for循环只使用dict中的第一个键

Question

I'm trying to calculate whether or not an item has been delayed. I have a dictionary containing 20+ keys and values, and am trying to loop through the rows in my dataset, and if the dataframe value is above the dict value, it should result in "True".

It seems like the loop is only using the first value in the dict, and I've tried changing it to no avail.

This is the current code I'm using:

def delaytest(df):
    for key, value in mydict.items():
        if (key == df["serviceType"]) & (value < (df["delayTime"]/60)):
            return True

        else:
            return False

df["delay"] = df.apply(delaytest, axis=1)

Example dict:

dict = {"key1": 5,
        "key2": 10,
        "key3": 15}

df before:
    serviceType   delayTime
    key1          6
    key2          11
    key2          12
    key1          4
    key3          16
    key3          14

df after:
       serviceType   delayTime  delay
    key1          6          True
    key2          11         False
    key2          12         False
    key1          4          False
    key3          16         False
    key3          14         False

df expected:
    serviceType   delayTime  delay
    key1          6          True  
    key2          11         True
    key2          12         True
    key1          4          False
    key3          16         True
    key3          14         False

Answer 1

if the dataframe value is above the dict value, it should result in "True".

Undestandable. But in your code, you also return False if you don't hit True . So with the first element, you always return something, therefore exiting the function. ;)

Get rid of the else with the return False and it should work:

def delaytest(df):
    for key, value in mydict.items():
        if (key == df["serviceType"]) & (value < (df["delayTime"]/60)):
            return True

df["delay"] = df.apply(delaytest, axis=1)

Answer 2

I think here is better use Series.map for new Series and compare by column delayTime :

df["delay"] = df['serviceType'].map(mydict) < df["delayTime"]
print (df)
  serviceType  delayTime  delay
0        key1          6   True
1        key2         11   True
2        key2         12   True
3        key1          4  False
4        key3         16   True
5        key3         14  False

Detail :

print (df['serviceType'].map(mydict))
0     5
1    10
2    10
3     5
4    15
5    15
Name: serviceType, dtype: int64

If map with value not matched, get missing values and output of comparison are False s:

mydict = {"key1": 5,
        "key2": 10,
        "key4": 15}

df["delay"] = df['serviceType'].map(mydict) < df["delayTime"]
print (df)
  serviceType  delayTime  delay
0        key1          6   True
1        key2         11   True
2        key2         12   True
3        key1          4  False
4        key3         16  False
5        key3         14  False

print (df['serviceType'].map(mydict))
0     5.0
1    10.0
2    10.0
3     5.0
4     NaN
5     NaN
Name: serviceType, dtype: float64

EDIT:

For possible map by multiple values create helper DataFrame and merge with left join:

pairs = [('cat1','prov1', 'content1'),
         ('cat2','prov2', 'content2'),
         ('cat3','prov3', 'content3')]
df1 = pd.DataFrame(pairs, columns=['category','provider','contentType'])
print (df1)

df = df.merge(df1, how='left')