diff --git a/doc/v2/howto/rnn/hrnn_rnn_api_compare_cn.rst b/doc/v2/howto/rnn/hrnn_rnn_api_compare_cn.rst index b05b66415f..67c7b774e9 100644 --- a/doc/v2/howto/rnn/hrnn_rnn_api_compare_cn.rst +++ b/doc/v2/howto/rnn/hrnn_rnn_api_compare_cn.rst @@ -134,7 +134,7 @@ **输入不等长** 是指recurrent_group的多个输入序列，在每个时间步的子序列长度可以不相等。但序列输出时，需要指定与某一个输入的序列信息是一致的。使用\ :red:`targetInlink`\ 可以指定哪一个输入和输出序列信息一致，默认指定第一个输入。 -示例3的配置分别为\ `单层不等长RNN `_\ 和\ `双层不等长RNN `_\ 。 +示例3的配置分别为\ `单层不等长RNN `_\ 和\ `双层不等长RNN `_\ 。 示例3对于单层RNN和双层RNN数据完全相同。 diff --git a/doc/v2/howto/rnn/hrnn_rnn_api_compare_en.rst b/doc/v2/howto/rnn/hrnn_rnn_api_compare_en.rst index e5aa05c117..ae997f0805 100644 --- a/doc/v2/howto/rnn/hrnn_rnn_api_compare_en.rst +++ b/doc/v2/howto/rnn/hrnn_rnn_api_compare_en.rst @@ -1,4 +1,226 @@ +.. _algo_hrnn_rnn_api_compare: + +##################### API comparision between RNN and hierarchical RNN -================================================ +##################### + +This article takes PaddlePaddle's hierarchical RNN unit test as an example. We will use several examples to illestrate the usage of single-layer and hierarchical RNNs. Each example has two model configurations, one for single-layer, and the other for hierarchical RNN. Although the implementations are different, both the two model configurations' effects are the same. All of the examples in this article only describe the API interface of the hierarchical RNN, while we do not use this hierarchical RNN to solve practical problems. If you want to understand the use of hierarchical RNN in specific issues, please refer to \ :ref:`algo_hrnn_demo`\ 。The unit test file used in this article's example is \ `test_RecurrentGradientMachine.cpp `_\ 。 + +Example 1：Hierarchical RNN without Memory between subsequences +================================ + +The classical case in the hierarchical RNN is to perform sequence operations on each time series data in the inner layers seperately. And the sequence operations in the inner layers is independent, that is, it does not need to use Memory. + +In this example, the network configuration of single-layer RNNs and hierarchical RNNs are all to use LSTM as en encoder to compress a word-segmented sentence into a vector. The difference is that, RNN uses a hierarchical RNN model, treating multiple sentences as a whole to use encoder to compress simultaneously. They are completely consistent in their semantic meanings. This pair of semantically identical example configurations is as follows： + +* RNN\: `sequence_layer_group.conf `_ +* Hierarchical RNN\: `sequence_nest_layer_group.conf `_ + + +Reading hierarchical sequence data +---------------- + +Firstly, the original data in this example is as follows \: + +- The original data in this example has 10 samples. Each of the sample includes two components: a lable(all 2 here), and a word-segmented sentence. This data is used by single RNN as well. + +.. literalinclude:: ../../../../paddle/gserver/tests/Sequence/tour_train_wdseg + :language: text + + +- The data for hierarchical RNN has 4 samples. Every sample is seperated by a blank line, while the content of the data is the same as the original data. But as for hierarchical LSTM, the first sample will encode two sentences into two vectors simultaneously. The sentence count dealed simultaneously by this 4 samples are \ :code:`[2, 3, 2, 3]`\ . + +.. literalinclude:: ../../../../paddle/gserver/tests/Sequence/tour_train_wdseg.nest + :language: text + +Secondly, as for these two types of different input data formats, the contrast of different DataProviders are as follows (`sequenceGen.py `_)\： + +.. literalinclude:: ../../../../paddle/gserver/tests/sequenceGen.py + :language: python + :lines: 21-39 + :linenos: + +- This is the DataProvider code for an ordinary single-layer time series. Its description is as follows: + + * DataProvider returns two parts, that are "words" and "label"，as line 19 in the above code. + + - "words" is a list of word table indices corresponding to each word in the sentence in the original data. Its data type is integer_value_sequence, that is integer list. So, "words" is a singler-layer time series in the data. + - "label" is the categorical label of each sentence, whose data type is integer_value. + +.. literalinclude:: ../../../../paddle/gserver/tests/sequenceGen.py + :language: python + :lines: 42-71 + :linenos: + +- As for the same data, the DataProvider code for hierarchical time series. Its description is as follows: + + - DataProvider returns two lists of data, that are "sentences" and "labels", corresponding to the sentences and labels in each group in the original data of hierarchical time series. + - "sentences" comes from the hierarchical time series original data. As it contains every sentences in each group internally, and each sentences are represented by a list of word table indices, so its data type is integer_value_sub_sequence, which is hierarchical time series. + - "labels" is the categorical lable of each sentence, so it is a sigle-layer time series. + + +Model configuration +------------------------------------------ + +Firstly, let's look at the configuration of single-layer RNN. The hightlighted part of line 9 to line 15 is the usage of single-layer RNN. Here we use the pre-defined RNN process function in PaddlePaddle. In this function, for each time step, RNN passes through an LSTM network. + +.. literalinclude:: ../../../../paddle/gserver/tests/sequence_layer_group.conf + :language: python + :lines: 38-63 + :linenos: + :emphasize-lines: 9-15 + + +Secondly, let's look at the model configuration of hierarchical RNN which has the same semantic meaning. \: + +* Most layers in PaddlePaddle do not care about whether the input is time series or not, e.g. \ :code:`embedding_layer`\ . In these layers, every operation is processed on each time step. + +* In the hightlighted part of line 7 to line 26 of this configuration, we transform the hierarchical time series data into single-layer time series data, then process each single-layer time series. + + * Use the function \ :code:`recurrent_group`\ to transform. Input sequences need to be passed in when transforming. As we want to transform hierarchical time series into single-layer sequences, we need to lable the input data as \ :code:`SubsequenceInput`\ . + + * In this example, we disassemble every group of the original data into sentences using \ :code:`recurrent_group`\ . Each of the disassembled sentences passes through an LSTM network. This is equivalent to single-layer RNN configuration. + +* Similar to single-layer RNN configuration, we only use the last vector after the encode of LSTM. So we use the operation of \ :code:`last_seq`\ to \ :code:`recurrent_group`\ . But unlike single-layer RNN, we use the last element of every subsequence, so we need to set \ :code:`agg_level=AggregateLevel.TO_SEQUENCE`\ . + +* Till now, \ :code:`lstm_last`\ has the same result as \ :code:`lstm_last`\ in single-layer RNN configuration. + +.. literalinclude:: ../../../../paddle/gserver/tests/sequence_nest_layer_group.conf + :language: python + :lines: 38-64 + :linenos: + :emphasize-lines: 7-26 + +Example 2：Hierarchical RNN with Memory between subsequences +================================ + +This example is intended to implement two fully-equivalent fully-connected RNNs using single-layer RNN and hierarchical RNN. + +* As for single-layer RNN, input is a full time series, e.g. \ :code:`[4, 5, 2, 0, 9, 8, 1, 4]`\ . + +* As for hierarchical RNN, input is a hierarchical time series which elements are arbitrarily combination of data in single-layer RNN, e.g. \ :code:`[ [4, 5, 2], [0, 9], [8, 1, 4]]`. + +model configuration +------------------ + +We select the different parts between single-layer RNN and hierarchical RNN configurations, to compare and analyze the reason why they have same semantic meanings. + +- single-layer RNN：passes through a simple recurrent_group. For each time step, the current input y and the last time step's output rnn_state pass through a fully-connected layer. + +.. literalinclude:: ../../../../paddle/gserver/tests/sequence_rnn.conf + :language: python + :lines: 36-48 + +- hierarchical RNN, the outer layer's memory is an element. + + - The recurrent_group of inner layer's inner_step is nearly the same as single-layer sequence, except for the case of boot_layer=outer_mem, which means using the outer layer's outer_mem as the initial state for the inner layer's memory. In the outer layer's out_step, outer_mem is the last vector of a subsequence, that is, the whole hierarchical group uses the last vector of the previous subsequence as the initial state for the next subsequence's memory. + - From the aspect of the input data, sentences from single-layer and hierarchical RNN are the same. The only difference is that, hierarchical RNN disassembes the sequence into subsequences. So in the hierarchical RNN configuration, we must use the last element of the previous subsequence as a boot_layer for the memory of the next subsequence, so that it makes no difference with "every time step uses the output of last time step" in the sigle-layer RNN configuration. + +.. literalinclude:: ../../../../paddle/gserver/tests/sequence_nest_rnn.conf + :language: python + :lines: 39-66 + +.. warning:: + Currently PaddlePaddle only supports the case that the lengths of the time series of Memory in each time step are the same. + +Example 3：hierarchical RNN with unequal length inputs +========================== + +.. role:: red + +.. raw:: html + + + +**unequal length inputs** means in the multiple input sequences of recurrent_group, the lengths of subsequences can be unequal. But the output of the sequence, needs to be consistent with one of the input sequences. Using \ :red:`targetInlink`\ can help you specify which of the input sequences and the output sequence can be consistent, by default is the first input. + +The configurations of Example 3 are \ `sequence_rnn_multi_unequalength_inputs `_ \ and \ `sequence_nest_rnn_multi_unequalength_inputs `_\ . + +The data for the configurations of Example 3's single-layer RNN and hierarchical RNN are exactly the same. + +* For the single-layer RNN, the data has two samples, which are \ :code:`[1, 2, 4, 5, 2], [5, 4, 1, 3, 1]`\ and \ :code:`[0, 2, 2, 5, 0, 1, 2], [1, 5, 4, 2, 3, 6, 1]`\ . Each of the data for the single-layer RNN has two group of features. + +* On the basis of the single-layer's data, hierarchical RNN's data randomly adds some partitions. For example, the first sample is transformed to \ :code:`[[0, 2], [2, 5], [0, 1, 2]],[[1, 5], [4], [2, 3, 6, 1]]`\ . + +* You need to pay attention that, PaddlePaddle only supports multiple input hierarchical RNNs that have same amount of subsequences currently. In this example, the two features both have 3 subsequences. Although the length of each subsequence can be different, the amount of subsequences should be the same. + + +model configuration +-------- + +Similar to Example 2's configuration, Example 3's configuration uses single-layer and hierarchical RNN to implement 2 fully-equivalent fully-connected RNNs. + +* single-layer RNN\: + +.. literalinclude:: ../../../../paddle/gserver/tests/sequence_rnn_multi_unequalength_inputs.py + :language: python + :lines: 42-59 + :linenos: + +* hierarchical RNN\ \: + +.. literalinclude:: ../../../../paddle/gserver/tests/sequence_nest_rnn_multi_unequalength_inputs.py + :language: python + :lines: 41-80 + :linenos: + +In the above code, the usage of single-layer and hierarchical RNNs are similar to Example 2, which difference is that it processes 2 inputs simultaneously. As for the hierarchical RNN, the lengths of the 2 input's subsequences are not equal. But we use the parameter \ :code:`targetInlink` \ to set the outper layer's \ :code:`recurrent_group` \ 's output format, so the shape of outer layer's output is the same as the shape of \ :code:`emb2`\ . + + +Glossary +====== + +.. _glossary_memory: + +Memory +------ + +Memory is a concept when PaddlePaddle is implementing RNN. RNN, recurrent neural