Зміст курсу
Preprocessing Data
Preprocessing Data
1. Data Exploration
3. Data Validation
4. Normalization & Standardization
5. Data Encoding
Missing and Wrong Data
As you already know, it is possible that raw data can contain some dirty data. It can be:
- NaN: undefined or missing data.
- empty strings.
- infinite: very large data.
- incorrect data: for example, 'Female' in the Price column, that contains numeric data (this value could be stored into the wrong cell accidentally). You may find impossible values of the user's age, for example, if this value should be entered by him manually (like -1, 110, 0, etc.).
- outliers: critically small or big values(for example, 250 cm in the Height column, or 112 yrs in the Age column), usually in a small amount. They may affect your result of analysis or model weights, so sometimes it makes sense to remove them.
Let's learn how to 'clean' your data and not to lose some useful info.
Перейдіть на комп'ютер для реальної практикиПродовжуйте з того місця, де ви зупинились, використовуючи один з наведених нижче варіантівВсе було зрозуміло?
Секція 2. Розділ 1
Missing and Wrong Data
As you already know, it is possible that raw data can contain some dirty data. It can be:
- NaN: undefined or missing data.
- empty strings.
- infinite: very large data.
- incorrect data: for example, 'Female' in the Price column, that contains numeric data (this value could be stored into the wrong cell accidentally). You may find impossible values of the user's age, for example, if this value should be entered by him manually (like -1, 110, 0, etc.).
- outliers: critically small or big values(for example, 250 cm in the Height column, or 112 yrs in the Age column), usually in a small amount. They may affect your result of analysis or model weights, so sometimes it makes sense to remove them.
Let's learn how to 'clean' your data and not to lose some useful info.
Перейдіть на комп'ютер для реальної практикиПродовжуйте з того місця, де ви зупинились, використовуючи один з наведених нижче варіантівВсе було зрозуміло?
Секція 2. Розділ 1
Missing and Wrong Data
As you already know, it is possible that raw data can contain some dirty data. It can be:
- NaN: undefined or missing data.
- empty strings.
- infinite: very large data.
- incorrect data: for example, 'Female' in the Price column, that contains numeric data (this value could be stored into the wrong cell accidentally). You may find impossible values of the user's age, for example, if this value should be entered by him manually (like -1, 110, 0, etc.).
- outliers: critically small or big values(for example, 250 cm in the Height column, or 112 yrs in the Age column), usually in a small amount. They may affect your result of analysis or model weights, so sometimes it makes sense to remove them.
Let's learn how to 'clean' your data and not to lose some useful info.
Перейдіть на комп'ютер для реальної практикиПродовжуйте з того місця, де ви зупинились, використовуючи один з наведених нижче варіантівВсе було зрозуміло?
As you already know, it is possible that raw data can contain some dirty data. It can be:
- NaN: undefined or missing data.
- empty strings.
- infinite: very large data.
- incorrect data: for example, 'Female' in the Price column, that contains numeric data (this value could be stored into the wrong cell accidentally). You may find impossible values of the user's age, for example, if this value should be entered by him manually (like -1, 110, 0, etc.).
- outliers: critically small or big values(for example, 250 cm in the Height column, or 112 yrs in the Age column), usually in a small amount. They may affect your result of analysis or model weights, so sometimes it makes sense to remove them.
Let's learn how to 'clean' your data and not to lose some useful info.
Перейдіть на комп'ютер для реальної практикиПродовжуйте з того місця, де ви зупинились, використовуючи один з наведених нижче варіантівСекція 2. Розділ 1
Перейдіть на комп'ютер для реальної практикиПродовжуйте з того місця, де ви зупинились, використовуючи один з наведених нижче варіантів