345244
+----------
+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. abspath ( _ _ file _ _ ) )
+bleu_metric: 48.431464491050775
+----------
+NL: Find octal characters matches from a string `str` using regex
+ref: print(re.findall("'\\\\[0-7]{1,3}'", str))
+Output: re. findall ('\ \ [ (? < = \ \ d ) \ \ ) ','- ', str )
+bleu_metric: 34.98979283499355
+----------
+NL: split string `input` based on occurrences of regex pattern '[ ](?=[A-Z]+\\b)'
+ref: re.split('[ ](?=[A-Z]+\\b)', input)
+Output: re. findall ('(? < = [ a - z ] ) (? = [ a - z ] ) (? = [ a
+bleu_metric: 16.927318970546587
+----------
+NL: Split string `input` at every space followed by an upper-case letter
+ref: re.split('[ ](?=[A-Z])', input)
+Output: re. split ('\ \ b ', text )
+bleu_metric: 23.369549958807866
+----------
+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: r = requests. get ( url )
+bleu_metric: 13.273006566054132
+----------
+NL: write bytes `bytes_` to a file `filename` in python 3
+ref: open('filename', 'wb').write(bytes_)
+Output: f. write ( struct. pack ('> i ', filename ) )
+bleu_metric: 21.042990347620457
+----------
+NL: get a list from a list `lst` with values mapped into a dictionary `dct`
+ref: [dct[k] for k in lst]
+Output: [ list ( lst [ i ] ) for i in lst ]
+bleu_metric: 28.787878181011266
+----------
+NL: find duplicate names in column 'name' of the dataframe `x`
+ref: x.set_index('name').index.get_duplicates()
+Output: print ( list ( df. index. get _ duplicates ( ) ) )
+bleu_metric: 19.923405658137924
+----------
+NL: truncate float 1.923328437452 to 3 decimal places
+ref: round(1.923328437452, 3)
+Output: round ( 1. 005, 2 )
+bleu_metric: 46.71379777282001
+----------
+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 = operator. itemgetter ( 1 ), reverse = true )
+bleu_metric: 19.26332463060877
+----------
+NL: place the radial ticks in plot `ax` at 135 degrees
+ref: ax.set_rlabel_position(135)
+Output: ax. set _ xticklabels ( [ ] )
+bleu_metric: 21.834177214239062
+----------
+NL: check if path `my_path` is an absolute path
+ref: os.path.isabs(my_path)
+Output: os. path. exists ('my _ path')
+bleu_metric: 31.61487584488943
+----------
+NL: get number of keys in dictionary `yourdict`
+ref: len(list(yourdict.keys()))
+Output: len ( mydict )
+bleu_metric: 11.141275535087015
+----------
+NL: count the number of keys in dictionary `yourdictfile`
+ref: len(set(open(yourdictfile).read().split()))
+Output: sum ( len ( v ) for v in yourdict. values ( ) )
+bleu_metric: 18.767030063813223
+----------
+NL: pandas dataframe get first row of each group by 'id'
+ref: df.groupby('id').first()
+Output: df. groupby ('id'). first ( ). first ( )
+bleu_metric: 75.62722459724604
+----------
+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 [ ['c ','df'] ]. apply ( lambda x : x [ 1 : - 1 ]. split (
+bleu_metric: 19.037861963633805
+----------
+NL: extract attributes 'src="js/([^"]*\\bjquery\\b[^"]*)"' from string `data`
+ref: re.findall('src="js/([^"]*\\bjquery\\b[^"]*)"', data)
+Output: " " " \ \ b " " "
+bleu_metric: 3.6253194206622026
+----------
+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 ( [ 1, 2, 3, 4, 5 ] )
+bleu_metric: 3.255351208100268
+----------
+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 ( ['shutdown ','/ r ','/ tmp / test. py'] )
+bleu_metric: 16.665685159230854
+----------
+NL: reverse a priority queue `q` in python without using classes
+ref: q.put((-n, n))
