Remove rows from Pandas dataframe where value only appears once

Change the len to count

df[df.groupby('ID').ID.transform('count') > 1]
Out[589]: 
   ID       Month  Metric1  Metric2
0   1  2018-01-01        4        3
1   1  2018-02-01        3        2
3   3  2018-01-01        4        2
4   3  2018-02-01        6        3

Try with pd.series.duplicated():

df1=df[df.ID.duplicated(keep=False)]
print(df1)

---

   ID       Month  Metric1  Metric2
0   1  2018-01-01        4        3
1   1  2018-02-01        3        2
3   3  2018-01-01        4        2
4   3  2018-02-01        6        3

filter

I cannot vouche for the speed of this but this is what this API was intended for...

df.groupby('ID').filter(lambda d: len(d) > 1)

   ID       Month  Metric1  Metric2
0   1  2018-01-01        4        3
1   1  2018-02-01        3        2
3   3  2018-01-01        4        2
4   3  2018-02-01        6        3

---

Numpy'd version of @Wen-Ben's answer

u, i = np.unique(df.ID.values, return_inverse=True)

df[np.bincount(i)[i] > 1]

   ID       Month  Metric1  Metric2
0   1  2018-01-01        4        3
1   1  2018-02-01        3        2
3   3  2018-01-01        4        2
4   3  2018-02-01        6        3

---

Because I was curious...

s0 = set()
s1 = set()

for i in df.ID:
    if i in s0:
        s1.add(i)
    s0.add(i)

df[df.ID.map(s1.__contains__)]

   ID       Month  Metric1  Metric2
0   1  2018-01-01        4        3
1   1  2018-02-01        3        2
3   3  2018-01-01        4        2
4   3  2018-02-01        6        3