4 examples of 'pandas read csv' in Python

37
def readcsv(filename, header=True):
38
    return pd.read_csv(filename, header=None) if not header else pd.read_csv(filename)

63
def _dataframe_from_csv(reader, delimiter, with_header, skipspace):
64
    """Returns csv data as a pandas Dataframe object"""
65
    sep = delimiter
66
    header = 0
67
    if not with_header:
68
        header = None
69
70
    return pd.read_csv(
71
        reader,
72
        header=header,
73
        sep=sep,
74
        skipinitialspace=skipspace,
75
        encoding='utf-8-sig'
76
    )

10
def read_data():
11
	'''
12
	reads data from csv that I generated from copying and pasting articles from
13
	5 different news sources
14
	'''
15
	wiki = pd.read_csv("data/wikipedia_data.csv")
16
	fox = pd.read_csv("data/fox_data.csv")
17
	npr = pd.read_csv("data/npr_data.csv")
18
	cnn = pd.read_csv("data/cnn_data.csv")
19
	cnn = cnn[cnn['text'].apply(str) != 'nan']
20
	data = pd.concat((wiki[['source', 'text', 'rating']], fox[['source', 'text', 'rating']], 
21
		npr[['source', 'text', 'rating']], cnn[['source', 'text', 'rating']] ), ignore_index=True)
22
	return data

554
def _csv_to_pandas_df(filepath,
555
                      separator=DEFAULT_SEPARATOR,
556
                      quote_char=DEFAULT_QUOTE_CHARACTER,
557
                      escape_char=DEFAULT_ESCAPSE_CHAR,
558
                      contain_headers=True,
559
                      lines_to_skip=0,
560
                      date_columns=None,
561
                      rowIdAndVersionInIndex=True):
562
    test_import_pandas()
563
    import pandas as pd
564
565
    # DATEs are stored in csv as unix timestamp in milliseconds
566
    def datetime_millisecond_parser(milliseconds): return pd.to_datetime(milliseconds, unit='ms', utc=True)
567
568
    if not date_columns:
569
        date_columns = []
570
571
    line_terminator = str(os.linesep)
572
573
    df = pd.read_csv(filepath,
574
                     sep=separator,
575
                     lineterminator=line_terminator if len(line_terminator) == 1 else None,
576
                     quotechar=quote_char,
577
                     escapechar=escape_char,
578
                     header=0 if contain_headers else None,
579
                     skiprows=lines_to_skip,
580
                     parse_dates=date_columns,
581
                     date_parser=datetime_millisecond_parser)
582
    if rowIdAndVersionInIndex and "ROW_ID" in df.columns and "ROW_VERSION" in df.columns:
583
        # combine row-ids (in index) and row-versions (in column 0) to
584
        # make new row labels consisting of the row id and version
585
        # separated by a dash.
586
        zip_args = [df["ROW_ID"], df["ROW_VERSION"]]
587
        if "ROW_ETAG" in df.columns:
588
            zip_args.append(df['ROW_ETAG'])
589
590
        df.index = row_labels_from_id_and_version(zip(*zip_args))
591
        del df["ROW_ID"]
592
        del df["ROW_VERSION"]
593
        if "ROW_ETAG" in df.columns:
594
            del df['ROW_ETAG']
595
596
    return df