+Output: t. start ( )
+bleu_metric: 12.151662434083677
+----------
+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: df. groupby ('o') ['value']. first ( )
+bleu_metric: 3.5795004593328636
+----------
+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 - z ] ) ( [ a - z ] ) ','abcd _ string
+bleu_metric: 33.336490646495264
+----------
+NL: Get the length of list `my_list`
+ref: len(my_list)
+Output: len ( my _ list )
+bleu_metric: 30.739407647563223
+----------
+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 _ list )
+bleu_metric: 30.739407647563223
+----------
+NL: Getting the length of `my_string`
+ref: len(my_string)
+Output: len ( my _ string )
+bleu_metric: 30.739407647563223
+----------
+NL: remove escape character from string "\\a"
+ref: """\\a""".decode('string_escape')
+Output: print ('\ n '. join ('[ _ name _ ]') )
+bleu_metric: 15.071676257541073
+----------
+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: a = re. sub ('[ a - za - z0 - 9 ] ','', line )
+bleu_metric: 6.256610657452671
+----------
+NL: remove directory tree '/folder_name'
+ref: shutil.rmtree('/folder_name')
+Output: os. path. dirname ( os. path. join ( dirpath, os. path. dirname ( _ _ file _ _ ) )
+bleu_metric: 6.4718242450883325
+----------
+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: df. apply ( lambda x : x. to _ frame ('new _ column'), axis = 1 )
+bleu_metric: 34.619282106807496
+----------
+NL: reverse sort Counter `x` by values
+ref: sorted(x, key=x.get, reverse=True)
+Output: sorted ( x, key = itemgetter ( 1 ) )
+bleu_metric: 40.540555470323305
+----------
+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: 15.53218858512746
+----------
+NL: append a numpy array 'b' to a numpy array 'a'
+ref: np.vstack((a, b))
+Output: np. concatenate ( ( a, b ) )
+bleu_metric: 76.91605673134586
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: print(concatenate((a, b), axis=0))
+Output: print ( concatenate ( ( a, b ), axis = 1 ) )
+bleu_metric: 82.82477531331041
+----------
+NL: numpy concatenate two arrays `a` and `b` along the second axis
+ref: print(concatenate((a, b), axis=1))
+Output: print ( concatenate ( ( a, b ), axis = 1 ) )
+bleu_metric: 100.0
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: c = np.r_[(a[None, :], b[None, :])]
+Output: print ( concatenate ( ( a, b ), axis = 1 ) )
+bleu_metric: 9.520068110441361
+----------
+NL: numpy concatenate two arrays `a` and `b` along the first axis
+ref: np.array((a, b))
+Output: print ( concatenate ( ( a, b ), axis = 1 ) )
+bleu_metric: 41.70507291634131
+----------
+NL: fetch address information for host 'google.com' ion port 80
+ref: print(socket.getaddrinfo('google.com', 80))
+Output: os. environ ['localhost : \ \ python27 \ \ python']
+bleu_metric: 9.672649511413093
+----------
+NL: add a column 'day' with value 'sat' to dataframe `df`
+ref: df.xs('sat', level='day', drop_level=False)
+Output: df ['value'] = df ['value']. apply ( lambda x : x ['value'] )
+bleu_metric: 8.179879050970094
+----------
+NL: return a 401 unauthorized in django
+ref: return HttpResponse('Unauthorized', status=401)
+Output: return httpresponse ( status = 204 )
+bleu_metric: 15.451666492113134
+----------
+NL: Flask set folder 'wherever' as the default template folder
+ref: Flask(__name__, template_folder='wherever')
+Output: app. run ( debug = true )
+bleu_metric: 16.89983564524028
+----------
+NL: How do I INSERT INTO t1 (SELECT * FROM t2) in SQLAlchemy?
