Exemplos Pandas DataFrame.

exemplos.py

# Useful Pandas Snippets.

# Data Types and Conversion.

# Convert Series datatype to numeric (will error if column has non-numeric values)

pd.to_numeric(df['Column Name'])

# Convert Series datatype to numeric, changing non-numeric values to NaN.
pd.to_numeric(df['Column Name'], errors='coerce)

# Change data type of DataFrame column.
df.column_name = df.column_name.astype(np.int64)

# ------------------------------------------------------

# Exploring and Filtering Data.

# Get a report of all duplicate records in DataFrame, based on specifc columns.

dupes = df[df.duplicated(['col1', 'col2', 'col3'], keep=False)]

# List unique values in DataFrame column.

df['Column Name'].unique()

# For each unique values in a DataFrame column, get a frequency count.

df['Column Name'].value_counts()

# Grab DataFrame rows when column = a specific value.

df = df.loc[df.column == 'somevalue']

# Grab DataFrame rows when column values is present in a list.

test_data = {'hi': 'yo', 'bye': 'later'}
df = pd.DataFrame(list(d.items()), column=['col1', 'col2'])
valuelist = ['yo', 'heya']
df[df.col2.isin(valuelist)]

# Grav DataFrame rows where column values is not present in a list.

test_data = {'h1': 'yo', 'bye': 'later'}
df = pd.DataFrame.(list(d.items()), columns=['col1', 'col2'])
valuelist = ['yo', 'later']
df[~df.col2.isin(valuelist)]

# Select from DataFrame using criteria from multiple columns (use | instead of & to do an OR)

newdf = df[(df['column_one'] > 2004]) & (df['column_two'] == 9)

# Loop through rows in a DataFrame.

for index, row in df.iterrows():
    print(index, row['some column']

# Much faster way to loop thhrough DataFrame rows if you can work with tuples.

for row in df.itertuples():
    print(row)

# Get top n for each group of columns in a sorted DataFrame.

top5 = df.groupby(['groupingcol1', 'goupingcol2']).head(5)

# Grab DataFrame rows where specific column is null/notnull.

newdf = df[df['column'].isnull()]

# Get quick count of rows in a DataFrame.

len(df.index)

# Get length of data in DataFrame column.
df.column_name.str.len()