else statement for nested for loop-CodePudding

I have defined a function which extracts words from sentences based on it's pos_tag.

def get_trigram(pos_1, pos_2, pos_3):
    all_trigram = []

    for j in range(len(df)):

        trigram = []

        if len(df['pos'][j]) >= 2:

            for i in range(len(df['pos'][j])):

                if df['pos'][j][i-2][1] == pos_1 and df['pos'][j][i-1][1] == pos_2 and df['pos'][j][i][1] == pos_3:
                    trigram.append(df['pos'][j][i-2][0]   " "   df['pos'][j][i-1][0]   " "   df['pos'][j][i][0])

            all_trigram.append(trigram)
      
    return all_trigram

The function runs and can work but the len of thelist all_trigram is less than the original len of dataframe which I am running on. I suspect it is because of this line of code

if len(df['pos'][j]) >= 2:

inside my function and thus for the rows that have less than 2, they are not captured inside as blanks. How can i construct the else statement and where can i place it such that the all_trigrams list can contain a blank list too for those rows that have less than 2?

Here is a sample of the df:

CodePudding user response：

Append an empty list to all_trigram when the condition len(df['pos'][j]) >= 2: is not satisfied. This will ensure that the size of all_trigram is same as the size of your dataframe.

def get_trigram(pos_1, pos_2, pos_3):
    all_trigram = []

    for j in range(len(df)):

        trigram = []

        if len(df['pos'][j]) >= 2:

            for i in range(len(df['pos'][j])):

                if df['pos'][j][i-2][1] == pos_1 and df['pos'][j][i-1][1] == pos_2 and df['pos'][j][i][1] == pos_3:
                    trigram.append(
                        df['pos'][j][i-2][0]   " "   df['pos'][j][i-1][0]   " "   df['pos'][j][i][0])

            all_trigram.append(trigram)
        else:
            all_trigram.append([])

    return all_trigram