network, usually requires some dependency between time steps, that is, the neural network in current time step depends on one of the neurons in the neural network in previous time steps, as the following figure shows: + +.. graphviz:: src/glossary_rnn.dot + +The dotted connections in the figure, is the network connections across time steps. When PaddlePaddle is implementing RNN, this connection accross time steps is implemented using a special neural network unit, called Memory. Memory can cache the output of one of the neurons in previous time step, then can be passed to another neuron in next time step. The implementation of an RNN using Memory is as follows: + +.. graphviz:: src/glossary_rnn_with_memory.dot + +With this method, PaddlePaddle can easily determine which outputs should cross time steps, and which should not. + +.. _glossary_timestep: + +time step +------ + +refers to time series + + +.. _glossary_sequence: + +time series +-------- + +Time series is a series of featured data. The order among these featured data is meaningful. So it is a list of features, not a set of features. As for each element of this list, or the featured data in each series, is called a time step. It must be noted that, the concepts of time series and time steps, are not necessarrily related to "time". As long as the "order" in a series of featured data is meaningful, it can be the input of time series. + +For example, in text classification task, we regard a sentence as a time series. So, each word in the sentence can become the index of the word in the word table. So this sentence can be represented as a list of these indices, e.g.:code:`[9, 2, 3, 5, 3]` . + +For a more detailed and accurate definition of the time series, please refer to `Wikipedia of Time series `_ or `Chinese Wikipedia of time series `_ . + +In additioin, Paddle always calls time series as :code:`Sequence` . They are a same concept in Paddle's documentations and APIs. + +.. _glossary_RNN: + +RNN +--- + +In PaddlePaddle's documentations, RNN is usually represented as :code:`Recurrent neural network` . For more information, please refer to `Wikipedia Recurrent neural network `_ or `Chinese Wikipedia `_ . + +In PaddlePaddle, RNN usually means, for the input data of a time series, the neural network between each time steps has a certain relevance. For example, the input of a certain neuron is the output of a certain neuron in the neural network of the last time step. Or, as for each time step, the network structure of the neural network has a directed ring structure. + +.. _glossary_hierarchical_RNN: + +hierarchical RNN +------- + +Hierarchical RNN, as the name suggests, means there is a nested relationship in RNNs. The input data is a time series, but for each of the inner featured data, it is also a time series, namely 2-dimentional array, or, array of array. Hierarchical RNN is a neural network that can process this type of input data. + +For example, the task of text classification of a paragragh, meaning to classify a paragraph of sentences. We can treat a paragraph as an array of sentences, and each sentence is an array of words. This is a type of the input data for the hierarchical RNN. We encode each sentence of this paragraph into a vector using LSTM, then encode each of the encoded vectors into a vector of this paragraph using LSTM. Finally we use this paragraph vector perform classification, which is the neural network structure of this hierarchical RNN. -TBD diff --git a/paddle/fluid/framework/details/broadcast_op_handle.h b/paddle/fluid/framework/details/broadcast_op_handle.h index b329242252..bc3e373488 100644 --- a/paddle/fluid/framework/details/broadcast_op_handle.h +++ b/paddle/fluid/framework/details/broadcast_op_handle.h @@ -29,9 +29,7 @@ namespace framework { namespace details { struct BroadcastOpHandle : public OpHandleBase { - const std::vector &local_scopes_; - const std::vector &places_; - + public: BroadcastOpHandle(const std::vector &local_scopes, const std::vector &places); @@ -41,6 +39,10 @@ struct BroadcastOpHandle : public OpHandleBase { protected: void RunImpl() override; + + private: + const std::vector &local_scopes_; + const std::vector &places_; }; } // namespace details diff --git a/paddle/fluid/framework/details/broadcast_op_handle_test.cc b/paddle/fluid/framework/details/broadcast_op_handle_test.cc index bcd61335be..efc7051582 100644 --- a/paddle/fluid/framework/details/broadcast_op_handle_test.cc +++ b/paddle/fluid/framework/details/broadcast_op_handle_test.cc @@ -90,7 +90,7 @@ struct TestBroadcastOpHandle { op_handle_->AddInput(dummy_var_handle); for (size_t j = 0; j < gpu_list_.size(); ++j) { - op_handle_->dev_ctxes_[gpu_list_[j]] = ctxs_[j].get(); + op_handle_->SetDeviceContext(gpu_list_[j], ctxs_[j].get()); VarHandle* out_var_handle = new VarHandle(2, j, "out", gpu_list_[j]); vars_.emplace_back(out_var_handle); op_handle_->AddOutput(out_var_handle); diff --git a/paddle/fluid/framework/details/computation_op_handle.cc b/paddle/fluid/framework/details/computation_op_handle.cc index ff6d91c1da..7ff0efe093 100644 --- a/paddle/fluid/framework/details/computation_op_handle.cc +++ b/paddle/fluid/framework/details/computation_op_handle.cc @@ -28,8 +28,8 @@ ComputationOpHandle::ComputationOpHandle(const OpDesc &op_desc, Scope *scope, void ComputationOpHandle::RunImpl() { auto *cur_ctx = dev_ctxes_[place_]; for (auto *in : inputs_) { - bool need_wait = - in->generated_op_ && in->generated_op_->dev_ctxes_[place_] != cur_ctx; + bool need_wait = in->generated_op_ && + in->generated_op_->DeviceContext(place_) != cur_ctx; if (need_wait) { in->generated_op_->Wait(cur_ctx); } diff --git a/paddle/fluid/framework/details/computation_op_handle.h b/paddle/fluid/framework/details/computation_op_handle.h index d6d2d731ca..c363b973d9 100644 --- a/paddle/fluid/framework/details/computation_op_handle.h +++ b/paddle/fluid/framework/details/computation_op_handle.h @@ -14,6 +14,9 @@ #pragma once +#include +#include + #include "paddle/fluid/framework/details/op_handle_base.h" #include "paddle/fluid/framework/op_registry.h" #include "paddle/fluid/framework/operator.h" @@ -24,10 +27,7 @@ namespace paddle { namespace framework { namespace details { struct ComputationOpHandle : public OpHandleBase { - std::unique_ptr op_; - Scope *scope_; - platform::Place place_; - + public: ComputationOpHandle(const OpDesc &op_desc, Scope *scope, platform::Place place); @@ -35,6 +35,11 @@ struct ComputationOpHandle : public OpHandleBase { protected: void RunImpl() override; + + private: + std::unique_ptr op_; + Scope *scope_; + platform::Place place_; }; } // namespace details } // namespace framework diff --git a/paddle/fluid/framework/details/fetch_op_handle.h b/paddle/fluid/framework/details/fetch_op_handle.h index 904b2d669f..b49f3df338 100644 --- a/paddle/fluid/framework/details/fetch_op_handle.h +++ b/paddle/fluid/framework/details/fetch_op_handle.h @@ -14,6 +14,9 @@ #pragma once +#include +#include + #include "paddle/fluid/framework/details/op_handle_base.h" #include "paddle/fluid/framework/feed_fetch_type.h" #include "paddle/fluid/framework/scope.h" @@ -24,11 +27,7 @@ namespace framework { namespace details { struct FetchOpHandle : public OpHandleBase { - FeedFetchList *data_; - size_t offset_; - std::vector *local_scopes_; - std::vector tensors_; - + public: FetchOpHandle(FeedFetchList *data, size_t offset, std::vector *local_scopes); @@ -42,6 +41,12 @@ struct FetchOpHandle : public OpHandleBase { protected: void RunImpl() override; + + private: + FeedFetchList *data_; + size_t offset_; + std::vector *local_scopes_; + std::vector tensors_; }; } // namespace details diff --git a/paddle/fluid/framework/details/gather_op_handle.h b/paddle/fluid/framework/details/gather_op_handle.h index 6c0231f642..d11ef8556a 100644 --- a/paddle/fluid/framework/details/gather_op_handle.h +++ b/paddle/fluid/framework/details/gather_op_handle.h @@ -29,9 +29,7 @@ namespace framework { namespace details { struct GatherOpHandle : public OpHandleBase { - const std::vector &local_scopes_; - const std::vector &places_; - + public: GatherOpHandle(const std::vector &local_scopes, const std::vector &places); @@ -41,6 +39,10 @@ struct GatherOpHandle : public OpHandleBase { protected: void RunImpl() override; + + private: + const std::vector &local_scopes_; + const std::vector &places_; }; } // namespace details diff --git a/paddle/fluid/framework/details/gather_op_handle_test.cc b/paddle/fluid/framework/details/gather_op_handle_test.cc index 2da8c89d2d..9481579f6c 100644 --- a/paddle/fluid/framework/details/gather_op_handle_test.cc +++ b/paddle/fluid/framework/details/gather_op_handle_test.cc @@ -78,7 +78,7 @@ struct TestGatherOpHandle { op_handle_.reset(new GatherOpHandle(local_scopes_, gpu_list_)); // add input for (size_t j = 0; j < gpu_list_.size(); ++j) { - op_handle_->dev_ctxes_[gpu_list_[j]] = ctxs_[j].get(); + op_handle_->SetDeviceContext(gpu_list_[j], ctxs_[j].get()); auto* in_var_handle = new VarHandle(1, j, "input", gpu_list_[j]); vars_.emplace_back(in_var_handle); op_handle_->AddInput(in_var_handle); diff --git a/paddle/fluid/framework/details/multi_devices_graph_builder.cc b/paddle/fluid/framework/details/multi_devices_graph_builder.cc index d2b6a35a5d..002952436e 100644 --- a/paddle/fluid/framework/details/multi_devices_graph_builder.cc +++ b/paddle/fluid/framework/details/multi_devices_graph_builder.cc @@ -60,7 +60,8 @@ void MultiDevSSAGraphBuilder::CreateOpHandleIOs(SSAGraph *result, const platform::Place &p, const size_t &i) const { auto *op_handle = result->ops_.back().get(); - op_handle->dev_ctxes_[p] = platform::DeviceContextPool::Instance().Get(p); + op_handle->SetDeviceContext(p, + platform::DeviceContextPool::Instance().Get(p)); auto var_names = op.InputArgumentNames(); diff --git a/paddle/fluid/framework/details/nccl_all_reduce_op_handle.h b/paddle/fluid/framework/details/nccl_all_reduce_op_handle.h index ad14a3c5cb..a0c321843e 100644 --- a/paddle/fluid/framework/details/nccl_all_reduce_op_handle.h +++ b/paddle/fluid/framework/details/nccl_all_reduce_op_handle.h @@ -27,10 +27,6 @@ namespace framework { namespace details { struct NCCLAllReduceOpHandle : public OpHandleBase { - const std::vector &local_scopes_; - const std::vector &places_; - const platform::NCCLContextMap &nccl_ctxs_; - NCCLAllReduceOpHandle(const std::vector &local_scopes, const std::vector &places, const platform::NCCLContextMap &ctxs); @@ -43,6 +39,11 @@ struct NCCLAllReduceOpHandle : public OpHandleBase { protected: void RunImpl() override; + + private: + const std::vector &local_scopes_; + const std::vector &places_; + const platform::NCCLContextMap &nccl_ctxs_; }; } // namespace details diff --git a/paddle/fluid/framework/details/op_handle_base.h b/paddle/fluid/framework/details/op_handle_base.h index a9a6c8d39c..00f213f3ed 100644 --- a/paddle/fluid/framework/details/op_handle_base.h +++ b/paddle/fluid/framework/details/op_handle_base.h @@ -27,28 +27,15 @@ namespace details { constexpr char kLocalExecScopeName[] = "@LCOAL_SCOPE@"; class OpHandleBase { - private: - DISABLE_COPY_AND_ASSIGN(OpHandleBase); - public: - std::vector inputs_; - std::vector outputs_; - std::unordered_map - dev_ctxes_; - -#ifdef PADDLE_WITH_CUDA - std::unordered_map events_; -#endif - OpHandleBase() {} + virtual ~OpHandleBase(); + std::string DebugString() const; virtual std::string Name() const = 0; - virtual ~OpHandleBase(); - void Run(bool use_event); virtual void Wait(platform::DeviceContext *waited_dev); @@ -61,6 +48,18 @@ class OpHandleBase { // will likely block other computations. virtual bool IsMultiDeviceTransfer() { return false; } + const platform::DeviceContext *DeviceContext(platform::Place place) { + return dev_ctxes_[place]; + } + + void SetDeviceContext(platform::Place place, platform::DeviceContext *ctx_) { + dev_ctxes_[place] = ctx_; + } + + const std::vector &Inputs() const { return inputs_; } + + const std::vector &Outputs() const { return outputs_; } + protected: void RunAndRecordEvent(const std::function &callback); @@ -68,6 +67,18 @@ class OpHandleBase { const std::function &callback); virtual void RunImpl() = 0; + + std::vector inputs_; + std::vector outputs_; + std::unordered_map + dev_ctxes_; + +#ifdef PADDLE_WITH_CUDA + std::unordered_map events_; +#endif + + DISABLE_COPY_AND_ASSIGN(OpHandleBase); }; } // namespace details diff --git a/paddle/fluid/framework/details/scale_loss_grad_op_handle.h b/paddle/fluid/framework/details/scale_loss_grad_op_handle.h index ab7353a4fc..d93d599d46 100644 --- a/paddle/fluid/framework/details/scale_loss_grad_op_handle.h +++ b/paddle/fluid/framework/details/scale_loss_grad_op_handle.h @@ -14,6 +14,8 @@ #pragma once +#include + #include "paddle/fluid/framework/details/op_handle_base.h" #include "paddle/fluid/framework/lod_tensor.h" #include "paddle/fluid/framework/scope.h" @@ -23,10 +25,6 @@ namespace framework { namespace details { struct ScaleLossGradOpHandle : public OpHandleBase { - float coeff_; - Scope *scope_; - platform::Place place_; - ScaleLossGradOpHandle(size_t num_dev, Scope *scope, platform::Place place, platform::DeviceContext *context); @@ -36,6 +34,11 @@ struct ScaleLossGradOpHandle : public OpHandleBase { protected: void RunImpl() override; + + private: + float coeff_; + Scope *scope_; + platform::Place place_; }; } // namespace details diff --git a/paddle/fluid/framework/details/send_op_handle.h b/paddle/fluid/framework/details/send_op_handle.h index 173f9d7261..2f78811fad 100644 --- a/paddle/fluid/framework/details/send_op_handle.h +++ b/paddle/fluid/framework/details/send_op_handle.h @@ -28,10 +28,6 @@ namespace framework { namespace details { struct SendOpHandle : public OpHandleBase { - std::unique_ptr op_; - const Scope* local_scope_; - const platform::Place& place_; - SendOpHandle(const framework::OpDesc& op_desc, const Scope* local_scope, const platform::Place& place); @@ -43,6 +39,11 @@ struct SendOpHandle : public OpHandleBase { protected: void RunImpl() override; + + private: + std::unique_ptr op_; + const Scope* local_scope_; + const platform::Place& place_; }; } // namespace details diff --git a/paddle/fluid/framework/details/ssa_graph_builder.cc b/paddle/fluid/framework/details/ssa_graph_builder.cc index 25e8c77bb4..6a56752755 100644 --- a/paddle/fluid/framework/details/ssa_graph_builder.cc +++ b/paddle/fluid/framework/details/ssa_graph_builder.cc @@ -117,12 +117,12 @@ void SSAGraphBuilder::PrintGraphviz(const SSAGraph &graph, std::ostream &sout) { std::string op_name = "op_" + std::to_string(op_id++); sout << op_name << " [label=\"" << op->Name() << "\", shape=rect]" << std::endl; - for (auto in : op->inputs_) { + for (auto in : op->Inputs()) { std::string var_name = "var_" + std::to_string(vars[in]); sout << var_name << " -> " << op_name << std::endl; } - for (auto out : op->outputs_) { + for (auto out : op->Outputs()) { std::string var_name = "var_" + std::to_string(vars[out]); sout << op_name << " -> " << var_name << std::endl; } @@ -133,7 +133,7 @@ void SSAGraphBuilder::PrintGraphviz(const SSAGraph &graph, std::ostream &sout) { void SSAGraphBuilder::AddOutputToLeafOps(SSAGraph *graph) { for (auto &op : graph->ops_) { - if (!op->outputs_.empty()) { + if (!op->Outputs().empty()) { continue; } auto *dummy_leaf = new DummyVarHandle(); diff --git a/paddle/fluid/framework/details/threaded_ssa_graph_executor.cc b/paddle/fluid/framework/details/threaded_ssa_graph_executor.cc index 3d2bd633af..14e75e7b7b 100644 --- a/paddle/fluid/framework/details/threaded_ssa_graph_executor.cc +++ b/paddle/fluid/framework/details/threaded_ssa_graph_executor.cc @@ -53,7 +53,7 @@ FeedFetchList ThreadedSSAGraphExecutor::Run( }; auto InsertPendingOp = [&pending_ops](OpHandleBase &op_instance) { - pending_ops.insert({&op_instance, op_instance.inputs_.size()}); + pending_ops.insert({&op_instance, op_instance.Inputs().size()}); }; // Transform SSAGraph to pending_ops & pending_vars @@ -69,7 +69,7 @@ FeedFetchList ThreadedSSAGraphExecutor::Run( } for (auto &op : graph_->ops_) { - if (op->inputs_.empty()) { // Special case, Op has no input. + if (op->Inputs().empty()) { // Special case, Op has no input. ready_ops.insert(op.get()); } else { InsertPendingOp(*op); @@ -99,7 +99,7 @@ FeedFetchList ThreadedSSAGraphExecutor::Run( fetch_ops.emplace_back(op); for (auto &p : places_) { - op->dev_ctxes_[p] = fetch_ctxs_.Get(p); + op->SetDeviceContext(p, fetch_ctxs_.Get(p)); } for (auto *var : vars) { @@ -180,7 +180,7 @@ void ThreadedSSAGraphExecutor::RunOp( op->Run(use_event_); VLOG(10) << op << " " << op->Name() << " Done "; running_ops_--; - ready_var_q->Extend(op->outputs_); + ready_var_q->Extend(op->Outputs()); VLOG(10) << op << " " << op->Name() << "Signal posted"; } catch (platform::EnforceNotMet ex) { exception_.reset(new platform::EnforceNotMet(ex)); diff --git a/paddle/fluid/framework/op_desc.h b/paddle/fluid/framework/op_desc.h index 614dd8cd00..cd6777e60a 100644 --- a/paddle/fluid/framework/op_desc.h +++ b/paddle/fluid/framework/op_desc.h @@ -119,7 +119,7 @@ class OpDesc { void InferVarType(BlockDesc *block) const; - void MarkAsTarget() { desc_.set_is_target(true); } + void SetIsTarget(bool is_target) { desc_.set_is_target(is_target); } void Flush(); diff --git a/paddle/fluid/operators/activation_op.cc b/paddle/fluid/operators/activation_op.cc index 9db718a550..56451f8f14 100644 --- a/paddle/fluid/operators/activation_op.cc +++ b/paddle/fluid/operators/activation_op.cc @@ -559,125 +559,125 @@ $$out = \frac{x}{1 + e^{- \beta x}}$$ namespace ops = paddle::operators; REGISTER_OPERATOR(sigmoid, ops::ActivationOp, ops::SigmoidOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(sigmoid_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(sigmoid_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(logsigmoid, ops::ActivationOp, ops::LogSigmoidOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(logsigmoid_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(logsigmoid_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(exp, ops::ActivationOp, ops::ExpOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(exp_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(exp_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(relu, ops::ActivationWithMKLDNNOp, ops::ReluOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(relu_grad, ops::ActivationWithMKLDNNOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(relu_grad, ops::ActivationWithMKLDNNOpGrad); REGISTER_OPERATOR(tanh, ops::ActivationWithMKLDNNOp, ops::TanhOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(tanh_grad, ops::ActivationWithMKLDNNOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(tanh_grad, ops::ActivationWithMKLDNNOpGrad); REGISTER_OPERATOR(tanh_shrink, ops::ActivationOp, ops::TanhShrinkOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(tanh_shrink_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(tanh_shrink_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(softshrink, ops::ActivationOp, ops::SoftShrinkOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(softshrink_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(softshrink_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(sqrt, ops::ActivationWithMKLDNNOp, ops::SqrtOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(sqrt_grad, ops::ActivationWithMKLDNNOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(sqrt_grad, ops::ActivationWithMKLDNNOpGrad); REGISTER_OPERATOR(abs, ops::ActivationWithMKLDNNOp, ops::AbsOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(abs_grad, ops::ActivationWithMKLDNNOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(abs_grad, ops::ActivationWithMKLDNNOpGrad); REGISTER_OPERATOR(ceil, ops::ActivationOp, ops::CeilOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(ceil_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(ceil_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(floor, ops::ActivationOp, ops::FloorOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(floor_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(floor_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(cos, ops::ActivationOp, ops::CosOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(cos_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(cos_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(sin, ops::ActivationOp, ops::SinOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(sin_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(sin_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(round, ops::ActivationOp, ops::RoundOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(round_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(round_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(reciprocal, ops::ActivationOp, ops::ReciprocalOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(reciprocal_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(reciprocal_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(log, ops::ActivationOp, ops::LogOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(log_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(log_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(square, ops::ActivationOp, ops::SquareOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(square_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(square_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(softplus, ops::ActivationOp, ops::SoftplusOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(softplus_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(softplus_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(softsign, ops::ActivationOp, ops::SoftsignOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(softsign_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(softsign_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(brelu, ops::ActivationOp, ops::BReluOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(brelu_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(brelu_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(leaky_relu, ops::ActivationOp, ops::LeakyReluOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(leaky_relu_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(leaky_relu_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(soft_relu, ops::ActivationOp, ops::SoftReluOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(soft_relu_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(soft_relu_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(elu, ops::ActivationOp, ops::ELUOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(elu_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(elu_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(relu6, ops::ActivationOp, ops::Relu6OpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(relu6_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(relu6_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(pow, ops::ActivationOp, ops::PowOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(pow_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(pow_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(stanh, ops::ActivationOp, ops::STanhOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(stanh_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(stanh_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(hard_shrink, ops::ActivationOp, ops::HardShrinkOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(hard_shrink_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(hard_shrink_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(thresholded_relu, ops::ActivationOp, ops::ThresholdedReluOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(thresholded_relu_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(thresholded_relu_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(hard_sigmoid, ops::ActivationOp, ops::HardSigmoidOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(hard_sigmoid_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(hard_sigmoid_grad, ops::ActivationOpGrad); REGISTER_OPERATOR(swish, ops::ActivationOp, ops::SwishOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(swish_grad, ops::ActivationOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(swish_grad, ops::ActivationOpGrad); #define