imperative.cc 13.3 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
/* Copyright (c) 2018 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. */

#include "paddle/fluid/pybind/imperative.h"
16

17
#include <Python.h>
18 19 20 21
#include <pybind11/chrono.h>
#include <pybind11/complex.h>
#include <pybind11/functional.h>
#include <pybind11/stl.h>
22
#include <memory>
23 24
#include <unordered_map>
#include <utility>
25

26
#include "paddle/fluid/framework/block_desc.h"
27 28
#include "paddle/fluid/imperative/layer.h"
#include "paddle/fluid/imperative/profiler.h"
29
#include "paddle/fluid/imperative/tracer.h"
M
minqiyang 已提交
30
#include "paddle/fluid/imperative/type_defs.h"
31

32 33
#include "paddle/fluid/pybind/pybind_boost_headers.h"

34 35 36
namespace paddle {
namespace pybind {

37 38
namespace py = ::pybind11;

39 40 41 42
class Layer : public imperative::Layer {
 public:
  using imperative::Layer::Layer;  // Inherit constructors

43 44 45 46 47
  std::vector<std::shared_ptr<imperative::VarBase>> Forward(
      const std::vector<std::shared_ptr<imperative::VarBase>> &inputs)
      override {
    PYBIND11_OVERLOAD(std::vector<std::shared_ptr<imperative::VarBase>>, Layer,
                      Forward,
48 49 50 51 52 53 54 55 56 57 58
                      inputs);  // NOLINT
  }
};

class PYBIND11_HIDDEN PyOpBase : public imperative::OpBase {
 public:
  using imperative::OpBase::OpBase;  // Inherit constructors

  PyOpBase(const std::string &name) : OpBase(name) {}
};

59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152
// Function like obj.attr_name in Python.
static PyObject *GetPythonAttribute(PyObject *obj, const char *attr_name) {
  // NOTE(zjl): PyObject_GetAttrString would return nullptr when attr_name
  // is not inside obj, but it would also set the error flag of Python.
  // If the error flag is set in C++, C++ code would not raise Exception,
  // but Python would raise Exception once C++ call ends.
  // To avoid unexpected Exception raised in Python, we check whether
  // attribute exists before calling PyObject_GetAttrString.
  //
  // Caution: PyObject_GetAttrString would increase reference count of PyObject.
  // Developer should call Py_DECREF manually after the attribute is not used.
  if (PyObject_HasAttrString(obj, attr_name)) {
    return PyObject_GetAttrString(obj, attr_name);
  } else {
    return nullptr;
  }
}

template <typename T>
static T PyObjectCast(PyObject *obj) {
  try {
    return py::cast<T>(py::handle(obj));
  } catch (py::cast_error &) {
    PADDLE_THROW("Python object is not type of %s", typeid(T).name());
  }
}

// NOTE(zjl): py::handle is a very light wrapper of PyObject *.
// Unlike py::object, py::handle does not change reference count of PyObject *.
static std::vector<std::shared_ptr<imperative::VarBase>>
GetVarBaseListFromPyHandle(const py::handle &handle) {
  PyObject *py_obj = handle.ptr();  // get underlying PyObject
  // Python None is not nullptr in C++!
  if (!py_obj || py_obj == Py_None) {
    return {};
  }

  const char *kIVarField = "_ivar";
  PyObject *py_ivar = GetPythonAttribute(py_obj, kIVarField);
  std::vector<std::shared_ptr<imperative::VarBase>> result;

  if (py_ivar) {  // Variable
    result.emplace_back(
        PyObjectCast<std::shared_ptr<imperative::VarBase>>(py_ivar));
    Py_DECREF(py_ivar);
  } else if (PyList_Check(py_obj)) {  // List of Variable
    size_t len = PyList_GET_SIZE(py_obj);
    result.reserve(len);
    for (size_t i = 0; i < len; ++i) {
      PyObject *py_ivar =
          PyObject_GetAttrString(PyList_GET_ITEM(py_obj, i), kIVarField);
      PADDLE_ENFORCE_NOT_NULL(py_ivar);
      result.emplace_back(
          PyObjectCast<std::shared_ptr<imperative::VarBase>>(py_ivar));
      Py_DECREF(py_ivar);
    }
  } else if (PyTuple_Check(py_obj)) {  // Tuple of Variable
    size_t len = PyTuple_GET_SIZE(py_obj);
    result.reserve(len);
    for (size_t i = 0; i < len; ++i) {
      PyObject *py_ivar =
          PyObject_GetAttrString(PyTuple_GET_ITEM(py_obj, i), kIVarField);
      PADDLE_ENFORCE_NOT_NULL(py_ivar);
      result.emplace_back(
          PyObjectCast<std::shared_ptr<imperative::VarBase>>(py_ivar));
      Py_DECREF(py_ivar);
    }
  } else {
    PADDLE_THROW(
        "unsupported type %s, must be Variable, List[Variable] or "
        "tuple[Variable]",
        py::str(handle));
  }

