eager_method.cc 5.8 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
/* Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License. */
// disable numpy compile error
#include <Python.h>

#include <string>
#include <vector>

#include "pybind11/numpy.h"
#include "pybind11/pybind11.h"

#include "paddle/fluid/eager/api/all.h"
#include "paddle/fluid/eager/autograd_meta.h"
22
#include "paddle/fluid/eager/utils.h"
23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
#include "paddle/fluid/memory/allocation/allocator.h"
#include "paddle/fluid/memory/memcpy.h"
#include "paddle/fluid/platform/enforce.h"
#include "paddle/fluid/pybind/eager.h"
#include "paddle/fluid/pybind/eager_utils.h"
#include "paddle/fluid/pybind/exception.h"
#include "paddle/pten/common/data_type.h"
#include "paddle/pten/core/convert_utils.h"
#include "paddle/pten/core/dense_tensor.h"
#include "paddle/pten/include/core.h"
namespace paddle {
namespace pybind {

extern PyTypeObject* pEagerTensorType;

static PyObject* eager_tensor_method_numpy(EagerTensorObject* self,
                                           PyObject* args, PyObject* kwargs) {
J
Jiabin Yang 已提交
40 41
  EAGER_SYNC_TRY
  if (!self->eager_tensor.initialized()) {
42 43 44
    Py_INCREF(Py_None);
    return Py_None;
  }
J
Jiabin Yang 已提交
45 46 47
  auto tensor_dims = self->eager_tensor.shape();
  auto numpy_dtype = TensorDtype2NumpyDtype(self->eager_tensor.type());
  auto sizeof_dtype = pten::DataTypeSize(self->eager_tensor.type());
48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
  Py_intptr_t py_dims[paddle::framework::DDim::kMaxRank];
  Py_intptr_t py_strides[paddle::framework::DDim::kMaxRank];
  size_t numel = 1;
  for (int i = tensor_dims.size() - 1; i >= 0; --i) {
    py_dims[i] = static_cast<size_t>(tensor_dims[i]);
    py_strides[i] = sizeof_dtype * numel;
    numel *= py_dims[i];
  }
  auto& api = pybind11::detail::npy_api::get();
  PyObject* array = api.PyArray_NewFromDescr_(
      api.PyArray_Type_, api.PyArray_DescrFromType_(numpy_dtype),
      tensor_dims.size(), py_dims, py_strides, nullptr,
      pybind11::detail::npy_api::NPY_ARRAY_ALIGNED_ |
          pybind11::detail::npy_api::NPY_ARRAY_WRITEABLE_,
      nullptr);

J
Jiabin Yang 已提交
64
  if (self->eager_tensor.is_cpu()) {
65
    auto dense_tensor =
J
Jiabin Yang 已提交
66
        std::dynamic_pointer_cast<pten::DenseTensor>(self->eager_tensor.impl());
67 68 69 70 71 72
    platform::CPUPlace place;
    // deep copy
    paddle::memory::Copy(place, reinterpret_cast<void*>(
                                    pybind11::detail::array_proxy(array)->data),
                         place, dense_tensor->data(), sizeof_dtype * numel);
#if defined(PADDLE_WITH_CUDA)
J
Jiabin Yang 已提交
73
  } else if (self->eager_tensor.is_cuda()) {
74
    auto dense_tensor =
J
Jiabin Yang 已提交
75
        std::dynamic_pointer_cast<pten::DenseTensor>(self->eager_tensor.impl());
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92

    paddle::platform::GpuMemcpySync(
        pybind11::detail::array_proxy(array)->data, dense_tensor->data(),
        pten::DataTypeSize(dense_tensor->dtype()) * dense_tensor->numel(),
        cudaMemcpyDeviceToHost);
#endif
  } else {
    PADDLE_THROW(platform::errors::InvalidArgument(
        "Tensor.numpy() only support cpu tensor."));
    Py_INCREF(Py_None);
    return Py_None;
  }

  return array;
  EAGER_CATCH_AND_THROW_RETURN_NULL
}

93 94 95
static PyObject* eager_tensor_method__is_initialized(EagerTensorObject* self,
                                                     PyObject* args,
                                                     PyObject* kwargs) {
J
Jiabin Yang 已提交
96 97
  EAGER_SYNC_TRY
  return ToPyObject(self->eager_tensor.initialized());
98 99 100
  EAGER_CATCH_AND_THROW_RETURN_NULL
}

101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134
static PyObject* eager_tensor_method__copy_to(EagerTensorObject* self,
                                              PyObject* args,
                                              PyObject* kwargs) {
  EAGER_SYNC_TRY
  bool blocking = CastPyArg2AttrBoolean(PyTuple_GET_ITEM(args, 0), 0);
  auto place = CastPyArg2Place(PyTuple_GET_ITEM(args, 1), 1);
  auto cp_tensor =
      self->eager_tensor.copy_to(pten::TransToPtenBackend(place), blocking);
  egr::EagerUtils::autograd_meta(&cp_tensor)->SetStopGradient(true);
  egr::EagerUtils::autograd_meta(&cp_tensor)
      ->SetPersistable(
          egr::EagerUtils::autograd_meta(&(self->eager_tensor))->Persistable());
  return ToPyObject(cp_tensor);
  EAGER_CATCH_AND_THROW_RETURN_NULL
}

static PyObject* eager_tensor_method_copy_(EagerTensorObject* self,
                                           PyObject* args, PyObject* kwargs) {
  EAGER_SYNC_TRY
  egr::EagerTensor src_tensor =
      CastPyArg2EagerTensor(PyTuple_GET_ITEM(args, 0), 0);
  bool blocking = CastPyArg2AttrBoolean(PyTuple_GET_ITEM(args, 1), 1);
  self->eager_tensor.copy_(src_tensor, blocking);
  egr::EagerUtils::autograd_meta(&(self->eager_tensor))
      ->SetStopGradient(
          egr::EagerUtils::autograd_meta(&(src_tensor))->StopGradient());
  egr::EagerUtils::autograd_meta(&(self->eager_tensor))
      ->SetPersistable(
          egr::EagerUtils::autograd_meta(&(src_tensor))->Persistable());
  Py_INCREF(Py_None);
  return Py_None;
  EAGER_CATCH_AND_THROW_RETURN_NULL
}

135 136 137 138
PyMethodDef variable_methods[] = {
    {"numpy", (PyCFunction)(void (*)(void))eager_tensor_method_numpy,
     METH_VARARGS | METH_KEYWORDS, NULL},
    {"_is_initialized",
139 140 141 142 143
     (PyCFunction)(void (*)(void))eager_tensor_method__is_initialized,
     METH_VARARGS | METH_KEYWORDS, NULL},
    {"_copy_to", (PyCFunction)(void (*)(void))eager_tensor_method__copy_to,
     METH_VARARGS | METH_KEYWORDS, NULL},
    {"copy_", (PyCFunction)(void (*)(void))eager_tensor_method_copy_,
144 145 146 147 148
     METH_VARARGS | METH_KEYWORDS, NULL},
    {NULL, NULL, 0, NULL}};

}  // namespace pybind
}  // namespace paddle