REGISTER_ACTIVATION_CPU_KERNEL(act_type, functor, grad_functor) \ REGISTER_OP_CPU_KERNEL( \ diff --git a/paddle/fluid/operators/bilinear_tensor_product_op.cc b/paddle/fluid/operators/bilinear_tensor_product_op.cc index 44e2af8e2e..e910ad92d1 100644 --- a/paddle/fluid/operators/bilinear_tensor_product_op.cc +++ b/paddle/fluid/operators/bilinear_tensor_product_op.cc @@ -155,9 +155,9 @@ class BilinearTensorProductOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(bilinear_tensor_product, ops::BilinearTensorProductOp, ops::BilinearTensorProductOpMaker, - paddle::framework::DefaultGradOpDescMaker) + paddle::framework::DefaultGradOpDescMaker); REGISTER_OPERATOR(bilinear_tensor_product_grad, - ops::BilinearTensorProductOpGrad) + ops::BilinearTensorProductOpGrad); REGISTER_OP_CPU_KERNEL( bilinear_tensor_product, ops::BilinearTensorProductKernel, diff --git a/paddle/fluid/operators/clip_op.cc b/paddle/fluid/operators/clip_op.cc index 3c2d8e8707..c71139fc7c 100644 --- a/paddle/fluid/operators/clip_op.cc +++ b/paddle/fluid/operators/clip_op.cc @@ -82,8 +82,8 @@ class ClipOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(clip, ops::ClipOp, ops::ClipOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(clip_grad, ops::ClipOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(clip_grad, ops::ClipOpGrad); REGISTER_OP_CPU_KERNEL( clip, ops::ClipKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/concat_op.cc b/paddle/fluid/operators/concat_op.cc index 5fbbe4d028..3bb3bd4eb1 100644 --- a/paddle/fluid/operators/concat_op.cc +++ b/paddle/fluid/operators/concat_op.cc @@ -105,10 +105,10 @@ class ConcatOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(concat, ops::ConcatOp, ops::ConcatOpMaker, paddle::framework::DefaultGradOpDescMaker< - false> /* set false to disable empty grad */) -REGISTER_OPERATOR(concat_grad, ops::ConcatOpGrad) + false> /* set false to disable empty grad */); +REGISTER_OPERATOR(concat_grad, ops::ConcatOpGrad); REGISTER_OP_CPU_KERNEL( - concat, ops::ConcatKernel) + concat, ops::ConcatKernel); REGISTER_OP_CPU_KERNEL( concat_grad, - ops::ConcatGradKernel) + ops::ConcatGradKernel); diff --git a/paddle/fluid/operators/conv_op.cc b/paddle/fluid/operators/conv_op.cc index 83e56f80ca..92748993c3 100644 --- a/paddle/fluid/operators/conv_op.cc +++ b/paddle/fluid/operators/conv_op.cc @@ -336,16 +336,16 @@ framework::OpKernelType ConvOpGrad::GetExpectedKernelType( namespace ops = paddle::operators; REGISTER_OPERATOR(conv2d, ops::ConvOp, ops::Conv2DOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(conv2d_grad, ops::ConvOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(conv2d_grad, ops::ConvOpGrad); // depthwise convolution op REGISTER_OPERATOR(depthwise_conv2d, ops::ConvOp, ops::Conv2DOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(depthwise_conv2d_grad, ops::ConvOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(depthwise_conv2d_grad, ops::ConvOpGrad); REGISTER_OPERATOR(conv3d, ops::ConvOp, ops::Conv3DOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(conv3d_grad, ops::ConvOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(conv3d_grad, ops::ConvOpGrad); // depthwise conv kernel // TODO(xingzhaolong): neon kernel for mobile diff --git a/paddle/fluid/operators/conv_shift_op.cc b/paddle/fluid/operators/conv_shift_op.cc index 46a675e936..82fdd30820 100644 --- a/paddle/fluid/operators/conv_shift_op.cc +++ b/paddle/fluid/operators/conv_shift_op.cc @@ -194,8 +194,8 @@ class ConvShiftGradKernel namespace ops = paddle::operators; REGISTER_OPERATOR(conv_shift, ops::ConvShiftOp, ops::ConvShiftOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(conv_shift_grad, ops::ConvShiftGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(conv_shift_grad, ops::ConvShiftGradOp); REGISTER_OP_CPU_KERNEL(conv_shift, ops::ConvShiftKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/conv_transpose_op.cc b/paddle/fluid/operators/conv_transpose_op.cc index c148237f85..d699dcafa4 100644 --- a/paddle/fluid/operators/conv_transpose_op.cc +++ b/paddle/fluid/operators/conv_transpose_op.cc @@ -300,8 +300,8 @@ namespace ops = paddle::operators; REGISTER_OPERATOR(conv2d_transpose, ops::ConvTransposeOp, ops::Conv2DTransposeOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(conv2d_transpose_grad, ops::ConvTransposeOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(conv2d_transpose_grad, ops::ConvTransposeOpGrad); REGISTER_OP_CPU_KERNEL( conv2d_transpose, @@ -315,8 +315,8 @@ REGISTER_OP_CPU_KERNEL( REGISTER_OPERATOR(conv3d_transpose, ops::ConvTransposeOp, ops::Conv3DTransposeOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(conv3d_transpose_grad, ops::ConvTransposeOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(conv3d_transpose_grad, ops::ConvTransposeOpGrad); REGISTER_OP_CPU_KERNEL( conv3d_transpose, diff --git a/paddle/fluid/operators/cos_sim_op.cc b/paddle/fluid/operators/cos_sim_op.cc index 8cde2cb077..04ca878e68 100644 --- a/paddle/fluid/operators/cos_sim_op.cc +++ b/paddle/fluid/operators/cos_sim_op.cc @@ -154,8 +154,8 @@ class CosSimOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(cos_sim, ops::CosSimOp, ops::CosSimOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(cos_sim_grad, ops::CosSimOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(cos_sim_grad, ops::CosSimOpGrad); REGISTER_OP_CPU_KERNEL( cos_sim, ops::CosSimKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/cross_entropy_op.cc b/paddle/fluid/operators/cross_entropy_op.cc index 0ad87e511e..0e0622e290 100644 --- a/paddle/fluid/operators/cross_entropy_op.cc +++ b/paddle/fluid/operators/cross_entropy_op.cc @@ -165,8 +165,8 @@ or not. But the output only shares the LoD information with input X. namespace ops = paddle::operators; REGISTER_OPERATOR(cross_entropy, ops::CrossEntropyOp, ops::CrossEntropyOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(cross_entropy_grad, ops::CrossEntropyGradientOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(cross_entropy_grad, ops::CrossEntropyGradientOp); REGISTER_OP_CPU_KERNEL(cross_entropy, ops::CrossEntropyOpKernel, ops::CrossEntropyOpKernel); REGISTER_OP_CPU_KERNEL(cross_entropy_grad, diff --git a/paddle/fluid/operators/cumsum_op.cc b/paddle/fluid/operators/cumsum_op.cc index 0da6f18852..f7c516a0ba 100644 --- a/paddle/fluid/operators/cumsum_op.cc +++ b/paddle/fluid/operators/cumsum_op.cc @@ -79,4 +79,4 @@ using CPU = paddle::platform::CPUDeviceContext; REGISTER_OPERATOR(cumsum, ops::CumOp, ops::CumsumOpMaker, ops::CumsumGradMaker); REGISTER_OP_CPU_KERNEL(cumsum, ops::CumKernel>, ops::CumKernel>, - ops::CumKernel>) + ops::CumKernel>); diff --git a/paddle/fluid/operators/cumsum_op.cu b/paddle/fluid/operators/cumsum_op.cu index 70e2a1de5e..eb5fd99ccb 100644 --- a/paddle/fluid/operators/cumsum_op.cu +++ b/paddle/fluid/operators/cumsum_op.cu @@ -19,4 +19,4 @@ using CUDA = paddle::platform::CUDADeviceContext; REGISTER_OP_CUDA_KERNEL(cumsum, ops::CumKernel>, ops::CumKernel>, - ops::CumKernel>) + ops::CumKernel>); diff --git a/paddle/fluid/operators/dropout_op.cc b/paddle/fluid/operators/dropout_op.cc index 3b9882ab94..4ed1b54884 100644 --- a/paddle/fluid/operators/dropout_op.cc +++ b/paddle/fluid/operators/dropout_op.cc @@ -102,8 +102,8 @@ class DropoutOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(dropout, ops::DropoutOp, ops::DropoutOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(dropout_grad, ops::DropoutOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(dropout_grad, ops::DropoutOpGrad); REGISTER_OP_CPU_KERNEL( dropout, ops::CPUDropoutKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/elementwise_div_op.cc b/paddle/fluid/operators/elementwise_div_op.cc index f3dabb9133..c7ddafcad1 100644 --- a/paddle/fluid/operators/elementwise_div_op.cc +++ b/paddle/fluid/operators/elementwise_div_op.cc @@ -32,8 +32,8 @@ class ElementwiseDivOpMaker : public ElementwiseOpMaker { namespace ops = paddle::operators; REGISTER_OPERATOR(elementwise_div, ops::ElementwiseOp, ops::ElementwiseDivOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(elementwise_div_grad, ops::ElementwiseOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(elementwise_div_grad, ops::ElementwiseOpGrad); REGISTER_OP_CPU_KERNEL( elementwise_div, ops::ElementwiseDivKernel, diff --git a/paddle/fluid/operators/elementwise_max_op.cc b/paddle/fluid/operators/elementwise_max_op.cc index 385159e8ec..a4fe386bb1 100644 --- a/paddle/fluid/operators/elementwise_max_op.cc +++ b/paddle/fluid/operators/elementwise_max_op.cc @@ -31,8 +31,8 @@ class ElementwiseMaxOpMaker : public ElementwiseOpMaker { namespace ops = paddle::operators; REGISTER_OPERATOR(elementwise_max, ops::ElementwiseOp, ops::ElementwiseMaxOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(elementwise_max_grad, ops::ElementwiseOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(elementwise_max_grad, ops::ElementwiseOpGrad); REGISTER_OP_CPU_KERNEL( elementwise_max, ops::ElementwiseMaxKernel, diff --git a/paddle/fluid/operators/elementwise_min_op.cc b/paddle/fluid/operators/elementwise_min_op.cc index 0b7ea4b1bf..68cd6ddb4a 100644 --- a/paddle/fluid/operators/elementwise_min_op.cc +++ b/paddle/fluid/operators/elementwise_min_op.cc @@ -31,8 +31,8 @@ class ElementwiseMinOpMaker : public ElementwiseOpMaker { namespace ops = paddle::operators; REGISTER_OPERATOR(elementwise_min, ops::ElementwiseOp, ops::ElementwiseMinOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(elementwise_min_grad, ops::ElementwiseOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(elementwise_min_grad, ops::ElementwiseOpGrad); REGISTER_OP_CPU_KERNEL( elementwise_min, ops::ElementwiseMinKernel, diff --git a/paddle/fluid/operators/elementwise_mul_op.cc b/paddle/fluid/operators/elementwise_mul_op.cc index 0e092924d7..2dec27136a 100644 --- a/paddle/fluid/operators/elementwise_mul_op.cc +++ b/paddle/fluid/operators/elementwise_mul_op.cc @@ -33,8 +33,8 @@ class ElementwiseMulOpMaker : public ElementwiseOpMaker { namespace ops = paddle::operators; REGISTER_OPERATOR(elementwise_mul, ops::ElementwiseOp, ops::ElementwiseMulOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(elementwise_mul_grad, ops::ElementwiseOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(elementwise_mul_grad, ops::ElementwiseOpGrad); REGISTER_OP_CPU_KERNEL( elementwise_mul, ops::ElementwiseMulKernel, diff --git a/paddle/fluid/operators/elementwise_sub_op.cc b/paddle/fluid/operators/elementwise_sub_op.cc index 675ff8860b..9d0598fc39 100644 --- a/paddle/fluid/operators/elementwise_sub_op.cc +++ b/paddle/fluid/operators/elementwise_sub_op.cc @@ -31,8 +31,8 @@ class ElementwiseSubOpMaker : public ElementwiseOpMaker { namespace ops = paddle::operators; REGISTER_OPERATOR(elementwise_sub, ops::ElementwiseOp, ops::ElementwiseSubOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(elementwise_sub_grad, ops::ElementwiseOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(elementwise_sub_grad, ops::ElementwiseOpGrad); REGISTER_OP_CPU_KERNEL( elementwise_sub, ops::ElementwiseSubKernel, diff --git a/paddle/fluid/operators/expand_op.cc b/paddle/fluid/operators/expand_op.cc index d69b769651..9c71ee6d3b 100644 --- a/paddle/fluid/operators/expand_op.cc +++ b/paddle/fluid/operators/expand_op.cc @@ -13,6 +13,7 @@ See the License for the specific language governing permissions and limitations under the License. */ #include "paddle/fluid/operators/expand_op.h" +#include #include @@ -131,8 +132,8 @@ class ExpandGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(expand, ops::ExpandOp, ops::ExpandOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(expand_grad, ops::ExpandGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(expand_grad, ops::ExpandGradOp); REGISTER_OP_CPU_KERNEL( expand, ops::ExpandKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/expand_op.h b/paddle/fluid/operators/expand_op.h index 2c2d5c7c42..75dbf1d8bf 100644 --- a/paddle/fluid/operators/expand_op.h +++ b/paddle/fluid/operators/expand_op.h @@ -14,13 +14,14 @@ limitations