Python Reference

Created by Nishant Kheterpal

Table Functions and Methods

In the examples in the left column, np refers to the NumPy module, as usual. Everything else is a function, a method, an example of an argument to a function or method, or an example of an object we might call the method on. For example, tbl refers to a table, array refers to an array, and num refers to a number. array.item(0) is an example call for the method item, and in that example, array is the name previously given to some array.

Name Chapter Description
Table() 6 Create an empty table, usually to extend with data
Table().read_table(filename) 6 Create a table from a data file
tbl.with_columns(name, values) tbl.with_columns(n1, v1, n2, v2,...) 6 A table with an additional or replaced column or columns. name is a string for the name of a column, values is an array
tbl.column(column_name_or_index) 6 The values of a column (an array)
tbl.num_rows 6 Compute the number of rows in a table
tbl.num_columns 6 Compute the number of columns in a table
tbl.labels 6 Lists the column labels in a table, col2, ...) 6 Create a copy of a table with only some of the columns. Each column is the column name or index.
tbl.drop(col1, col2, ...) 6 Create a copy of a table without some of the columns. Each column is the column name or index.
tbl.relabel(old_label, new_label) 6 Modifies the existing table in place, changing the column heading in the first argument to the second
tbl.relabeled(old_label, new_label) 6 Returns a new table with the column heading in the first argument changed to the second
tbl.sort(column_name_or_index) 6.1 Create a copy of a table sorted by the values in a column. Defaults to ascending order unless "descending = True" is included
tbl.where(column, predicate) 6.2 Create a copy of a table with only the rows that match some predicate See Table.where predicates below
tbl.take(row_indices) 6.2 A table with only the rows at the given indices. row_indices is an array of indices.
tbl.scatter(x_column, y_column) 7 Draws a scatter plot consisting of one point for each row of the table. Note that x_column and y_column must be strings specifying column names.
tbl.plot(x_column, y_column) 7 Draw a line graph consisting of one point for each row of the table.
tbl.barh(categories) tbl.barh(categories, values) 7.1 Displays a bar chart with bars for each category in a column, with height proportional to the corresponding frequency. values argument unnecessary if table has only a column of categories and a column of values.
tbl.hist(column, unit, bins) 7.2 Generates a histogram of the numerical values in a column. unit and bins are optional arguments, used to label the axes and group the values into intervals (bins), respectively. Bins have the form [a, b).
tbl.apply(function, column) 8.1 Returns an array of values resulting from applying a function to each item in a column., func) 8.2 Group rows by unique values or combinations of values in a column(s). Multiple columns must be entered in array or list form. Other values aggregated by count (default) or optional argument func.
tbl.pivot(col1, col2, vals, collect) tbl.pivot(col1, col2) 8.3 A pivot table where each unique value in col1 has its own column and each unique value in ccol2 has its own row. Count or aggregate values from a third column, collect with some function. Default vals and collect return counts in cells.
tblA.join(colA, tblB, colB) tblA.join(colA, tblB) 8.4 Generate a table with the columns of tblA and tblB, containing rows for all values of a column that appear in both tables. Default colB is colA. colA and colB must be strings specifying column names.
tbl.sample(n) tbl.sample(n, with_replacement) 10 A new table where n rows are randomly sampled from the original table. Default is with replacement. For sampling without replacement, use argument with_replacement=False. For a non-uniform sample, provide a third argument weights=distribution where distribution is an array or list containing the probability of each row.
proportions_from_distribution 11.1 Returns a new table with an additional column whose values correspond to a random sample (of specified size) based on proportions in a specified column.

Array Functions and Methods

Name Chapter Description
max(array) 3.3 Returns the maximum value of an array
min(array) 3.3 Returns the minimum value of an array
sum(array) 3.3 Returns the sum of the values in an array
abs(num), np.abs(array) 3.3 Take the absolute value of number or each number in an array.
round(num), np.round(array) 3.3 Round number or array of numbers to the nearest integer.
len(array) 3.3 Returns the length (number of elements) of an array
make_array(val1, val2, ...) 5 Makes a numpy array with the values passed in
np.average(array) np.mean(array) 5.1 Returns the mean value of an array
np.diff(array) 5.1 Returns a new array of size len(arr)-1 with elements equal to the difference between adjacent elements; val_2 - val_1, val_3 - val_2, etc.
np.sqrt(array) 5.1 Returns an array with the square root of each element
np.arange(start, stop, step) np.arange(start, stop) np.arange(stop) 5.2 An array of numbers starting with start, going up in increments of step, and going up to but excluding stop. When start and/or step are left out, default values are used in their place. Default step is 1; default start is 0.
array.item(index) 5.3 Returns the i-th item in an array (remember Python indices start at 0!)
np.random.choice(array, n) np.random.choice(array) 9 Picks one (by default) or some number 'n' of items from an array at random. By default, with replacement.
np.count_nonzero(array) 9 Returns the number of non-zero (or True) elements in an array.
np.append(array, item) 9.2 Returns a copy of the input array with item (must be the same type as the other entries in the array) appended to the end.
percentile(percentile, array) 12.1 Returns the corresponding percentile of an array.