  PADDLE_ENFORCE(PyErr_Occurred() == nullptr,
                 py::str(py::handle(PyErr_Occurred())));

  return result;
}

using PyVarBaseMap = std::unordered_map<std::string, py::handle>;

static imperative::VarBasePtrMap ConvertToVarBasePtrMap(
    const PyVarBaseMap &map) {
  imperative::VarBasePtrMap result;
  for (auto &pair : map) {
    auto var_vec = GetVarBaseListFromPyHandle(pair.second);
    if (!var_vec.empty()) {
      result.emplace(pair.first, std::move(var_vec));
    }
  }
  return result;
}

153
// Bind Methods
154 155 156 157
void BindImperative(pybind11::module *m_ptr) {
  auto &m = *m_ptr;

  py::class_<imperative::detail::BackwardStrategy> backward_strategy(
158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
      m, "BackwardStrategy", R"DOC(

    BackwardStrategy is a descriptor of a how to run the backward process. Now it has:

    1. :code:`sort_sum_gradient`, which will sum the gradient by the reverse order of trace.

    Examples:

     .. code-block:: python
        import numpy as np
        import paddle.fluid as fluid
        from paddle.fluid import FC

        x = np.ones([2, 2], np.float32)
        with fluid.dygraph.guard():
            inputs2 = []
            for _ in range(10):
                inputs2.append(fluid.dygraph.base.to_variable(x))
            ret2 = fluid.layers.sums(inputs2)
            loss2 = fluid.layers.reduce_sum(ret2)
            backward_strategy = fluid.dygraph.BackwardStrategy()
            backward_strategy.sort_sum_gradient = True
            loss2.backward(backward_strategy)
      )DOC");
182 183 184 185 186 187 188 189 190 191 192 193 194 195 196
  backward_strategy.def(py::init())
      .def_property("sort_sum_gradient",
                    [](const imperative::detail::BackwardStrategy &self) {
                      return self.sorted_sum_gradient_;
                    },
                    [](imperative::detail::BackwardStrategy &self,
                       bool sorted_sum_gradient) {
                      self.sorted_sum_gradient_ = sorted_sum_gradient;
                    });

  m.def("start_imperative_gperf_profiler",
        []() { imperative::StartProfile(); });

  m.def("stop_imperative_gperf_profiler", []() { imperative::StopProfile(); });

197 198
  py::class_<imperative::VarBase, std::shared_ptr<imperative::VarBase>>(
      m, "VarBase", R"DOC()DOC")
199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261
      .def(
          py::init<const std::string &, paddle::framework::proto::VarType::Type,
                   const std::vector<int64_t>, const paddle::platform::CPUPlace,
                   bool, bool>())
      .def(
          py::init<const std::string &, paddle::framework::proto::VarType::Type,
                   const std::vector<int64_t>,
                   const paddle::platform::CUDAPlace, bool, bool>())
      .def("_run_backward",
           [](imperative::VarBase &self,
              const imperative::detail::BackwardStrategy &bckst) {
             self.RunBackward(bckst);
           })
      .def("_grad_name", &imperative::VarBase::GradName)
      .def("_grad_value", &imperative::VarBase::GradValue)
      .def("_clear_gradient", &imperative::VarBase::ClearGradient)
      .def("_grad_ivar",
           [](const imperative::VarBase &self) { return self.grads_; },
           py::return_value_policy::reference)
      .def("_copy_to",
           [](const imperative::VarBase &self, const platform::CPUPlace &place,
              bool blocking) {
             return self.NewVarBase(place, blocking).release();
           },
           py::return_value_policy::take_ownership)
      .def("_copy_to",
           [](const imperative::VarBase &self, const platform::CUDAPlace &place,
              bool blocking) {
             return self.NewVarBase(place, blocking).release();
           },
           py::return_value_policy::take_ownership)
      .def("value",
           [](const imperative::VarBase &self) { return self.var_.get(); },
           py::return_value_policy::reference)
      .def_property("name", &imperative::VarBase::Name,
                    &imperative::VarBase::SetName)
      .def_property_readonly("shape", &imperative::VarBase::Shape)
      .def_property_readonly("dtype", &imperative::VarBase::DataType)
      .def_property("persistable", &imperative::VarBase::IsPersistable,
                    &imperative::VarBase::SetPersistable)
      .def_property("stop_gradient", &imperative::VarBase::IsStopGradient,
                    &imperative::VarBase::SetStopGradient);

