Vishal Kumar | Pandas Handbook

Data Scientist 10+ Projects MSc Data Science JK Tyres R&D
437
Total Functions
3
Data Columns
Python
Technology
Free
Open Source
All Columns Function Meaning Example

Pandas DataFrame Functions Reference

Showing 437 entries
Function Meaning Example (Python)
T Transpose DataFrame - swap rows and columns df.T()
abs Return absolute value of each element df.abs()
add Add elements to DataFrame df.add(other)
add_prefix Add prefix to column labels df.add_prefix('col_')
add_suffix Add suffix to column labels df.add_suffix('_2024')
agg Aggregate using one or more operations df.agg(['sum', 'mean'])
aggregate Aggregate using one or more operations df.aggregate(['sum', 'min'])
align Align two DataFrames on axes df1.align(df2)
all Return whether all elements are True df.all(axis=1)
any Return whether any element is True df.any(axis=0)
append Append rows of other DataFrame df.append(other)
apply Apply function along axis df.apply(lambda x: x*2)
applymap Apply function element-wise df.applymap(str.upper)
asfreq Convert time series to specified frequency df.asfreq('D')
asof Return last non-NaN value df.asof('2024-01-01')
assign Assign new columns to DataFrame df.assign(new_col=df['a']*2)
astype Cast object to dtype df.astype('float32')
at Access single value by label df.at[0, 'column']
at_time Select values at particular time df.at_time('09:30')
attrs Dictionary of global attributes df.attrs['description']
axes Return list of axis labels df.axes
backfill Backward fill NaN values df.backfill()
between_time Select values between times df.between_time('09:00', '17:00')
bfill Backward fill NaN values df.bfill()
bool Return bool of single element df.bool()
boxplot Make box plot from DataFrame df.boxplot()
clip Trim values at thresholds df.clip(lower=0, upper=100)
columns Column labels of DataFrame df.columns
combine Combine DataFrames element-wise df1.combine(df2, np.minimum)
combine_first Combine with other DataFrame, filling nulls df1.combine_first(df2)
compare Compare DataFrames and show differences df1.compare(df2)
convert_dtypes Convert columns to best possible dtypes df.convert_dtypes()
copy Copy DataFrame df.copy()
corr Compute pairwise correlation df.corr()
corrwith Compute pairwise correlation with another df.corrwith(other)
count Count non-NA cells df.count()
cov Compute covariance df.cov()
cummax Cumulative maximum df.cummax()
cummin Cumulative minimum df.cummin()
cumprod Cumulative product df.cumprod()
cumsum Cumulative sum df.cumsum()
describe Generate descriptive statistics df.describe()
diff First discrete difference df.diff()
div Floating division df.div(other)
divide Floating division df.divide(other)
dot Matrix multiplication df.dot(other)
drop Drop specified labels df.drop(['col1'], axis=1)
drop_duplicates Drop duplicate rows df.drop_duplicates()
droplevel Drop levels from MultiIndex df.droplevel(1)
dropna Drop missing values df.dropna()
dtypes Return dtypes of columns df.dtypes
duplicated Return boolean Series for duplicates df.duplicated()
empty Check if DataFrame is empty df.empty
eq Equal comparison df.eq(other)
equals Check if two DataFrames are equal df.equals(other)
eval Evaluate string expression df.eval('A + B')
ewm Exponential weighted functions df.ewm(span=3).mean()
expanding Expanding window functions df.expanding().mean()
explode Transform list-like to rows df.explode('col')
ffill Forward fill NaN values df.ffill()
fillna Fill NaN values df.fillna(0)
filter Filter DataFrame by labels df.filter(items=['A', 'B'])
first Select first periods df.first('5D')
first_valid_index Return index of first non-NA value df.first_valid_index()
flags Get flags of DataFrame df.flags
floordiv Integer division df.floordiv(other)
from_dict Create DataFrame from dict pd.DataFrame.from_dict(data)
from_records Create DataFrame from records pd.DataFrame.from_records(records)
ge Greater than or equal comparison df.ge(other)
get Get item from object df.get('column', default=0)
groupby Group DataFrame by mapping df.groupby('col').mean()
gt Greater than comparison df.gt(other)
head Return first n rows df.head(10)
hist Plot histogram df.hist()
iat Access single value by position df.iat[0, 1]
idxmax Return index of maximum df.idxmax()
idxmin Return index of minimum df.idxmin()
iloc Purely integer-location indexing df.iloc[0:5, 0:2]
index Index of DataFrame df.index
infer_objects Infer better dtypes df.infer_objects()
info Print concise summary df.info()
insert Insert column at location df.insert(1, 'new', values)
interpolate Interpolate NaN values df.interpolate()
isetitem Set item by position df.isetitem(0, values)
isin Check if values in Series/DataFrame df.isin([1, 2, 3])
isna Detect missing values df.isna()
isnull Detect missing values df.isnull()
items Iterate over columns for label, content in df.items()
iterrows Iterate over DataFrame rows for index, row in df.iterrows()
itertuples Iterate as namedtuples for row in df.itertuples()
join Join columns with other DataFrame df.join(other, on='key')
keys Get axis labels df.keys()
kurt Unbiased kurtosis df.kurt()
kurtosis Unbiased kurtosis df.kurtosis()
last Select last periods df.last('5D')
last_valid_index Index of last non-NA value df.last_valid_index()
le Less than or equal comparison df.le(other)
loc Label-location indexing df.loc[:, 'A':'C']