under the License. */ #pragma once +#include + #include #include #include #include #include #include -#include #include "paddle/fluid/framework/eigen.h" #include "paddle/fluid/framework/op_registry.h" #include "paddle/fluid/framework/operator.h" diff --git a/paddle/fluid/operators/fc_op.cc b/paddle/fluid/operators/fc_op.cc index 5070a4b78d..45e4d5b2b8 100644 --- a/paddle/fluid/operators/fc_op.cc +++ b/paddle/fluid/operators/fc_op.cc @@ -99,5 +99,5 @@ FCOpMaker::FCOpMaker(OpProto* proto, OpAttrChecker* op_checker) } // namespace paddle REGISTER_OPERATOR(fc, paddle::operators::FCOp, paddle::operators::FCOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(fc_grad, paddle::operators::FCOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(fc_grad, paddle::operators::FCOpGrad); diff --git a/paddle/fluid/operators/gather_op.cc b/paddle/fluid/operators/gather_op.cc index 60075d9777..4c82f5c429 100644 --- a/paddle/fluid/operators/gather_op.cc +++ b/paddle/fluid/operators/gather_op.cc @@ -101,7 +101,7 @@ Out = [[3, 4], namespace ops = paddle::operators; REGISTER_OPERATOR(gather, ops::GatherOp, ops::GatherOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(gather_grad, ops::GatherGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(gather_grad, ops::GatherGradOp); REGISTER_OP_CPU_KERNEL(gather, ops::GatherOpKernel); REGISTER_OP_CPU_KERNEL(gather_grad, ops::GatherGradientOpKernel); diff --git a/paddle/fluid/operators/gather_op.cu b/paddle/fluid/operators/gather_op.cu index 3819549c71..7e014dd1cb 100644 --- a/paddle/fluid/operators/gather_op.cu +++ b/paddle/fluid/operators/gather_op.cu @@ -12,10 +12,10 @@ 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 "gather.cu.h" #include "paddle/fluid/framework/eigen.h" +#include "paddle/fluid/operators/gather.cu.h" #include "paddle/fluid/operators/gather_op.h" -#include "scatter.cu.h" +#include "paddle/fluid/operators/scatter.cu.h" namespace paddle { namespace operators { diff --git a/paddle/fluid/operators/gather_op.h b/paddle/fluid/operators/gather_op.h index 5a8b1ebbe3..2dd726bebb 100644 --- a/paddle/fluid/operators/gather_op.h +++ b/paddle/fluid/operators/gather_op.h @@ -13,10 +13,10 @@ See the License for the specific language governing permissions and limitations under the License. */ #pragma once -#include "gather.h" #include "paddle/fluid/framework/eigen.h" #include "paddle/fluid/framework/op_registry.h" -#include "scatter.h" +#include "paddle/fluid/operators/gather.h" +#include "paddle/fluid/operators/scatter.h" namespace paddle { namespace operators { diff --git a/paddle/fluid/operators/gather_test.cc b/paddle/fluid/operators/gather_test.cc index 7625bd45d9..9c0561b016 100644 --- a/paddle/fluid/operators/gather_test.cc +++ b/paddle/fluid/operators/gather_test.cc @@ -12,38 +12,37 @@ 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/operators/gather.h" -#include "paddle/fluid/framework/ddim.h" -#include "paddle/fluid/framework/tensor.h" -#include "paddle/fluid/platform/place.h" - #include #include #include -TEST(Gather, GatherData) { - using namespace paddle::framework; - using namespace paddle::platform; - using namespace paddle::operators; +#include "paddle/fluid/framework/ddim.h" +#include "paddle/fluid/framework/tensor.h" +#include "paddle/fluid/operators/gather.h" +#include "paddle/fluid/platform/place.h" - Tensor* src = new Tensor(); - Tensor* index = new Tensor(); - Tensor* output = new Tensor(); +TEST(Gather, GatherData) { + paddle::framework::Tensor* src = new paddle::framework::Tensor(); + paddle::framework::Tensor* index = new paddle::framework::Tensor(); + paddle::framework::Tensor* output = new paddle::framework::Tensor(); int* p_src = nullptr; int* p_index = nullptr; - p_src = src->mutable_data(make_ddim({3, 4}), CPUPlace()); - p_index = index->mutable_data(make_ddim({2}), CPUPlace()); + p_src = src->mutable_data(paddle::framework::make_ddim({3, 4}), + paddle::platform::CPUPlace()); + p_index = index->mutable_data(paddle::framework::make_ddim({2}), + paddle::platform::CPUPlace()); for (int i = 0; i < 12; ++i) p_src[i] = i; p_index[0] = 1; p_index[1] = 0; - int* p_output = output->mutable_data(make_ddim({2, 4}), CPUPlace()); + int* p_output = output->mutable_data( + paddle::framework::make_ddim({2, 4}), paddle::platform::CPUPlace()); auto* cpu_place = new paddle::platform::CPUPlace(); paddle::platform::CPUDeviceContext ctx(*cpu_place); - CPUGather(ctx, *src, *index, output); + paddle::operators::CPUGather(ctx, *src, *index, output); for (int i = 0; i < 4; ++i) EXPECT_EQ(p_output[i], i + 4); for (int i = 4; i < 8; ++i) EXPECT_EQ(p_output[i], i - 4); diff --git a/paddle/fluid/operators/get_places_op.cc b/paddle/fluid/operators/get_places_op.cc index 9002ce4717..0d7219ac5c 100644 --- a/paddle/fluid/operators/get_places_op.cc +++ b/paddle/fluid/operators/get_places_op.cc @@ -12,7 +12,7 @@ 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 +#include // NOLINT #include "paddle/fluid/framework/op_registry.h" #include "paddle/fluid/operators/detail/safe_ref.h" #include "paddle/fluid/platform/place.h" diff --git a/paddle/fluid/operators/gru_op.cc b/paddle/fluid/operators/gru_op.cc index b717c59091..0a524c914d 100644 --- a/paddle/fluid/operators/gru_op.cc +++ b/paddle/fluid/operators/gru_op.cc @@ -217,8 +217,8 @@ class GRUGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(gru, ops::GRUOp, ops::GRUOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(gru_grad, ops::GRUGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(gru_grad, ops::GRUGradOp); REGISTER_OP_CPU_KERNEL( gru, ops::GRUKernel, ops::GRUKernel); diff --git a/paddle/fluid/operators/hinge_loss_op.cc b/paddle/fluid/operators/hinge_loss_op.cc index d14935e771..086b5a97de 100644 --- a/paddle/fluid/operators/hinge_loss_op.cc +++ b/paddle/fluid/operators/hinge_loss_op.cc @@ -104,8 +104,8 @@ class HingeLossGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(hinge_loss, ops::HingeLossOp, ops::HingeLossOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(hinge_loss_grad, ops::HingeLossGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(hinge_loss_grad, ops::HingeLossGradOp); REGISTER_OP_CPU_KERNEL( hinge_loss, ops::HingeLossKernel); diff --git a/paddle/fluid/operators/huber_loss_op.cc b/paddle/fluid/operators/huber_loss_op.cc index 0789c89bd1..74d8e0e2b7 100644 --- a/paddle/fluid/operators/huber_loss_op.cc +++ b/paddle/fluid/operators/huber_loss_op.cc @@ -122,8 +122,8 @@ class HuberLossGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(huber_loss, ops::HuberLossOp, ops::HuberLossOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(huber_loss_grad, ops::HuberLossGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(huber_loss_grad, ops::HuberLossGradOp); REGISTER_OP_CPU_KERNEL( huber_loss, ops::HuberLossKernel); diff --git a/paddle/fluid/operators/im2sequence_op.cc b/paddle/fluid/operators/im2sequence_op.cc index 593cf60c11..8c120eec86 100644 --- a/paddle/fluid/operators/im2sequence_op.cc +++ b/paddle/fluid/operators/im2sequence_op.cc @@ -149,8 +149,8 @@ class Im2SequenceGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(im2sequence, ops::Im2SequenceOp, ops::Im2SequenceOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(im2sequence_grad, ops::Im2SequenceGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(im2sequence_grad, ops::Im2SequenceGradOp); REGISTER_OP_CPU_KERNEL( im2sequence, ops::Im2SequenceKernel); diff --git a/paddle/fluid/operators/increment_op.cc b/paddle/fluid/operators/increment_op.cc index ec2e641679..5d8710a9b3 100644 --- a/paddle/fluid/operators/increment_op.cc +++ b/paddle/fluid/operators/increment_op.cc @@ -89,4 +89,4 @@ REGISTER_OP_CPU_KERNEL( increment, ops::IncrementKernel, ops::IncrementKernel, ops::IncrementKernel, - ops::IncrementKernel) + ops::IncrementKernel); diff --git a/paddle/fluid/operators/increment_op.cu b/paddle/fluid/operators/increment_op.cu index 7fb6425fe9..228063bf3d 100644 --- a/paddle/fluid/operators/increment_op.cu +++ b/paddle/fluid/operators/increment_op.cu @@ -19,4 +19,4 @@ REGISTER_OP_CUDA_KERNEL( increment, ops::IncrementKernel, ops::IncrementKernel, ops::IncrementKernel, - ops::IncrementKernel) + ops::IncrementKernel); diff --git a/paddle/fluid/operators/iou_similarity_op.cc b/paddle/fluid/operators/iou_similarity_op.cc old mode 100755 new mode 100644 diff --git a/paddle/fluid/operators/iou_similarity_op.cu b/paddle/fluid/operators/iou_similarity_op.cu old mode 100755 new mode 100644 diff --git a/paddle/fluid/operators/l1_norm_op.cc b/paddle/fluid/operators/l1_norm_op.cc index ba7577c510..0c143b7c8a 100644 --- a/paddle/fluid/operators/l1_norm_op.cc +++ b/paddle/fluid/operators/l1_norm_op.cc @@ -68,8 +68,8 @@ $$Out = \sum{|X|}$$ namespace ops = paddle::operators; REGISTER_OPERATOR(l1_norm, ops::L1NormOp, ops::L1NormOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(l1_norm_grad, ops::L1NormGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(l1_norm_grad, ops::L1NormGradOp); REGISTER_OP_CPU_KERNEL( l1_norm, ops::L1NormKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/label_smooth_op.cc b/paddle/fluid/operators/label_smooth_op.cc index 663adc5700..a73c626032 100644 --- a/paddle/fluid/operators/label_smooth_op.cc +++ b/paddle/fluid/operators/label_smooth_op.cc @@ -118,8 +118,8 @@ class LabelSmoothGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(label_smooth, ops::LabelSmoothOp, ops::LabelSmoothOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(label_smooth_grad, ops::LabelSmoothGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(label_smooth_grad, ops::LabelSmoothGradOp); REGISTER_OP_CPU_KERNEL( label_smooth, ops::LabelSmoothKernel, diff --git a/paddle/fluid/operators/layer_norm_op.cc b/paddle/fluid/operators/layer_norm_op.cc index e033da857b..de1056aef7 100644 --- a/paddle/fluid/operators/layer_norm_op.cc +++ b/paddle/fluid/operators/layer_norm_op.cc @@ -163,8 +163,8 @@ class LayerNormGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(layer_norm, ops::LayerNormOp, ops::LayerNormOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(layer_norm_grad, ops::LayerNormGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(layer_norm_grad, ops::LayerNormGradOp); REGISTER_OP_CPU_KERNEL( layer_norm, ops::LayerNormKernel, ops::LayerNormKernel); diff --git a/paddle/fluid/operators/linear_chain_crf_op.cc b/paddle/fluid/operators/linear_chain_crf_op.cc index 24b845528d..2f29e377fd 100644 --- a/paddle/fluid/operators/linear_chain_crf_op.cc +++ b/paddle/fluid/operators/linear_chain_crf_op.cc @@ -258,8 +258,8 @@ class LinearChainCRFGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(linear_chain_crf, ops::LinearChainCRFOp, ops::LinearChainCRFOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(linear_chain_crf_grad, ops::LinearChainCRFGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(linear_chain_crf_grad, ops::LinearChainCRFGradOp); REGISTER_OP_CPU_KERNEL( linear_chain_crf, ops::LinearChainCRFOpKernel, diff --git a/paddle/fluid/operators/lod_reset_op.cc b/paddle/fluid/operators/lod_reset_op.cc index fd1e1ffd46..92ebfc274b 100644 --- a/paddle/fluid/operators/lod_reset_op.cc +++ b/paddle/fluid/operators/lod_reset_op.cc @@ -156,8 +156,8 @@ class LoDResetGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(lod_reset, ops::LoDResetOp, ops::LoDResetOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(lod_reset_grad, ops::LoDResetGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(lod_reset_grad, ops::LoDResetGradOp); REGISTER_OP_CPU_KERNEL( lod_reset, ops::LoDResetKernel, ops::LoDResetKernel, diff --git a/paddle/fluid/operators/log_loss_op.cc b/paddle/fluid/operators/log_loss_op.cc index b1a68d2887..a8258a1afd 100644 --- a/paddle/fluid/operators/log_loss_op.cc +++ b/paddle/fluid/operators/log_loss_op.cc @@ -107,8 +107,8 @@ class LogLossGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(log_loss, ops::LogLossOp, ops::LogLossOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(log_loss_grad, ops::LogLossGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(log_loss_grad, ops::LogLossGradOp); REGISTER_OP_CPU_KERNEL( log_loss, ops::LogLossKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/lrn_op.cc