detect_groups_with_common_vals.py

"""
Detect groups having common values

Example: unpack.csv

PATID, HASH_1, HASH_2
pat_1, a, b
pat_2, x
pat_3, , y
pat_4, a, c
pat_5, d, b
pat_6, e, b
pat_7, e, f

"""
import pandas as pd
import numpy as np


filename = 'unpack.csv'

df = pd.read_csv(filename, sep=",", skipinitialspace=True, dtype=object)
df.fillna('', inplace=True)

df['COUNT_1'] = 0
df['COUNT_2'] = 0

print(df)

grp = df.groupby('HASH_1', as_index=False)
gr1 = df.groupby('HASH_1', as_index=False).agg(np.size)


print(gr1)
found_1 = gr1.loc[gr1['COUNT_1'] > 1]

print("Found: \n{}".format(found_1))


# df_a = df.groupby('HASH_1').agg(lambda x: ','.join(x))
#
#
# for i in df_a.index:
#     print("\n -> Name: {}".format(i))
#     print(df_a.loc[i]['PATID'])
#

for name, group in grp:
    print(name)
    print("Group size: {}".format(len(group)))