+ref: session.execute('INSERT INTO t1 (SELECT * FROM t2)')
+Output: session. execute ('select * from from t1 ;')
+bleu_metric: 34.9740896964685
+----------
+NL: sort a list of lists 'c2' such that third row comes first
+ref: c2.sort(key=lambda row: row[2])
+Output: sorted ( l, key = lambda x : ( - x [ 1 ], x [ 2 ] ) )
+bleu_metric: 22.169194554222997
+----------
+NL: Sorting a list of lists in Python
+ref: c2.sort(key=lambda row: (row[2], row[1], row[0]))
+Output: [ ( row [ i ], row [ i + 2 ] ) for i in range ( 0, len ( row ), row [ 2 ]
+bleu_metric: 26.189293164591476
+----------
+NL: Sorting a list of lists in Python
+ref: c2.sort(key=lambda row: (row[2], row[1]))
+Output: [ ( row [ i ], row [ i + 2 ] ) for i in range ( 0, len ( row ), row [ 2 ]
+bleu_metric: 22.303080489498637
+----------
+NL: set font `Arial` to display non-ascii characters in matplotlib
+ref: matplotlib.rc('font', **{'sans-serif': 'Arial', 'family': 'sans-serif'})
+Output: ax. set _ ylabel ('y - r ', title ='\ u2022')
+bleu_metric: 6.906418747057865
+----------
+NL: Convert DateTime column 'date' of pandas dataframe 'df' to ordinal
+ref: df['date'].apply(lambda x: x.toordinal())
+Output: df ['date'] = df ['date']. map ( lambda x : x. date ( ) )
+bleu_metric: 51.29497107827519
+----------
+NL: get html source of Selenium WebElement `element`
+ref: element.get_attribute('innerHTML')
+Output: print ( element. get _ attribute ('innerhtml') )
+bleu_metric: 18.69300079996002
+----------
+NL: Get the integer location of a key `bob` in a pandas data frame
+ref: df.index.get_loc('bob')
+Output: df. reset _ index ( )
+bleu_metric: 19.840332506055578
+----------
+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 ('echo $ filename. exe')
+bleu_metric: 12.044064401053362
+----------
+NL: add an item with key 'third_key' and value 1 to an dictionary `my_dict`
+ref: my_dict.update({'third_key': 1})
+Output: my _ dict ['my _ dict'] = var
+bleu_metric: 10.579369505074823
+----------
+NL: declare an array
+ref: my_list = []
+Output: variable = [ ]
+bleu_metric: 75.98356856515926
+----------
+NL: Insert item `12` to a list `my_list`
+ref: my_list.append(12)
+Output: my _ list. append ('{ }, { } '. format ( c. c ) )
+bleu_metric: 12.259149958656423
+----------
+NL: add an entry 'wuggah' at the beginning of list `myList`
+ref: myList.insert(0, 'wuggah')
+Output: mylist = [ 'an ','i ','j']
+bleu_metric: 11.502783619900047
+----------
+NL: convert a hex-string representation to actual bytes
+ref: """\\xF3\\xBE\\x80\\x80""".replace('\\x', '').decode('hex')
+Output: " " " \ \ xc3 \ \ x85あ " " ". encode ('hex')
+bleu_metric: 24.712086993335753
+----------
+NL: select the last column of dataframe `df`
+ref: df[df.columns[-1]]
+Output: df [ df. columns [ - 1 ] ]
+bleu_metric: 100.0
+----------
+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 ['value'] = df. groupby ('value') ['value']. apply ( lambda x : x
+bleu_metric: 17.966302274629616
+----------
+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: 35.61693754184905
+----------
+NL: get the type of `i`
+ref: type(i)
+Output: print ( len ( a ) )
+bleu_metric: 22.679164443904003
+----------
+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 ( ) ['a']
+bleu_metric: 16.784459625186194
+----------
+NL: get the 5th item of a generator
+ref: next(itertools.islice(range(10), 5, 5 + 1))
+Output: next ( ( iter ( range ( n ) ) ) for i in range ( n ) )
+bleu_metric: 18.81478574691708
+----------
+NL: Print a string `word` with string format
+ref: print('"{}"'.format(word))
+Output: print ('\ n '. join ('% s = % s'% ( value, ) ) )
+bleu_metric: 16.331948281960493
+----------
+NL: join a list of strings `list` using a space ' '
+ref: """ """.join(list)
+Output: list (''. join ( map ( str, list ) ) )
