Working with DataFrames — quiz

How do you stop 07030 becoming 7030?

1. astype(int)
2. dtype={'zip': str} when reading
3. parse_dates
4. encoding="utf-8"

Answer: dtype={'zip': str} when reading — Read identifier-like columns as strings.

pd.read_excel(..., sheet_name=None) returns:

1. The first sheet
2. A dictionary of every sheet
3. An error
4. An empty DataFrame

Answer: A dictionary of every sheet — Keyed by sheet name.

info() shows order_date as object. This means:

1. It is a date
2. It is stored as text
3. It contains nulls
4. It is an index

Answer: It is stored as text — object means text; convert with to_datetime.

An integer column showing float64 usually indicates:

1. A bug
2. Missing values are present
3. Large numbers
4. Wrong encoding

Answer: Missing values are present — NaN forces the column to float.

df[['a']] returns:

1. A Series
2. A DataFrame
3. A list
4. An index

Answer: A DataFrame — Double brackets always give a DataFrame.

df.rename(columns={...}) without assignment:

1. Renames in place
2. Returns a new DataFrame and changes nothing
3. Raises an error
4. Renames the index

Answer: Returns a new DataFrame and changes nothing — Assign the result, or pass inplace=True.

df[df['a'] > 1 and df['b'] < 5] raises an error because:

1. a is not numeric
2. pandas requires & rather than and
3. Brackets are wrong
4. It needs .loc

Answer: pandas requires & rather than and — The "truth value is ambiguous" error means and was used instead of &.

df.loc[0:5] versus df.iloc[0:5]:

1. Identical
2. loc includes row 5, iloc excludes it
3. iloc includes 5, loc excludes it
4. loc only takes conditions

Answer: loc includes row 5, iloc excludes it — loc is label-based and slice-inclusive.

Which handles several conditions in order?

1. np.where
2. np.select
3. apply
4. astype

Answer: np.select — np.select takes condition and choice lists, first match wins.

SettingWithCopyWarning is best fixed by:

1. Ignoring it
2. Adding .copy() after filtering, or assigning via df.loc
3. Using apply
4. Resetting the index

Answer: Adding .copy() after filtering, or assigning via df.loc — It means pandas cannot tell whether you hold a view or a copy.