[][src]Module tensorflow::expr

This module builds computation graphs.

This module is unfinished.

Structs

Add

Expression resulting from adding two subexpressions.

Assign

Expression that assigns a value to a variable.

Compiler

A Compiler compiles Exprs to Operations.

Constant

Expression for a constant.

Div

Expression resulting from dividing two subexpressions.

Expr

A operation in an expression tree, which is a thin wrapper around an ExprImpl.

Mul

Expression resulting from multiplying two subexpressions.

Neg

Expression resulting from negation of an expression.

Placeholder

Expression for a placeholder.

Rem

Expression resulting from taking a modulus.

Sub

Expression resulting from subtracting two subexpressions.

TruncateDiv

Expression that assigns a value to a variable.

Variable

Expression for a variable.

Enums

OpLevel

Denotes operator precedence. Used for displaying expressions as strings.

ShapeHint

Enum of an expr's possible shape states

Traits

AnyExpr

An AnyExpr is just an Expr<T> for some unknown T. Clients should not implement this.

ExprImpl

Trait implemented by all expression types. Most users will want to store an Expr instead.