adam_op_npu.cc 7.3 KB
Newer Older
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 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
/* 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. */

#include <memory>
#include <string>

#include "paddle/fluid/operators/npu_op_runner.h"
#include "paddle/fluid/operators/optimizers/adam_op.h"

namespace paddle {
namespace operators {

using Tensor = framework::Tensor;
using LoDTensor = framework::LoDTensor;

template <typename DeviceContext, typename T>
class AdamNPUKernel : public framework::OpKernel<T> {
 public:
  void Compute(const framework::ExecutionContext& ctx) const override {
    const auto* param_var = ctx.InputVar("Param");
    PADDLE_ENFORCE_EQ(param_var->IsType<framework::LoDTensor>(), true,
                      platform::errors::InvalidArgument(
                          "The Var(%s)'s type should be LoDTensor, "
                          "but the received is %s",
                          ctx.InputNames("Param").front(),
                          framework::ToTypeName(param_var->Type())));
    T epsilon = static_cast<T>(ctx.Attr<float>("epsilon"));
    auto* param = ctx.Input<LoDTensor>("Param");
    auto* grad_var = ctx.InputVar("Grad");
    PADDLE_ENFORCE_EQ(grad_var->IsType<framework::LoDTensor>(), true,
                      platform::errors::InvalidArgument(
                          "The Grad(%s)'s type should be LoDTensor, "
                          "but the received is %s",
                          ctx.InputNames("Grad").front(),
                          framework::ToTypeName(param_var->Type())));
    auto* grad = ctx.Input<LoDTensor>("Grad");
    auto* mom1 = ctx.Input<LoDTensor>("Moment1");
    auto* mom2 = ctx.Input<LoDTensor>("Moment2");
    auto* lr = ctx.Input<LoDTensor>("LearningRate");

    auto* beta1_pow = ctx.Input<LoDTensor>("Beta1Pow");
    auto* beta2_pow = ctx.Input<LoDTensor>("Beta2Pow");

    auto* param_out = ctx.Output<LoDTensor>("ParamOut");
    auto* mom1_out = ctx.Output<LoDTensor>("Moment1Out");
    auto* mom2_out = ctx.Output<LoDTensor>("Moment2Out");
    auto* beta1_pow_out = ctx.Output<LoDTensor>("Beta1PowOut");
    auto* beta2_pow_out = ctx.Output<LoDTensor>("Beta2PowOut");

    param_out->mutable_data<T>(ctx.GetPlace());
    mom1_out->mutable_data<T>(ctx.GetPlace());
    mom2_out->mutable_data<T>(ctx.GetPlace());
64 65 66

    // NOTE(zhiqiu): beta1_pow and beta2_pow may on CPU and not transform place.
    if (beta1_pow->place() == platform::CPUPlace()) {
67
      T beta1 = *beta1_pow->data<T>();
P
pangyoki 已提交
68 69
      // `mutable_data` operation needs to be done after getting data
      beta1_pow_out->mutable_data<T>(ctx.GetPlace());
70
      FillNpuTensorWithConstant<T>(beta1_pow_out, beta1);
P
pangyoki 已提交
71 72
    } else {
      beta1_pow_out->mutable_data<T>(ctx.GetPlace());
73 74
    }
    if (beta2_pow->place() == platform::CPUPlace()) {
75
      T beta2 = *beta2_pow->data<T>();
P
pangyoki 已提交
76
      beta2_pow_out->mutable_data<T>(ctx.GetPlace());
77
      FillNpuTensorWithConstant<T>(beta2_pow_out, beta2);
P
pangyoki 已提交
78 79
    } else {
      beta2_pow_out->mutable_data<T>(ctx.GetPlace());
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

