25 #ifndef TVM_TE_AUTODIFF_H_ 26 #define TVM_TE_AUTODIFF_H_ 56 Tensor
Jacobian(
const Tensor& output,
const Tensor& input);
90 TVM_DLL Array<Tensor>
Gradient(
const Tensor& output,
const Array<Tensor>& inputs,
91 const Tensor& head = Tensor());
96 #endif // TVM_TE_AUTODIFF_H_ runtime implementation for LibTorch/TorchScript.
Definition: analyzer.h:36
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.
A managed object in the TVM runtime.
Tensor VectorJacobianProduct(const Tensor &output, const Tensor &input, const Tensor &head)
The building block for reverse-mode AD.
Array< Tensor > Gradient(const Tensor &output, const Array< Tensor > &inputs, const Tensor &head=Tensor())
Perform reverse mode automatic differentiation.