Compare 2 df.columns with multiple conditions have wrong result

I have a pandas column df.a. which have values of 1, 0, -1, and compare to a new column df.c. The requirement is as follows:

1. The count(df.c) must be between -1 and 1.
2. If the df.a is greater than the previous df.c by 1 or less than the previous df.c by -1, the previous df.c will be put into the current df.c.
3. Otherwise, if df.a is equal to zero. it will also put the previous df.c to current df.c.
4. If df.a equal to 1 or -1, it will calculate df.a with previous df.c and put into current df.c

My coding is as follows:

Initialize df.c as df['a'][0] == 1.

a=df['a'] + df['count'].shift(1) < -1 
b=df['a'] + df['count'].shift(1) > 1

c=df['count'].shift(1) == -1
d=df['count'].shift(1) == 1
e=df['count'].shift(1) == '0'

g=df['a']=='-1'
h=df['a']=='1'
i=df['a']=='0'

conditions=[a|b|i&d|i&c,h&e|g&e,e&i|c&h|d&g]
choices=[df['count'].shift(1), df['a'], '0']
df['count'] = np.select(conditions, choices, default=df['count'])

My question is when I compare df['a'][4], df['count'][6] should be (-1, '1') but the result is 1, it should fall into (e) -1 + 1 = 0. Any suggestion? Have I done something wrong?

    a   c   expected result

0   1   nan 1
1   1   1.0 1
2   0   1.0 1
3   1   1.0 1
4   -1  0   0 
5   0   1.0 0
6   1   1.0 1
7   1   1.0 1
8   -1  0   0
9   -1  0   -1
10  -1  0   -1
11  0   1.0 -1
12  -1  0   -1
13  1   1.0 0
14  0   1.0 0
15  1   1.0 1