Pandas Dataframe - Adding Else?-CodePudding

I want to generate Test Data for my Bayesian Network. This is my current Code:

data = np.random.randint(2, size=(5, 6))
columns = ['p_1', 'p_2', 'OP1', 'OP2', 'OP3', 'OP4']
df = pd.DataFrame(data=data, columns=columns)

df.loc[(df['p_1'] == 1) & (df['p_2'] == 1), 'OP1'] = 1

df.loc[(df['p_1'] == 1) & (df['p_2'] == 0), 'OP2'] = 1

df.loc[(df['p_1'] == 0) & (df['p_2'] == 1), 'OP3'] = 1

df.loc[(df['p_1'] == 0) & (df['p_2'] == 0), 'OP4'] = 1


print(df)

So every time, for example, p_1 has a 1 and p_2 has a 1, the OP1 should be 1 as well, all the other values should output 0 in the column. When p_1 is 1 and p_2 is 0, then OP2 should be 1 an d all others 0, and so on.

But my current Output is the following:

p_1	p_2	OP1	OP2	OP3	OP4
0	0	0	0	0	1
1	0	1	1	1	1
0	0	1	1	0	1
0	1	1	1	1	1
1	0	0	1	1	0

Is there any way to fix it? What did I do wrong?

I didn't really understand the solutions to other peoples questions, so I thought Id ask here.

I hope that someone can help me.

CodePudding user response：

The problem is that when you instantiate df, the "OP" columns already have some values:

data = np.random.randint(2, size=(5, 6)) 
columns = ['p_1', 'p_2', 'OP1', 'OP2', 'OP3', 'OP4'] 
df = pd.DataFrame(data=data, columns=columns) 

df                                                                      

   p_1  p_2  OP1  OP2  OP3  OP4
0    1    1    0    1    0    0
1    0    0    1    1    0    1
2    0    1    1    1    0    0
3    1    1    1    1    0    1
4    0    1    1    0    1    0

One way of fixing it with your code is forcing all "OP" columns to 0 before:

df["OP1"] = df["OP2"] = df["OP3"] df["OP4"] = 0

But then you are generating too many random numbers. I'd do this instead:

data = np.random.randint(2, size=(5, 2)) 
columns = ['p_1', 'p_2'] 
df = pd.DataFrame(data=data, columns=columns) 
df["OP1"] = ((df['p_1'] == 0) & (df['p_2'] == 1)).astype(int)

CodePudding user response：

You can defined tuples for test and create new columns by casting values of mask to inetegers for True/False to 1/0 mapping:

vals = [(1,1),(1,0),(0,1),(0,0)]
for i, (a, b) in enumerate(vals, 1):
    df[f'OP{i}'] = ((df['p_1'] == a) & (df['p_2'] == b)).astype(int)
print(df)
   p_1  p_2  OP1  OP2  OP3  OP4
0    0    0    0    0    0    1
1    0    1    0    0    1    0
2    0    1    0    0    1    0
3    0    1    0    0    1    0
4    1    0    0    1    0    0

In your solution set 0 first, because already are set 1 values in original DataFrame:

cols = ['OP1', 'OP2', 'OP3', 'OP4']
df[cols] = 0