25 #ifndef TVM_TE_AUTODIFF_H_
26 #define TVM_TE_AUTODIFF_H_
Reference to PrimExprNode.
Definition: expr.h:115
Array, container representing a contiguous sequence of ObjectRefs.
Definition: array.h:289
Tensor structure representing a possible input, or intermediate computation result.
Definition: tensor.h:102
a named variable in TIR
Definition: var.h:89
Array< Tensor > Gradient(const Tensor &output, const Array< Tensor > &inputs, const Tensor &head=Tensor())
Perform reverse mode automatic differentiation.
Tensor VectorJacobianProduct(const Tensor &output, const Tensor &input, const Tensor &head)
The building block for reverse-mode AD.
PrimExpr Derivative(const PrimExpr &expr, const Var &var)
Take the derivative of the expression with respect to the given variable.
Tensor Jacobian(const Tensor &output, const Tensor &input)
Get the tensor representing the Jacobian of the output with respect to the input.
Var var(std::string name_hint, DataType t=DataType::Int(32))
Construct a new Var expression.
runtime implementation for LibTorch/TorchScript.
Definition: analyzer.h:36
A managed object in the TVM runtime.