+bleu_metric: 33.918247054808944
+----------
+NL: create list `y` containing two empty lists
+ref: y = [[] for n in range(2)]
+Output: y = [ ( x, y ) for x in zip ( l1, l2 ) ]
+bleu_metric: 18.951629567590743
+----------
+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: new _ list = [ line. rstrip ('\ t') for line in open ('filename. txt') ]
+bleu_metric: 37.34651115614923
+----------
+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: 19.720950806796015
+----------
+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: " " " a b : b. c '! ='a')
+bleu_metric: 23.234178119893244
+----------
+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: " " " " " ". join ( [ char for char in'it is icy'if char! ='i'] )
+bleu_metric: 100.0
+----------
+NL: Drop rows of pandas dataframe `df` having NaN in column at index "1"
+ref: df.dropna(subset=[1])
+Output: df. drop ( df. index. duplicated ( )
+bleu_metric: 17.996531271765896
+----------
+NL: get elements from list `myList`, that have a field `n` value 30
+ref: [x for x in myList if x.n == 30]
+Output: mylist = [ i for i in range ( 10 ) if i % 2 = = 0 ]
+bleu_metric: 11.709061630923735
+----------
+NL: converting list of strings `intstringlist` to list of integer `nums`
+ref: nums = [int(x) for x in intstringlist]
+Output: nums = [ int ( x ) for x in intstringlist ]
+bleu_metric: 100.0
+----------
+NL: convert list of string numbers into list of integers
+ref: map(int, eval(input('Enter the unfriendly numbers: ')))
+Output: [ int ( s ) for s in example _ string. split ( ',') ]
+bleu_metric: 11.433361115787454
+----------
+NL: print "." without newline
+ref: sys.stdout.write('.')
+Output: sys. stdout. write ( '.')
+bleu_metric: 100.0
+----------
+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 ( 1123. 456789, 0 ) )
+bleu_metric: 39.832871551569504
+----------
+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') :
+bleu_metric: 45.30799450827704
+----------
+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') :
+bleu_metric: 95.90965597935381
+----------
+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') :
+bleu_metric: 50.25808484829805
+----------
+NL: plot dataframe `df` without a legend
+ref: df.plot(legend=False)
+Output: df. legend ( ). plot ( kind ='bar')
+bleu_metric: 21.305413619585085
+----------
+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: 192. setsockopt ('192. 168. 1. 1 ','192. 168. 1. 255')
+bleu_metric: 13.252382555750852
+----------
+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: 192. setsockopt ('192. 168. 1. 1 ','192. 168. 1. 255')
+bleu_metric: 5.334065332387164
+----------
+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: 192. setsockopt ('192. 168. 1. 1 ','192. 168. 1. 255')
+bleu_metric: 61.58601992013119
+----------
+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: float ( [ x for x in list ( d. values ( ) ) ] )
+bleu_metric: 11.156857921595945
+----------
+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: result = [ line. split ( ) for line in open ('filename. txt ','r') ]
+bleu_metric: 7.848329424931549
+----------
+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: [ a for b in c for a in list _ of _ b ]
+bleu_metric: 8.372912227572577
+----------
+NL: Print new line character as `\n` in a string `foo\nbar`
+ref: print('foo\nbar'.encode('string_escape'))
+Output: print ('\ n '. join ( chr ( i ) for i in text ) )
+bleu_metric: 19.268479640608692
+----------
+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, y )
+bleu_metric: 8.67666492910357
+----------
+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: np. array ( [ [ 2, 3, 4 ], [ 2, 5, 6 ] ] )
+bleu_metric: 9.410489957951636
+----------
+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='