9.4.7 Rational approximation

Floating point numbers are considered approximations, while integers and rational numbers are considered exact. The float2rational or exact command finds a rational approximation to a floating point number.

• float2rational takes expr, an expression.
• float2rational(expr) returns expr with all the floating point numbers in expr replaced by rational numbers; any floating point number x is replaced by a rational r with |r−x|<ε, where ε is given by epsilon in the CAS configuration (see Section 2.5.7, item 2.5.7).

Examples