4 examples of 'python read tsv' in Python

26
def load_tsv():
27
  import pandas as pd
28
29
  print('loading with pandas')
30
31
  filename = 'mat_cats.tsv'
32
  tmp = pd.read_table(filename, sep='\t', index_col=[0,1], skipinitialspace=True, na_values='NaN', keep_default_na=False, usecols=range(12), skip_blank_lines=True)
33
34
  # filename = 'mat_2cc_1rc.txt'
35
  # tmp = pd.read_table(filename, index_col=[0,1], header=[0,1])
36
37
  mat = tmp.values
38
  row_names = tmp.index.tolist()
39
  col_names = tmp.columns.tolist()
40
41
  print(mat)
42
  print('\nrow_names')
43
  print(row_names)
44
  print(len(row_names))
45
  print('\ncol names')
46
  print(col_names)
47
  print(len(col_names))

52
def tsvreader(file):
53
    reader = csv.reader(file, delimiter='\t')
54
    for row in reader:
55
        yield row

32
def parse_tsv(tsv_fpath):
33
    """ Parse a tsv ';' seperated file which was downloaded from vizier
34
    """
35
    out_dict = {'data_lists':{},
36
                'data_units_dict':{}}
37
    icol_to_colname = {}
38
    lines = open(tsv_fpath)
39
    i_signif_line = 0
40
    for line_raw in lines:
41
        line = line_raw.strip()
42
        if len(line) == 0:
43
            continue
44
        elif line[0] == '#':
45
            continue
46
        elif i_signif_line == 0:
47
            # _RAJ2000;_DEJ2000;Name;RAJ2000;DEJ2000;Bmag;Vmag;SpType;FileName
48
            col_names = line.split(';')
49
            for i, col_name in enumerate(col_names):
50
                out_dict['data_lists'][col_name] = []
51
                icol_to_colname[i] = col_name
52
        elif i_signif_line == 1:
53
            unit_str_list = line.split(';')
54
            for i, unit_str in enumerate(unit_str_list):
55
                out_dict['data_units_dict'][icol_to_colname[i]] = unit_str
56
        elif i_signif_line == 2:
57
            pass # skip this but also increment i_signif_line below
58
        else:
59
            elems = line.split(';')
60
            for i, elem in enumerate(elems):
61
                out_dict['data_lists'][icol_to_colname[i]].append(elem)
62
        i_signif_line += 1
63
    return out_dict

58
def read_tsv(self):
59
    lines = []
60
    with self.annotation_file.open('r') as ann_file:
61
        reader = csv.reader(ann_file, delimiter="\t", quotechar=None)
62
        for idx, line in enumerate(reader):
63
            if idx == 0:
64
                continue
65
            guid = "dev-{}".format(idx)
66
            language = line[0]
67
            if self.language_filter and language not in self.language_filter:
68
                continue
69
            label = reversed_label_map[line[1]]
70
            text_a = line[6]
71
            text_b = line[7]
72
            lines.append(InputExample(guid, text_a, text_b, label))
73
74
    return lines