Hello everybody, I’m not stuck at any part at the moment but, at the end of the guided project for Pandas and NumPy Fundamentals I found the following extra challenge:
Find the most common brand/model combinations.
And I was challenged enough to find the solution, therefore, I wrote the piece of code below.
# Most common brand/model combinations. # Importing pandas library. import pandas as pd # Opening the dataset. autos_copy = pd.read_csv('autos.csv',encoding='Latin-1') autos_copy["brand_model"] = autos_copy.brand + '|' + autos_copy.model # Had to get rid of this "brand", it is kind of a general brand. unique_brand = autos_copy.loc[autos_copy["brand"] != 'sonstige_autos',"brand"].unique() brand_model_list =  for each_brand in unique_brand: #print('-----------',each_brand) brand,model = autos_copy.loc[autos_copy.loc[:,"brand"] == each_brand,"brand_model"].value_counts().sort_values(ascending=False).index.split("|") ads = autos_copy.loc[autos_copy.loc[:,"brand"] == each_brand,"brand_model"].value_counts().iloc brand_model_list.append([brand,model,ads]) #print("Brand:",brand," Model:",model," Ads:",ads) columns_names = ["brand","model","ads"] common_brand_model = pd.DataFrame(data=brand_model_list,columns = columns_names) print("Most common brand/model combinations") common_brand_model.sort_values("ads",ascending=False).head()
I didn’t get any error, I only want to know if the result is correct, and if the logic is not that wrong, I wasn’t able to come up with a different solution.