lt Less than comparison df.lt(other)
map Apply function element-wise df['col'].map(func)
mask Replace values where condition is True df.mask(df < 0, 0)
max Maximum of values df.max()
mean Mean of values df.mean()
median Median of values df.median()
melt Unpivot DataFrame df.melt(id_vars=['id'])
memory_usage Memory usage of DataFrame df.memory_usage()
merge Merge DataFrames df.merge(df2, on='key')
min Minimum of values df.min()
mod Modulo df.mod(other)
mode Mode of values df.mode()
mul Multiplication df.mul(other)
multiply Multiplication df.multiply(other)
ndim Number of dimensions df.ndim
ne Not equal comparison df.ne(other)
nlargest Return n largest values df.nlargest(5, 'col')
notna Detect non-missing values df.notna()
notnull Detect non-missing values df.notnull()
nsmallest Return n smallest values df.nsmallest(5, 'col')
nunique Count unique values df.nunique()
pad Forward fill NaN values df.pad()
pct_change Percentage change df.pct_change()
pipe Apply chain of functions df.pipe(func1).pipe(func2)
pivot Reshape data using pivot df.pivot(index='date', columns='type')
pivot_table Create pivot table df.pivot_table(values='value', index='date')
plot Plot DataFrame df.plot()
pop Pop column and return df.pop('col')
pow Exponentiation df.pow(other)
prod Product of values df.prod()
product Product of values df.product()
quantile Return quantile df.quantile(0.5)
query Query DataFrame with expression df.query('col > 5')
radd Reverse addition df.radd(other)
rank Rank of values df.rank()
rdiv Reverse division df.rdiv(other)
reindex Conform DataFrame to new index df.reindex(new_index)
reindex_like Reindex to match other DataFrame df.reindex_like(other)
rename Rename columns/index df.rename(columns={'old': 'new'})
rename_axis Set axis name df.rename_axis('date')
reorder_levels Reorder index levels df.reorder_levels([1,0])
replace Replace values df.replace(old, new)
resample Resample time-series data df.resample('M').mean()
reset_index Reset index to default df.reset_index()
rfloordiv Reverse integer division df.rfloordiv(other)
rmod Reverse modulo df.rmod(other)
rmul Reverse multiplication df.rmul(other)
rolling Rolling window calculations df.rolling(window=3).mean()
round Round values df.round(2)
rpow Reverse exponentiation df.rpow(other)
rsub Reverse subtraction df.rsub(other)
rtruediv Reverse floating division df.rtruediv(other)
sample Random sample of rows df.sample(5)
select_dtypes Select columns by dtype df.select_dtypes(include=['number'])
sem Standard error of mean df.sem()
set_axis Set axis labels df.set_axis(new_labels, axis=1)
set_flags Set flags df.set_flags(allows_duplicate_labels=False)
set_index Set DataFrame index df.set_index('col')
shape Tuple of dimensions df.shape
shift Shift index by periods df.shift(1)
size Number of elements df.size
skew Unbiased skewness df.skew()
sort_index Sort by index df.sort_index()
sort_values Sort by values df.sort_values('col')
squeeze Squeeze 1D axes df.squeeze()
stack Stack columns to rows df.stack()
std Standard deviation df.std()
sub Subtraction df.sub(other)
subtract Subtraction df.subtract(other)
sum Sum of values df.sum()
swapaxes Swap axes df.swapaxes(0, 1)
swaplevel Swap levels in MultiIndex df.swaplevel(0, 1)
tail Return last n rows df.tail(5)
take Take by positions df.take([0, 2, 4])
to_clipboard Copy to clipboard df.to_clipboard()
to_csv Write to CSV df.to_csv('file.csv')
to_dict Convert to dictionary df.to_dict()
to_excel Write to Excel df.to_excel('file.xlsx')
to_feather Write to Feather format df.to_feather('file.feather')
to_gbq Write to BigQuery df.to_gbq('dataset.table')
to_hdf Write to HDF5 df.to_hdf('file.h5', 'key')
to_html Render as HTML table df.to_html('file.html')
to_json Write to JSON df.to_json('file.json')
to_latex Render as LaTeX table df.to_latex()
to_markdown Render as Markdown df.to_markdown()
to_numpy Convert to numpy array df.to_numpy()
to_orc Write to ORC format df.to_orc('file.orc')
to_parquet Write to Parquet df.to_parquet('file.parquet')
to_period Convert to PeriodIndex df.to_period('M')
to_pickle Serialize to pickle df.to_pickle('file.pkl')
to_records Convert to record array df.to_records()
to_sql Write to SQL database df.to_sql('table', con)
to_stata Write to Stata format df.to_stata('file.dta')
to_string Render as string print(df.to_string())
to_timestamp Convert to Timestamp df.to_timestamp()
to_xarray Convert to xarray df.to_xarray()
to_xml Render as XML df.to_xml('file.xml')
transform Apply function and return same shape df.transform('sqrt')
transpose Transpose index and columns df.transpose()
truediv Floating division df.truediv(other)
truncate Truncate before/after df.truncate(before='2024-01-01')
tz_convert Convert timezone df.tz_convert('US/Eastern')
tz_localize Localize timezone df.tz_localize('UTC')
unstack Unstack level to columns df.unstack()
update Update with other DataFrame df.update(other)
value_counts Count unique values df.value_counts()
values Return numpy array df.values
var Variance df.var()
where Replace values where condition is False df.where(df > 0, 0)
xs Cross-section from Series/DataFrame df.xs('key', level='level')
Function - Name of the DataFrame method Meaning - Description of what it does Example - Usage example in Python

Created for education by Vishal Kumar

Last Updated: February 2026