b/paddle/fluid/operators/lrn_op.cc index 6ff9a68ba4..f5c0e47fda 100644 --- a/paddle/fluid/operators/lrn_op.cc +++ b/paddle/fluid/operators/lrn_op.cc @@ -277,8 +277,8 @@ class LRNOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(lrn, ops::LRNOp, ops::LRNOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(lrn_grad, ops::LRNOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(lrn_grad, ops::LRNOpGrad); REGISTER_OP_CPU_KERNEL( lrn, ops::LRNKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/lstm_op.cc b/paddle/fluid/operators/lstm_op.cc index 75b9c65f18..084ee1cfe6 100644 --- a/paddle/fluid/operators/lstm_op.cc +++ b/paddle/fluid/operators/lstm_op.cc @@ -274,8 +274,8 @@ class LSTMGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(lstm, ops::LSTMOp, ops::LSTMOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(lstm_grad, ops::LSTMGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(lstm_grad, ops::LSTMGradOp); REGISTER_OP_CPU_KERNEL( lstm, ops::LSTMKernel, ops::LSTMKernel); diff --git a/paddle/fluid/operators/lstm_unit_op.cc b/paddle/fluid/operators/lstm_unit_op.cc index 16d2dabd1d..e1157ef6c6 100644 --- a/paddle/fluid/operators/lstm_unit_op.cc +++ b/paddle/fluid/operators/lstm_unit_op.cc @@ -98,8 +98,8 @@ class LstmUnitGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(lstm_unit, ops::LstmUnitOp, ops::LstmUnitOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(lstm_unit_grad, ops::LstmUnitGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(lstm_unit_grad, ops::LstmUnitGradOp); REGISTER_OP_CPU_KERNEL(lstm_unit, ops::LstmUnitKernel, ops::LstmUnitKernel); diff --git a/paddle/fluid/operators/lstmp_op.cc b/paddle/fluid/operators/lstmp_op.cc index a575ade472..f9261323f0 100644 --- a/paddle/fluid/operators/lstmp_op.cc +++ b/paddle/fluid/operators/lstmp_op.cc @@ -323,8 +323,8 @@ class LSTMPGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(lstmp, ops::LSTMPOp, ops::LSTMPOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(lstmp_grad, ops::LSTMPGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(lstmp_grad, ops::LSTMPGradOp); REGISTER_OP_CPU_KERNEL( lstmp, ops::LSTMPKernel, ops::LSTMPKernel); diff --git a/paddle/fluid/operators/margin_rank_loss_op.cc b/paddle/fluid/operators/margin_rank_loss_op.cc index b3f6431233..0b41a3e1ff 100644 --- a/paddle/fluid/operators/margin_rank_loss_op.cc +++ b/paddle/fluid/operators/margin_rank_loss_op.cc @@ -113,8 +113,8 @@ namespace ops = paddle::operators; REGISTER_OPERATOR(margin_rank_loss, ops::MarginRankLossOp, ops::MarginRankLossOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(margin_rank_loss_grad, ops::MarginRankLossGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(margin_rank_loss_grad, ops::MarginRankLossGradOp); REGISTER_OP_CPU_KERNEL( margin_rank_loss, ops::MarginRankLossKernel); diff --git a/paddle/fluid/operators/matmul_op.cc b/paddle/fluid/operators/matmul_op.cc index 6a3507fbfc..e5d33fbc36 100644 --- a/paddle/fluid/operators/matmul_op.cc +++ b/paddle/fluid/operators/matmul_op.cc @@ -238,8 +238,8 @@ class MatMulOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(matmul, ops::MatMulOp, ops::MatMulOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(matmul_grad, ops::MatMulOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(matmul_grad, ops::MatMulOpGrad); REGISTER_OP_CPU_KERNEL( matmul, ops::MatMulKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/maxout_op.cc b/paddle/fluid/operators/maxout_op.cc index 9144d1fab9..e2bcba5a5e 100644 --- a/paddle/fluid/operators/maxout_op.cc +++ b/paddle/fluid/operators/maxout_op.cc @@ -102,8 +102,8 @@ class MaxOutOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(maxout, ops::MaxOutOp, ops::MaxOutOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(maxout_grad, ops::MaxOutOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(maxout_grad, ops::MaxOutOpGrad); REGISTER_OP_CPU_KERNEL( maxout, ops::MaxOutKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/modified_huber_loss_op.cc b/paddle/fluid/operators/modified_huber_loss_op.cc index 042a977d2e..3a0fc74584 100644 --- a/paddle/fluid/operators/modified_huber_loss_op.cc +++ b/paddle/fluid/operators/modified_huber_loss_op.cc @@ -110,8 +110,8 @@ class ModifiedHuberLossGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(modified_huber_loss, ops::ModifiedHuberLossOp, ops::ModifiedHuberLossOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(modified_huber_loss_grad, ops::ModifiedHuberLossGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(modified_huber_loss_grad, ops::ModifiedHuberLossGradOp); REGISTER_OP_CPU_KERNEL( modified_huber_loss, diff --git a/paddle/fluid/operators/mul_op.cc b/paddle/fluid/operators/mul_op.cc index 9a99e3878a..bfb20fefba 100644 --- a/paddle/fluid/operators/mul_op.cc +++ b/paddle/fluid/operators/mul_op.cc @@ -161,8 +161,8 @@ class MulGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(mul, ops::MulOp, ops::MulOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(mul_grad, ops::MulGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(mul_grad, ops::MulGradOp); REGISTER_OP_CPU_KERNEL( mul, ops::MulKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/multiclass_nms_op.cc b/paddle/fluid/operators/multiclass_nms_op.cc index 0f80f752c9..a12b975326 100644 --- a/paddle/fluid/operators/multiclass_nms_op.cc +++ b/paddle/fluid/operators/multiclass_nms_op.cc @@ -173,8 +173,8 @@ class MultiClassNMSKernel : public framework::OpKernel { void MultiClassNMS(const framework::ExecutionContext& ctx, const Tensor& scores, const Tensor& bboxes, - std::map>& indices, - int& num_nmsed_out) const { + std::map>* indices, + int* num_nmsed_out) const { int64_t background_label = ctx.Attr("background_label"); int64_t nms_top_k = ctx.Attr("nms_top_k"); int64_t keep_top_k = ctx.Attr("keep_top_k"); @@ -189,15 +189,15 @@ class MultiClassNMSKernel : public framework::OpKernel { if (c == background_label) continue; Tensor score = scores.Slice(c, c + 1); NMSFast(bboxes, score, score_threshold, nms_threshold, nms_eta, nms_top_k, - &(indices[c])); - num_det += indices[c].size(); + &((*indices)[c])); + num_det += (*indices)[c].size(); } - num_nmsed_out = num_det; + *num_nmsed_out = num_det; const T* scores_data = scores.data(); if (keep_top_k > -1 && num_det > keep_top_k) { std::vector>> score_index_pairs; - for (const auto& it : indices) { + for (const auto& it : *indices) { int label = it.first; const T* sdata = scores_data + label * predict_dim; const std::vector& label_indices = it.second; @@ -220,13 +220,13 @@ class MultiClassNMSKernel : public framework::OpKernel { int idx = score_index_pairs[j].second.second; new_indices[label].push_back(idx); } - new_indices.swap(indices); - num_nmsed_out = keep_top_k; + new_indices.swap(*indices); + *num_nmsed_out = keep_top_k; } } void MultiClassOutput(const Tensor& scores, const Tensor& bboxes, - std::map>& selected_indices, + const std::map>& selected_indices, Tensor* outs) const { int predict_dim = scores.dims()[1]; auto* scores_data = scores.data(); @@ -273,7 +273,7 @@ class MultiClassNMSKernel : public framework::OpKernel { std::map> indices; int num_nmsed_out = 0; - MultiClassNMS(ctx, ins_score, ins_boxes, indices, num_nmsed_out); + MultiClassNMS(ctx, ins_score, ins_boxes, &indices, &num_nmsed_out); all_indices.push_back(indices); batch_starts.push_back(batch_starts.back() + num_nmsed_out); } diff --git a/paddle/fluid/operators/nccl_op.cu.cc b/paddle/fluid/operators/nccl_op.cu.cc index ad623e1fe0..8de974bc2b 100644 --- a/paddle/fluid/operators/nccl_op.cu.cc +++ b/paddle/fluid/operators/nccl_op.cu.cc @@ -135,8 +135,9 @@ class NCCLBcastKernel : public framework::OpKernel { auto* x = ctx.Input("X"); VLOG(3) << "gpu : " << gpu_id << " invoke Bcast. send " << x->numel(); PADDLE_ENFORCE(platform::dynload::ncclBcast( - (void*)x->data(), x->numel(), NCCLTypeWrapper::type, root, - comm->comms().at(idx), ctx.cuda_device_context().stream())); + reinterpret_cast(const_cast(x->data())), x->numel(), + NCCLTypeWrapper::type, root, comm->comms().at(idx), + ctx.cuda_device_context().stream())); VLOG(3) << "gpu : " << gpu_id << " finished Bcast."; } else { auto* out = ctx.Output("Out"); diff --git a/paddle/fluid/operators/nce_op.cc b/paddle/fluid/operators/nce_op.cc index b471a7e594..192bdf8ea5 100644 --- a/paddle/fluid/operators/nce_op.cc +++ b/paddle/fluid/operators/nce_op.cc @@ -182,8 +182,8 @@ class NCEOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(nce, ops::NCEOp, ops::NCEOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(nce_grad, ops::NCEOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(nce_grad, ops::NCEOpGrad); REGISTER_OP_CPU_KERNEL(nce, ops::NCEKernel, ops::NCEKernel); REGISTER_OP_CPU_KERNEL(nce_grad, diff --git a/paddle/fluid/operators/nce_op.h b/paddle/fluid/operators/nce_op.h index 9420763847..2c4c97f28b 100644 --- a/paddle/fluid/operators/nce_op.h +++ b/paddle/fluid/operators/nce_op.h @@ -16,6 +16,7 @@ limitations under the License. */ #include #include +#include #include "paddle/fluid/framework/eigen.h" #include "paddle/fluid/framework/op_registry.h" #include "unsupported/Eigen/CXX11/Tensor" @@ -108,7 +109,7 @@ class NCEKernel : public framework::OpKernel { auto weight_mat = EigenMatrix::From(*(context.Input("Weight"))); for (int64_t i = 0; i < sample_labels->numel(); ++i) { Eigen::Tensor result = - (input_mat.chip((int)(i / sample_labels->dims()[1]), 0) * + (input_mat.chip(static_cast(i / sample_labels->dims()[1]), 0) * weight_mat.chip(sample_labels_data[i], 0)) .sum(); sample_out_data[i] += result(0); @@ -190,7 +191,7 @@ class NCEGradKernel : public framework::OpKernel { auto x_matrix = EigenMatrix::From(*(context.Input("Input"))); for (int64_t i = 0; i < sample_labels->numel(); ++i) { d_w_matrix.chip(sample_labels_data[i], 0) += - x_matrix.chip((int)(i / sample_labels->dims()[1]), 0) * + x_matrix.chip(static_cast(i / sample_labels->dims()[1]), 0) * sample_grad_data[i]; } } @@ -202,7 +203,7 @@ class NCEGradKernel : public framework::OpKernel { auto d_x_matrix = EigenMatrix::From(*d_x); auto w_matrix = EigenMatrix::From(*(context.Input("Weight"))); for (int64_t i = 0; i < sample_labels->numel(); ++i) { - d_x_matrix.chip((int)(i / sample_labels->dims()[1]), 0) += + d_x_matrix.chip(static_cast(i / sample_labels->dims()[1]), 0) += w_matrix.chip(sample_labels_data[i], 0) * sample_grad_data[i]; } } diff --git a/paddle/fluid/operators/norm_op.cc b/paddle/fluid/operators/norm_op.cc index ff4d6ec69f..30a991224f 100644 --- a/paddle/fluid/operators/norm_op.cc +++ b/paddle/fluid/operators/norm_op.cc @@ -86,8 +86,8 @@ class NormOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(norm, ops::NormOp, ops::NormOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(norm_grad, ops::NormOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(norm_grad, ops::NormOpGrad); REGISTER_OP_CPU_KERNEL( norm, ops::NormKernel, ops::NormKernel); diff --git a/paddle/fluid/operators/pool_op.cc b/paddle/fluid/operators/pool_op.cc index 371100fd74..f2de075e0d 100644 --- a/paddle/fluid/operators/pool_op.cc +++ b/paddle/fluid/operators/pool_op.cc @@ -334,19 +334,19 @@ Example: namespace ops = paddle::operators; REGISTER_OPERATOR(pool2d, ops::PoolOp, ops::Pool2dOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(pool2d_grad, ops::PoolOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(pool2d_grad, ops::PoolOpGrad); REGISTER_OP_CPU_KERNEL( pool2d, ops::PoolKernel, ops::PoolKernel); REGISTER_OP_CPU_KERNEL( pool2d_grad, ops::PoolGradKernel, - ops::PoolGradKernel) + ops::PoolGradKernel); REGISTER_OPERATOR(pool3d, ops::PoolOp, ops::Pool3dOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(pool3d_grad, ops::PoolOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(pool3d_grad, ops::PoolOpGrad); REGISTER_OP_CPU_KERNEL( pool3d, ops::PoolKernel, diff --git a/paddle/fluid/operators/pool_with_index_op.cc b/paddle/fluid/operators/pool_with_index_op.cc index a633beab3b..848cd61b23 100644 --- a/paddle/fluid/operators/pool_with_index_op.cc +++ b/paddle/fluid/operators/pool_with_index_op.cc @@ -260,8 +260,8 @@ namespace ops = paddle::operators; REGISTER_OPERATOR(max_pool2d_with_index, ops::MaxPoolWithIndexOp, ops::MaxPool2dWithIndexOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(max_pool2d_with_index_grad, ops::MaxPoolWithIndexOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(max_pool2d_with_index_grad, ops::MaxPoolWithIndexOpGrad); REGISTER_OP_CPU_KERNEL( max_pool2d_with_index, @@ -273,12 +273,12 @@ REGISTER_OP_CPU_KERNEL( ops::MaxPoolWithIndexGradKernel, ops::MaxPoolWithIndexGradKernel) + int>); REGISTER_OPERATOR(max_pool3d_with_index, ops::MaxPoolWithIndexOp, ops::MaxPool3dWithIndexOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(max_pool3d_with_index_grad, ops::MaxPoolWithIndexOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(max_pool3d_with_index_grad, ops::MaxPoolWithIndexOpGrad); REGISTER_OP_CPU_KERNEL( max_pool3d_with_index, @@ -290,4 +290,4 @@ REGISTER_OP_CPU_KERNEL( ops::MaxPoolWithIndexGradKernel, ops::MaxPoolWithIndexGradKernel) + int>); diff --git a/paddle/fluid/operators/pool_with_index_op.cu.cc b/paddle/fluid/operators/pool_with_index_op.cu.cc index 5fc418b6fd..5497dcbd9c 100644 --- a/paddle/fluid/operators/pool_with_index_op.cu.cc +++ b/paddle/fluid/operators/pool_with_index_op.cu.cc @@ -27,7 +27,7 @@ REGISTER_OP_CUDA_KERNEL( ops::MaxPoolWithIndexGradKernel, ops::MaxPoolWithIndexGradKernel) + int>); REGISTER_OP_CUDA_KERNEL( max_pool3d_with_index, @@ -40,4 +40,4 @@ REGISTER_OP_CUDA_KERNEL( ops::MaxPoolWithIndexGradKernel, ops::MaxPoolWithIndexGradKernel) + int>); diff --git a/paddle/fluid/operators/prelu_op.cc b/paddle/fluid/operators/prelu_op.cc index ef28114ef7..a066b3e06e 100644 --- a/paddle/fluid/operators/prelu_op.cc +++ b/paddle/fluid/operators/prelu_op.cc @@ -84,8 +84,8 @@ class PReluGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(prelu, ops::PReluOp, ops::PReluOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(prelu_grad, ops::PReluGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(prelu_grad, ops::PReluGradOp); REGISTER_OP_CPU_KERNEL( prelu, ops::PReluKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/rank_loss_op.cc b/paddle/fluid/operators/rank_loss_op.cc index 865f03ec90..eb9ff8de3e 100644 --- a/paddle/fluid/operators/rank_loss_op.cc +++ b/paddle/fluid/operators/rank_loss_op.cc @@ -122,8 +122,8 @@ class RankLossGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(rank_loss, ops::RankLossOp, ops::RankLossOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(rank_loss_grad, ops::RankLossGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(rank_loss_grad, ops::RankLossGradOp); REGISTER_OP_CPU_KERNEL( rank_loss, ops::RankLossKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/reduce_op.cc b/paddle/fluid/operators/reduce_op.cc index 97bbc1dba6..093db96647 100644 --- a/paddle/fluid/operators/reduce_op.cc +++ b/paddle/fluid/operators/reduce_op.cc @@ -191,24 +191,24 @@ class ReduceProdOpMaker : public ReduceOpMaker { namespace ops = paddle::operators; REGISTER_OPERATOR(reduce_sum, ops::ReduceOp, ops::ReduceSumOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(reduce_sum_grad, ops::ReduceGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(reduce_sum_grad, ops::ReduceGradOp); REGISTER_OPERATOR(reduce_mean, ops::ReduceOp, ops::ReduceMeanOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(reduce_mean_grad, ops::ReduceGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(reduce_mean_grad, ops::ReduceGradOp); REGISTER_OPERATOR(reduce_max, ops::ReduceOp, ops::ReduceMaxOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(reduce_max_grad, ops::ReduceGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(reduce_max_grad, ops::ReduceGradOp); REGISTER_OPERATOR(reduce_min, ops::ReduceOp, ops::ReduceMinOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(reduce_min_grad, ops::ReduceGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(reduce_min_grad, ops::ReduceGradOp); REGISTER_OPERATOR(reduce_prod, ops::ReduceOp, ops::ReduceProdOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(reduce_prod_grad, ops::ReduceGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(reduce_prod_grad, ops::ReduceGradOp); #define REGISTER_REDUCE_CPU_KERNEL(reduce_type, functor, grad_functor) \ REGISTER_OP_CPU_KERNEL(reduce_type, \ diff --git a/paddle/fluid/operators/reduce_op.h b/paddle/fluid/operators/reduce_op.h index b28dd7f209..e42b4bfe42 100644 --- a/paddle/fluid/operators/reduce_op.h +++ b/paddle/fluid/operators/reduce_op.h @@ -35,77 +35,77 @@ using EigenVector = framework::EigenVector; struct SumFunctor { template - void operator()(const DeviceContext& place, X& x, Y& y, const Dim& dim) { - y.device(place) = x.sum(dim); + void operator()(const DeviceContext& place, X* x, Y* y, const Dim& dim) { + y->device(place) = x->sum(dim); } }; struct SumGradFunctor { template - void operator()(const DeviceContext& place, X& x, Y& y, DX& dx, DY& dy, + void operator()(const DeviceContext& place, X* x, Y* y, DX* dx, DY* dy, const Dim& dim, int size) { - dx.device(place) = dy.broadcast(dim); + dx->device(place) = dy->broadcast(dim); } }; struct MeanFunctor { template - void operator()(const DeviceContext& place, X& x, Y& y, const Dim& dim) { - y.device(place) = x.mean(dim); + void operator()(const DeviceContext& place, X* x, Y* y, const Dim& dim) { + y->device(place) = x->mean(dim); } }; struct MeanGradFunctor { template - void operator()(const DeviceContext& place, X& x, Y& y, DX& dx, DY& dy, + void operator()(const DeviceContext& place, X* x, Y* y, DX* dx, DY* dy, const Dim& dim, int size) { - dx.device(place) = dy.broadcast(dim) / dx.constant(size); + dx->device(place) = dy->broadcast(dim) / dx->constant(size); } }; struct MaxFunctor { template - void operator()(const DeviceContext& place, X& x, Y& y, const Dim& dim) { - y.device(place) = x.maximum(dim); + void operator()(const DeviceContext& place, X* x, Y* y, const Dim& dim) { + y->device(place) = x->maximum(dim); } }; struct MinFunctor { template - void operator()(const DeviceContext& place, X& x, Y& y, const Dim& dim) { - y.device(place) = x.minimum(dim); + void operator()(const DeviceContext& place, X* x, Y* y, const Dim& dim) { + y->device(place) = x->minimum(dim); } }; struct MaxOrMinGradFunctor { template - void operator()(const DeviceContext& place, X& x, Y& y, DX& dx, DY& dy, + void operator()(const DeviceContext& place, X* x, Y* y, DX* dx, DY* dy, const Dim& dim, int size) { - auto equals = x == y.broadcast(dim); - auto ones = dx.constant(1); - auto zeros = dx.constant(0); + auto equals = (*x) == y->broadcast(dim); + auto ones = dx->constant(1); + auto zeros = dx->constant(0); // If there are multiple minimum or maximum elements, the subgradient of // each is the set [0, 1], and we pass gradient to all of them here. - dx.device(place) = dy.broadcast(dim) * equals.select(ones, zeros); + dx->device(place) = dy->broadcast(dim) * equals.select(ones, zeros); } }; struct ProdFunctor { template - void operator()(const DeviceContext& place, X& x, Y& y, const Dim& dim) { - y.device(place) = x.prod(dim); + void operator()(const DeviceContext& place, X* x, Y* y, const Dim& dim) { + y->device(place) = x->prod(dim); } }; struct ProdGradFunctor { template - void operator()(const DeviceContext& place, X& x, Y& y, DX& dx, DY& dy, + void operator()(const DeviceContext& place, X* x, Y* y, DX* dx, DY* dy, const Dim& dim, int size) { - dx.device(place) = dy.broadcast(dim) * y.broadcast(dim) * x.inverse(); + dx->device(place) = dy->broadcast(dim) * y->broadcast(dim) * x->inverse(); } }; @@ -125,7 +125,7 @@ class ReduceKernel : public framework::OpKernel { *context.template device_context().eigen_device(); auto reduce_dim = Eigen::array({{0}}); Functor functor; - functor(place, x, out, reduce_dim); + functor(place, &x, &out, reduce_dim); } else { int rank = context.Input("X")->dims().size(); switch (rank) { @@ -178,10 +178,10 @@ class ReduceKernel : public framework::OpKernel { if (D == 1) { auto out = EigenScalar::From(*output); - functor(place, x, out, reduce_dim); + functor(place, &x, &out, reduce_dim); } else { auto out = EigenTensor::From(*output, dims); - functor(place, x, out, reduce_dim); + functor(place, &x, &out, reduce_dim); } } }; @@ -206,7 +206,7 @@ class ReduceGradKernel : public framework::OpKernel { auto broadcast_dim = Eigen::array({{static_cast(input0->numel())}}); Functor functor; - functor(place, x, x_reduce, x_grad, x_reduce_grad, broadcast_dim, + functor(place, &x, &x_reduce, &x_grad, &x_reduce_grad, broadcast_dim, broadcast_dim[0]); } else { int rank = context.Input("X")->dims().size(); @@ -258,7 +258,7 @@ class ReduceGradKernel : public framework::OpKernel { auto& place = *context.template device_context().eigen_device(); Functor functor; - functor(place, x, x_reduce, x_grad, x_reduce_grad, broadcast_dim, + functor(place, &x, &x_reduce, &x_grad, &x_reduce_grad, broadcast_dim, broadcast_dim[dim]); } }; diff --git a/paddle/fluid/operators/reshape_op.cc b/paddle/fluid/operators/reshape_op.cc index e8ade16bde..5e5ccc3ded 100644 --- a/paddle/fluid/operators/reshape_op.cc +++ b/paddle/fluid/operators/reshape_op.cc @@ -114,8 +114,8 @@ namespace ops = paddle::operators; using CPU = paddle::platform::CPUDeviceContext; REGISTER_OPERATOR(reshape, ops::ReshapeOp, ops::ReshapeOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(reshape_grad, ops::ReshapeGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(reshape_grad, ops::ReshapeGradOp); REGISTER_OP_CPU_KERNEL(reshape, ops::ReshapeKernel, ops::ReshapeKernel, ops::ReshapeKernel, diff --git a/paddle/fluid/operators/roi_pool_op.cc b/paddle/fluid/operators/roi_pool_op.cc index 4b0ea68e0e..224ec93d28 100644 --- a/paddle/fluid/operators/roi_pool_op.cc +++ b/paddle/fluid/operators/roi_pool_op.cc @@ -154,8 +154,8 @@ https://stackoverflow.com/questions/43430056/what-is-roi-layer-in-fast-rcnn namespace ops = paddle::operators; REGISTER_OPERATOR(roi_pool, ops::ROIPoolOp, ops::ROIPoolOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(roi_pool_grad, ops::ROIPoolGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(roi_pool_grad, ops::ROIPoolGradOp); REGISTER_OP_CPU_KERNEL( roi_pool, ops::CPUROIPoolOpKernel, diff --git a/paddle/fluid/operators/row_conv_op.cc b/paddle/fluid/operators/row_conv_op.cc index 7e3d8d7d2f..23f720da0b 100644 --- a/paddle/fluid/operators/row_conv_op.cc +++ b/paddle/fluid/operators/row_conv_op.cc @@ -251,8 +251,8 @@ class RowConvGradKernel namespace ops = paddle::operators; REGISTER_OPERATOR(row_conv, ops::RowConvOp, ops::RowConvOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(row_conv_grad, ops::RowConvGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(row_conv_grad, ops::RowConvGradOp); REGISTER_OP_CPU_KERNEL( row_conv, ops::RowConvKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/save_load_combine_op_test.cc b/paddle/fluid/operators/save_load_combine_op_test.cc index 286f75df4c..2773c32a0a 100644 --- a/paddle/fluid/operators/save_load_combine_op_test.cc +++ b/paddle/fluid/operators/save_load_combine_op_test.cc @@ -23,17 +23,17 @@ USE_NO_KERNEL_OP(load_combine); int* CreateForSaveCombineOp(int x, int y, const std::vector& lod_info, std::string var_name, - paddle::platform::CPUPlace& place, - paddle::framework::Scope& scope, - paddle::framework::LoD& expect_lod) { - auto var = scope.Var(var_name); + const paddle::platform::CPUPlace& place, + paddle::framework::Scope* scope, + paddle::framework::LoD* expect_lod) { + auto var = scope->Var(var_name); auto tensor = var->GetMutable(); tensor->Resize({x, y}); - expect_lod.resize(1); + expect_lod->resize(1); for (size_t i = 0; i < lod_info.size(); i++) { - expect_lod[0].push_back(lod_info[i]); + (*expect_lod)[0].push_back(lod_info[i]); } - tensor->set_lod(expect_lod); + tensor->set_lod(*expect_lod); int* expect = tensor->mutable_data(place); for (int64_t i = 0; i < tensor->numel(); ++i) { expect[i] = static_cast(i); @@ -42,17 +42,17 @@ int* CreateForSaveCombineOp(int x, int y, const std::vector& lod_info, } paddle::framework::LoDTensor* GeneratePlaceholderBeforeLoad( - const std::string