File: te_wrapper.h

package info (click to toggle)
pytorch-cuda 2.6.0%2Bdfsg-7
  • links: PTS, VCS
  • area: contrib
  • in suites: forky, sid, trixie
  • size: 161,620 kB
  • sloc: python: 1,278,832; cpp: 900,322; ansic: 82,710; asm: 7,754; java: 3,363; sh: 2,811; javascript: 2,443; makefile: 597; ruby: 195; xml: 84; objc: 68
file content (44 lines) | stat: -rw-r--r-- 1,145 bytes parent folder | download | duplicates (3)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
#pragma once

#include <torch/csrc/jit/tensorexpr/codegen.h>
#include <torch/csrc/jit/tensorexpr/ir.h>
#include <torch/csrc/jit/tensorexpr/ir_simplifier.h>
#include <torch/csrc/jit/tensorexpr/llvm_codegen.h>
#include <torch/csrc/jit/tensorexpr/loopnest.h>

namespace torch::jit {

class TEWrapper {
 public:
  TEWrapper() = default;
  void call(const std::vector<void*>& args);

  template <typename ExpectedType>
  bool checkInput(const at::Tensor& t) {
#ifdef TORCH_ENABLE_LLVM
    return t.is_contiguous() && t.dtype().Match<ExpectedType>();
#else
    return false;
#endif
  }

#ifdef TORCH_ENABLE_LLVM
  void update(std::unique_ptr<tensorexpr::LLVMCodeGen>&& cg_);
#endif

 private:
#ifdef TORCH_ENABLE_LLVM
  std::unique_ptr<tensorexpr::LLVMCodeGen> cg;
#endif
};

std::shared_ptr<TEWrapper> createDiv();
std::shared_ptr<TEWrapper> createLogit();
std::shared_ptr<TEWrapper> createRelu();
std::shared_ptr<TEWrapper> createTanh();
std::shared_ptr<TEWrapper> createSigmoid();
std::shared_ptr<TEWrapper> createSignedLog1p();
std::shared_ptr<TEWrapper> createClamp();
std::shared_ptr<TEWrapper> createClampNanToNum();

} // namespace torch::jit