接受用户输入并使用Pandas Python在CSV中搜索

Question

I'm trying to take user input and search the csv file by asking the make, model, and year of the car but it is not filtering the cars correctly when I get to model. It is still showing all the car models even when I want only Toyota cars. I am also getting Empty dataframe error when I finish the input.

sData.csv

import pandas

# reads in vehicle Data
df = pandas.read_csv('sData.csv')
pandas.set_option('display.max_columns', None)
pandas.set_option('display.width', 400)

def get_choice(data, column):

    #Gets user choice
    nums = [val for val in range(len(df[column].unique()))]
    choices = list(zip(nums, df[column].unique()))
    print("Select '%s' of the car\n" % column)
    for v in choices:
        print("%s.  %s" % (v))
    user_input = input("Answer: ")
    user_answer = [val[1] for val in choices if val[0]==int(user_input)][0]
    print("'%s' = %s\n" % (column, user_answer))
    return user_answer

def main():

    make_input = get_choice(data=df, column="make")
    model_input = get_choice(data=df, column="model")
    year_input = get_choice(data=df, column="year")
    newdf = df.loc[(df["make"]==make_input)&(df["model"]==model_input)&(df["year"]==year_input)]
    print(newdf)

if __name__ == "__main__":
    main()

Answer 1

The issue you are having with regards to models not being filtered by make is caused by the fact that you are not modifying df or passing a modified copy of it to the subsequent calls to make_choice . Essentially, each time you call make_choice the results are not filtered by prior choices.

There are multiple possible solutions, one would be to simply update df as you make choices, but this wouldn't keep the original df .

make_input = get_choice(data=df, column="make")
df = df.loc[df.make == make_input, :]
...

Alternatively:

filtered_df = df.copy() # Could be costly if df is large
make_input = get_choice(data=filtered_df, column="make")
filtered_df = filtered_df.loc[filtered_df.make == make_input, :]
model_input = get_choice(data=filtered_df, column="model")
...

As for the empty data frame error, I would advise to use a debugger and see what each step of your code is yielding. If you do not already know how to use it, I would recommend you to look at pdb .