  py::class_<imperative::OpBase, PyOpBase>(m, "OpBase", R"DOC()DOC")
      .def(py::init<const std::string &>())
      .def("register_backward_hooks",
           [](imperative::OpBase &self, const py::object &callable) {
             self.RegisterBackwardHooks(callable);
           })
      .def_property("_trace_id",
                    [](const imperative::OpBase &self) {
                      py::gil_scoped_release release;
                      return self.trace_id_;
                    },
                    [](imperative::OpBase &self, int trace_id) {
                      py::gil_scoped_release release;
                      self.trace_id_ = trace_id;
                    },
                    py::return_value_policy::reference)
      .def_property_readonly("type", &imperative::OpBase::Type);

  py::class_<imperative::Layer, Layer /* <--- trampoline*/> layer(m, "Layer");
  layer.def(py::init<>())
262 263 264 265 266
      .def("forward",
           [](imperative::Layer &self,
              const std::vector<std::shared_ptr<imperative::VarBase>> &inputs) {
             return self.Forward(inputs);
           });
267

268 269 270 271 272
  // NOTE(zjl): Tracer use PyVarBaseMap as its parameter but not VarBasePtrMap.
  // We call Python C-API to convert PyVarBaseMap to VarBasePtrMap, instead
  // making conversion in Python code. This speed up Tracer.trace() about 6%
  // in ptb model and make time cost in Python to be nearly zero.
  py::class_<imperative::Tracer>(m, "Tracer", "")
273
      .def("__init__",
274
           [](imperative::Tracer &self, framework::BlockDesc *root_block) {
M
minqiyang 已提交
275
             new (&self) imperative::Tracer(root_block);
276
           })
M
minqiyang 已提交
277
      .def("trace",
278
           [](imperative::Tracer &self, imperative::OpBase *op,
279
              const PyVarBaseMap &inputs, const PyVarBaseMap &outputs,
280
              framework::AttributeMap attrs_map,
M
minqiyang 已提交
281 282
              const platform::CPUPlace expected_place,
              const bool stop_gradient = false) {
283 284 285 286 287 288 289
             auto ins = ConvertToVarBasePtrMap(inputs);
             auto outs = ConvertToVarBasePtrMap(outputs);
             {
               py::gil_scoped_release release;
               self.Trace(op, std::move(ins), &outs, attrs_map, expected_place,
                          stop_gradient);
             }
M
minqiyang 已提交
290
           })
291
      .def("trace", [](imperative::Tracer &self, imperative::OpBase *op,
292
                       const PyVarBaseMap &inputs, const PyVarBaseMap &outputs,
293 294 295
                       framework::AttributeMap attrs_map,
                       const platform::CUDAPlace expected_place,
                       const bool stop_gradient = false) {
296 297 298 299 300 301 302
        auto ins = ConvertToVarBasePtrMap(inputs);
        auto outs = ConvertToVarBasePtrMap(outputs);
        {
          py::gil_scoped_release release;
          self.Trace(op, std::move(ins), &outs, attrs_map, expected_place,
                     stop_gradient);
        }
303
      });
304 305

  // define parallel context
306 307 308
  py::class_<imperative::ParallelStrategy> parallel_strategy(
      m, "ParallelStrategy", "");
  parallel_strategy.def(py::init())
309 310
      .def_property(
          "nranks",
311 312
          [](const imperative::ParallelStrategy &self) { return self.nranks_; },
          [](imperative::ParallelStrategy &self, int nranks) {
313 314 315
            self.nranks_ = nranks;
          })
      .def_property("local_rank",
316
                    [](const imperative::ParallelStrategy &self) {
317 318
                      return self.local_rank_;
                    },
319
                    [](imperative::ParallelStrategy &self, int local_rank) {
320 321 322 323
                      self.local_rank_ = local_rank;
                    })
      .def_property(
          "trainer_endpoints",
324
          [](const imperative::ParallelStrategy &self) {
325 326
            return self.trainer_endpoints_;
          },
327
          [](imperative::ParallelStrategy &self, std::vector<std::string> eps) {
328 329 330
            self.trainer_endpoints_ = eps;
          })
      .def_property("current_endpoint",
331
                    [](const imperative::ParallelStrategy &self) {
332 333
                      return self.current_endpoint_;
                    },
334 335
                    [](imperative::ParallelStrategy &self,
                       const std::string &ep) { self.current_endpoint_ = ep; });
336
#if defined(PADDLE_WITH_CUDA) && !defined(_WIN32)
337 338
  py::class_<imperative::NCCLParallelContext> nccl_ctx(m,
                                                       "NCCLParallelContext");
339 340

  nccl_ctx
341 342 343
      .def(py::init<const imperative::ParallelStrategy &,
                    const platform::CUDAPlace &>())
      .def("init", [](imperative::NCCLParallelContext &self) { self.Init(); });
344
#endif
345 346 347 348
}

}  // namespace pybind
}  // namespace paddle