강의

챌린지

멘토링

로드맵

커뮤니티

BEST
Data Science

/

Certificate (Data Science)

[Side Projects After Work] Big Data Analysis Engineer Practical Exam (Task Types 1, 2, and 3)

We’ll guide non-majors and beginners to quickly earn the Big Data Analysis Engineer practical certification! Keep the theory light and the practical training solid. Even without complex background knowledge, you’ll focus on past exam questions and study only the key points that are sure to appear on the test.

(4.9) 944 reviews

6,197 learners

Level Beginner

Course period 12 months

  • roadmap
Engineer Big Data Analysis
Engineer Big Data Analysis
Big Data
Big Data
Python
Python
Pandas
Pandas
Machine Learning(ML)
Machine Learning(ML)
Engineer Big Data Analysis
Engineer Big Data Analysis
Big Data
Big Data
Python
Python
Pandas
Pandas
Machine Learning(ML)
Machine Learning(ML)
News list
roadmap님의 프로필 이미지
roadmap

Encoding method taught by GPT, align❣️

Recently, I've been getting a lot of questions asking me if it's okay to use GPT, which provides a very simple way to align when the number of unique columns in train and test are different :)

First of all , you can use it. 🤩 GPT is the best

However, it does have its drawbacks, so it would be better to use it knowing the pros and cons 💪 💪 💪 💪

 train = pd.get_dummies(train) test = pd.get_dummies(test) train, test = train.align(test, join='left', axis=1, fill_value=0)

✅ train.align(test, join='left', axis=1, fill_value=0)

Code to match test to train's column order and structure

💚 Advantages

  • It's very simple
    You can unify the column order and structure of test according to train criteria in one line.

  • Preventing errors during model training
    You can prevent errors that occur due to mismatched train/test columns in advance.

  • You can fill missing columns with fill_value
    Flexible processing is possible, such as automatically filling in columns that do not exist in the test with 0.

❤ Cons

  • Columns that exist only in test are deleted
    Since join='left' , columns that are not in train are forcibly removed from test.

Comment