    T beta1 = static_cast<T>(ctx.Attr<float>("beta1"));
    if (ctx.HasInput("Beta1Tensor")) {
      auto* beta1_tensor = ctx.Input<framework::Tensor>("Beta1Tensor");
      PADDLE_ENFORCE_EQ(beta1_tensor->numel(), 1,
                        platform::errors::InvalidArgument(
                            "Input(Beta1Tensor) size must be 1, but get %d",
                            beta1_tensor->numel()));
      beta1 = static_cast<T>(GetAttrFromTensor(beta1_tensor));
    }
    T beta2 = static_cast<T>(ctx.Attr<float>("beta2"));
    if (ctx.HasInput("Beta2Tensor")) {
      auto* beta2_tensor = ctx.Input<framework::Tensor>("Beta2Tensor");
      PADDLE_ENFORCE_EQ(beta2_tensor->numel(), 1,
                        platform::errors::InvalidArgument(
                            "Input(Beta2Tensor) size must be 1, but get %d",
                            beta2_tensor->numel()));
      beta2 = static_cast<T>(GetAttrFromTensor(beta2_tensor));
    }
    VLOG(3) << "beta1_pow.numel() : " << beta1_pow->numel()
            << "beta2_pow.numel() : " << beta2_pow->numel();
    VLOG(3) << "param.numel(): " << param->numel();

    PADDLE_ENFORCE_EQ(beta1_pow_out->numel(), 1,
                      platform::errors::InvalidArgument(
                          "beta1 pow output size should be 1, but received "
                          "value is:%d.",
                          beta1_pow_out->numel()));

    PADDLE_ENFORCE_EQ(beta2_pow_out->numel(), 1,
                      platform::errors::InvalidArgument(
                          "beta2 pow output size should be 1, but received "
                          "value is:%d.",
                          beta2_pow_out->numel()));

    // reshape
    Tensor beta1_tensor(framework::proto::VarType::FP32);
118 119
    beta1_tensor.mutable_data<T>({1}, ctx.GetPlace());
    FillNpuTensorWithConstant<T>(&beta1_tensor, beta1);
120
    Tensor beta2_tensor(framework::proto::VarType::FP32);
121 122
    beta2_tensor.mutable_data<T>({1}, ctx.GetPlace());
    FillNpuTensorWithConstant<T>(&beta2_tensor, beta2);
123 124 125

    Tensor epsilon_tensor(framework::proto::VarType::FP32);
    epsilon_tensor.mutable_data<T>({1}, ctx.GetPlace());
126
    FillNpuTensorWithConstant<T>(&epsilon_tensor, epsilon);
127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
    auto stream =
        ctx.template device_context<paddle::platform::NPUDeviceContext>()
            .stream();
    auto runner =
        NpuOpRunner("ApplyAdamD",
                    {
                        *param, *mom1, *mom2, *beta1_pow, *beta2_pow, *lr,
                        beta1_tensor, beta2_tensor, epsilon_tensor, *grad,
                    },
                    {
                        *param_out, *mom1_out, *mom2_out,
                    },
                    {});
    runner.Run(stream);

    // NOTE(zhiqiu): ApplyAdamD updates params inplace, so
    // if param and param_out is not same, we need to do copy.
    if (param_out->data<T>() != param->data<T>()) {
145 146 147
      framework::TensorCopy(
          *param, ctx.GetPlace(),
          ctx.template device_context<platform::DeviceContext>(), param_out);
148 149
    }
    if (mom1_out->data<T>() != mom1->data<T>()) {
150 151 152
      framework::TensorCopy(
          *mom1, ctx.GetPlace(),
          ctx.template device_context<platform::DeviceContext>(), mom1_out);
153 154
    }
    if (mom2_out->data<T>() != mom2->data<T>()) {
155 156 157
      framework::TensorCopy(
          *mom2, ctx.GetPlace(),
          ctx.template device_context<platform::DeviceContext>(), mom2_out);
158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176
    }
    auto runner_m1 =
        NpuOpRunner("Mul", {*beta1_pow, beta1_tensor}, {*beta1_pow_out}, {});
    runner_m1.Run(stream);
    auto runner_m2 =
        NpuOpRunner("Mul", {*beta2_pow, beta2_tensor}, {*beta2_pow_out}, {});
    runner_m2.Run(stream);
  }
};

}  // namespace operators
}  // namespace paddle

namespace ops = paddle::operators;

REGISTER_OP_NPU_KERNEL(
    adam, ops::AdamNPUKernel<paddle::platform::NPUDeviceContext, float>,
    ops::AdamNPUKernel<paddle::platform::NPUDeviceContext,
                       paddle::platform::float16>);