Exploring Data with Pandas: Intermediate

How to make first column of dataframe as row label?

by index_col = 0

f500 = pd.read_csv("f500.csv",index_col = 0)

How to name row label column?

df.index.name = Name of row label

E.g.: f500.index.name = "Company"

How to name column label column?

df.columns.name= Name of columns label

E.g.: f500.columns.name = "Metric"

How to select first value in third row?

df.iloc[2, 0]

How to select all the rows in first column using iloc?

first_column = f500.iloc[:, 0]

How to get null value of a column and get three column values based on that?

prev_rank_null = f500["previous_rank"].isnull()

null_prev_rank = f500[prev_rank_null][["company", "rank", "previous_rank"]]

print(null_prev_rank)

Note: We can use notnull() for the opposite operation.

What are the other operators we can use apart from ==, !=, >, <?

&, |, ~

How to invert the boolean indexing?

df[~(df["A"] == X)]

~ will invert the original value. That means the operation is not equal to X.

How to sort dataframe based on a column name?

selected_rows = f500.loc[f500.loc[:, "country"] == "Japan"]

sorted_rows = selected_rows.sort_values("profits")

In the above example, the dataframe is sorted in ascending order by “profits” column. We can sort them in descending order by below syntax.

selected_rows = f500.loc[f500.loc[:, "country"] == "Japan"]

sorted_rows = selected_rows.sort_values("profits", ascending = False)

How to get unique values from a column?

f500.loc[:, "country"].unique()

How to get highest roa of each sector?

f500["roa"] = f500.loc[:, "profits"] / f500.loc[:, "assets"]

top_roa_by_sector = {}

sectors = f500.loc[:, "sector"].unique()

for s in sectors:

selected_companies = f500.loc[f500.loc[:, "sector"] == s]

sorted_companies = selected_companies.sort_values("roa", ascending = False)

top_roa_by_sector[s] = sorted_companies.iloc[0,0]