out_var_name, paddle::framework::Scope& scope) { - auto load_var = scope.Var(out_var_name); + const std::string out_var_name, paddle::framework::Scope* scope) { + auto load_var = scope->Var(out_var_name); auto target = load_var->GetMutable(); return target; } int* GetValuesAfterLoadCombineOp(paddle::framework::LoDTensor* target, - paddle::framework::Scope& scope, - paddle::framework::LoD& actual_lod) { + const paddle::framework::Scope& scope, + paddle::framework::LoD* actual_lod) { int* actual = target->data(); - actual_lod = target->lod(); + *actual_lod = target->lod(); return actual; } @@ -78,26 +78,26 @@ TEST(SaveLoadCombineOp, CPU) { std::vector lod1 = {0, 1, 2, 3, 10}; int numel1 = 100; paddle::framework::LoD expect_lod1; - int* expect1 = CreateForSaveCombineOp(10, 10, lod1, "test_var1", place, scope, - expect_lod1); + int* expect1 = CreateForSaveCombineOp(10, 10, lod1, "test_var1", place, + &scope, &expect_lod1); std::vector lod2 = {0, 2, 5, 10}; int numel2 = 200; paddle::framework::LoD expect_lod2; - int* expect2 = CreateForSaveCombineOp(10, 20, lod2, "test_var2", place, scope, - expect_lod2); + int* expect2 = CreateForSaveCombineOp(10, 20, lod2, "test_var2", place, + &scope, &expect_lod2); std::vector lod3 = {0, 2, 3, 20}; int numel3 = 4000; paddle::framework::LoD expect_lod3; int* expect3 = CreateForSaveCombineOp(20, 200, lod3, "test_var3", place, - scope, expect_lod3); + &scope, &expect_lod3); std::vector lod4 = {0, 1, 20}; int numel4 = 1000; paddle::framework::LoD expect_lod4; - int* expect4 = CreateForSaveCombineOp(20, 50, lod4, "test_var4", place, scope, - expect_lod4); + int* expect4 = CreateForSaveCombineOp(20, 50, lod4, "test_var4", place, + &scope, &expect_lod4); // Set attributes std::string filename = "check_tensor.ls"; @@ -111,10 +111,10 @@ TEST(SaveLoadCombineOp, CPU) { save_combine_op->Run(scope, place); // Set up output vars - auto target1 = GeneratePlaceholderBeforeLoad("out_var1", scope); - auto target2 = GeneratePlaceholderBeforeLoad("out_var2", scope); - auto target3 = GeneratePlaceholderBeforeLoad("out_var3", scope); - auto target4 = GeneratePlaceholderBeforeLoad("out_var4", scope); + auto target1 = GeneratePlaceholderBeforeLoad("out_var1", &scope); + auto target2 = GeneratePlaceholderBeforeLoad("out_var2", &scope); + auto target3 = GeneratePlaceholderBeforeLoad("out_var3", &scope); + auto target4 = GeneratePlaceholderBeforeLoad("out_var4", &scope); // Run the load_combine_op auto load_combine_op = paddle::framework::OpRegistry::CreateOp( @@ -123,10 +123,10 @@ TEST(SaveLoadCombineOp, CPU) { load_combine_op->Run(scope, place); paddle::framework::LoD actual_lod1, actual_lod2, actual_lod3, actual_lod4; - int* actual1 = GetValuesAfterLoadCombineOp(target1, scope, actual_lod1); - int* actual2 = GetValuesAfterLoadCombineOp(target2, scope, actual_lod2); - int* actual3 = GetValuesAfterLoadCombineOp(target3, scope, actual_lod3); - int* actual4 = GetValuesAfterLoadCombineOp(target4, scope, actual_lod4); + int* actual1 = GetValuesAfterLoadCombineOp(target1, scope, &actual_lod1); + int* actual2 = GetValuesAfterLoadCombineOp(target2, scope, &actual_lod2); + int* actual3 = GetValuesAfterLoadCombineOp(target3, scope, &actual_lod3); + int* actual4 = GetValuesAfterLoadCombineOp(target4, scope, &actual_lod4); CheckValues(expect1, actual1, expect_lod1, actual_lod1, numel1); CheckValues(expect2, actual2, expect_lod2, actual_lod2, numel2); diff --git a/paddle/fluid/operators/scatter_op.cc b/paddle/fluid/operators/scatter_op.cc index 0ad9e2ca2e..95b12455ea 100644 --- a/paddle/fluid/operators/scatter_op.cc +++ b/paddle/fluid/operators/scatter_op.cc @@ -103,7 +103,7 @@ $$ namespace ops = paddle::operators; REGISTER_OPERATOR(scatter, ops::ScatterOp, ops::ScatterOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(scatter_grad, ops::ScatterGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(scatter_grad, ops::ScatterGradOp); REGISTER_OP_CPU_KERNEL(scatter, ops::ScatterOpKernel); REGISTER_OP_CPU_KERNEL(scatter_grad, ops::ScatterGradientOpKernel); diff --git a/paddle/fluid/operators/sequence_concat_op.cc b/paddle/fluid/operators/sequence_concat_op.cc index 55631c2b91..3c21903e3a 100644 --- a/paddle/fluid/operators/sequence_concat_op.cc +++ b/paddle/fluid/operators/sequence_concat_op.cc @@ -127,7 +127,7 @@ namespace ops = paddle::operators; REGISTER_OPERATOR(sequence_concat, ops::SequenceConcatOp, ops::SequenceConcatOpMaker, paddle::framework::DefaultGradOpDescMaker< - false> /* set false to disable empty grad */) + false> /* set false to disable empty grad */); REGISTER_OPERATOR(sequence_concat_grad, ops::SequenceConcatGradOp); REGISTER_OP_CPU_KERNEL( sequence_concat, diff --git a/paddle/fluid/operators/sequence_conv_op.cc b/paddle/fluid/operators/sequence_conv_op.cc index 57a1febcc4..94f4b49b00 100644 --- a/paddle/fluid/operators/sequence_conv_op.cc +++ b/paddle/fluid/operators/sequence_conv_op.cc @@ -177,8 +177,8 @@ context_length, context_stride and context_start. namespace ops = paddle::operators; REGISTER_OPERATOR(sequence_conv, ops::SequenceConvOp, ops::SequenceConvOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(sequence_conv_grad, ops::SequenceConvGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(sequence_conv_grad, ops::SequenceConvGradOp); REGISTER_OP_CPU_KERNEL( sequence_conv, diff --git a/paddle/fluid/operators/sequence_expand_op.cc b/paddle/fluid/operators/sequence_expand_op.cc index ae05f94577..84a35d7172 100644 --- a/paddle/fluid/operators/sequence_expand_op.cc +++ b/paddle/fluid/operators/sequence_expand_op.cc @@ -202,8 +202,8 @@ class SequenceExpandOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(sequence_expand, ops::SequenceExpandOp, ops::SequenceExpandOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(sequence_expand_grad, ops::SequenceExpandOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(sequence_expand_grad, ops::SequenceExpandOpGrad); REGISTER_OP_CPU_KERNEL( sequence_expand, ops::SequenceExpandKernel, diff --git a/paddle/fluid/operators/sequence_slice_op.cc b/paddle/fluid/operators/sequence_slice_op.cc index df88121e6f..7cd620af07 100644 --- a/paddle/fluid/operators/sequence_slice_op.cc +++ b/paddle/fluid/operators/sequence_slice_op.cc @@ -122,8 +122,8 @@ NOTE: The first dimension size of input, the size of offset and Length, should b namespace ops = paddle::operators; REGISTER_OPERATOR(sequence_slice, ops::SequenceSliceOp, ops::SequenceSliceOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(sequence_slice_grad, ops::SequenceSliceGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(sequence_slice_grad, ops::SequenceSliceGradOp); REGISTER_OP_CPU_KERNEL( sequence_slice, ops::SequenceSliceOpKernel); diff --git a/paddle/fluid/operators/sequence_slice_op.cu b/paddle/fluid/operators/sequence_slice_op.cu old mode 100755 new mode 100644 diff --git a/paddle/fluid/operators/sequence_softmax_cudnn_op.cu.cc b/paddle/fluid/operators/sequence_softmax_cudnn_op.cu.cc index 5661f4b42f..0ddacb5710 100644 --- a/paddle/fluid/operators/sequence_softmax_cudnn_op.cu.cc +++ b/paddle/fluid/operators/sequence_softmax_cudnn_op.cu.cc @@ -99,7 +99,7 @@ class SequenceSoftmaxGradCUDNNKernel : public framework::OpKernel { namespace ops = paddle::operators; REGISTER_OP_KERNEL(sequence_softmax, CUDNN, ::paddle::platform::CUDAPlace, ops::SequenceSoftmaxCUDNNKernel, - ops::SequenceSoftmaxCUDNNKernel) + ops::SequenceSoftmaxCUDNNKernel); REGISTER_OP_KERNEL(sequence_softmax_grad, CUDNN, ::paddle::platform::CUDAPlace, ops::SequenceSoftmaxGradCUDNNKernel, - ops::SequenceSoftmaxGradCUDNNKernel) + ops::SequenceSoftmaxGradCUDNNKernel); diff --git a/paddle/fluid/operators/sequence_softmax_op.cc b/paddle/fluid/operators/sequence_softmax_op.cc index 47ba9a7445..a0d47c12ba 100644 --- a/paddle/fluid/operators/sequence_softmax_op.cc +++ b/paddle/fluid/operators/sequence_softmax_op.cc @@ -157,8 +157,8 @@ class SequenceSoftmaxGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(sequence_softmax, ops::SequenceSoftmaxOp, ops::SequenceSoftmaxOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(sequence_softmax_grad, ops::SequenceSoftmaxGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(sequence_softmax_grad, ops::SequenceSoftmaxGradOp); REGISTER_OP_CPU_KERNEL( sequence_softmax, ops::SequenceSoftmaxKernel, diff --git a/paddle/fluid/operators/sequence_softmax_op.cu.cc b/paddle/fluid/operators/sequence_softmax_op.cu.cc index 57adea3a1b..397df75415 100644 --- a/paddle/fluid/operators/sequence_softmax_op.cu.cc +++ b/paddle/fluid/operators/sequence_softmax_op.cu.cc @@ -18,7 +18,7 @@ namespace ops = paddle::operators; REGISTER_OP_CUDA_KERNEL( sequence_softmax, ops::SequenceSoftmaxKernel, - ops::SequenceSoftmaxKernel) + ops::SequenceSoftmaxKernel); REGISTER_OP_CUDA_KERNEL( sequence_softmax_grad, ops::SequenceSoftmaxGradKernel, diff --git a/paddle/fluid/operators/sigmoid_cross_entropy_with_logits_op.cc b/paddle/fluid/operators/sigmoid_cross_entropy_with_logits_op.cc index 442e1fef4c..5db77d0493 100644 --- a/paddle/fluid/operators/sigmoid_cross_entropy_with_logits_op.cc +++ b/paddle/fluid/operators/sigmoid_cross_entropy_with_logits_op.cc @@ -138,9 +138,9 @@ namespace ops = paddle::operators; REGISTER_OPERATOR(sigmoid_cross_entropy_with_logits, ops::SigmoidCrossEntropyWithLogitsOp, ops::SigmoidCrossEntropyWithLogitsOpMaker, - paddle::framework::DefaultGradOpDescMaker) + paddle::framework::DefaultGradOpDescMaker); REGISTER_OPERATOR(sigmoid_cross_entropy_with_logits_grad, - ops::SigmoidCrossEntropyWithLogitsGradOp) + ops::SigmoidCrossEntropyWithLogitsGradOp); REGISTER_OP_CPU_KERNEL(sigmoid_cross_entropy_with_logits, ops::SigmoidCrossEntropyWithLogitsKernel< paddle::platform::CPUDeviceContext, float>); diff --git a/paddle/fluid/operators/smooth_l1_loss_op.cc b/paddle/fluid/operators/smooth_l1_loss_op.cc index 3c15f0542b..322581fdef 100644 --- a/paddle/fluid/operators/smooth_l1_loss_op.cc +++ b/paddle/fluid/operators/smooth_l1_loss_op.cc @@ -133,8 +133,8 @@ class SmoothL1LossGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(smooth_l1_loss, ops::SmoothL1LossOp, ops::SmoothL1LossOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(smooth_l1_loss_grad, ops::SmoothL1LossGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(smooth_l1_loss_grad, ops::SmoothL1LossGradOp); REGISTER_OP_CPU_KERNEL( smooth_l1_loss, ops::SmoothL1LossKernel); diff --git a/paddle/fluid/operators/softmax_op.cc b/paddle/fluid/operators/softmax_op.cc index 7c75a45fee..2741ba95bc 100644 --- a/paddle/fluid/operators/softmax_op.cc +++ b/paddle/fluid/operators/softmax_op.cc @@ -161,8 +161,8 @@ class SoftmaxOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(softmax, ops::SoftmaxOp, ops::SoftmaxOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(softmax_grad, ops::SoftmaxOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(softmax_grad, ops::SoftmaxOpGrad); REGISTER_OP_CPU_KERNEL( softmax, ops::SoftmaxKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/spp_op.cc b/paddle/fluid/operators/spp_op.cc index f286807159..1cada95501 100644 --- a/paddle/fluid/operators/spp_op.cc +++ b/paddle/fluid/operators/spp_op.cc @@ -93,8 +93,8 @@ class SppOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(spp, ops::SppOp, ops::SppOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(spp_grad, ops::SppOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(spp_grad, ops::SppOpGrad); REGISTER_OP_CPU_KERNEL( spp, ops::SppKernel, ops::SppKernel); diff --git a/paddle/fluid/operators/squared_l2_distance_op.cc b/paddle/fluid/operators/squared_l2_distance_op.cc index 11e5faac39..c32f575b54 100644 --- a/paddle/fluid/operators/squared_l2_distance_op.cc +++ b/paddle/fluid/operators/squared_l2_distance_op.cc @@ -111,8 +111,8 @@ class SquaredL2DistanceGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(squared_l2_distance, ops::SquaredL2DistanceOp, ops::SquaredL2DistanceOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(squared_l2_distance_grad, ops::SquaredL2DistanceGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(squared_l2_distance_grad, ops::SquaredL2DistanceGradOp); REGISTER_OP_CPU_KERNEL( squared_l2_distance, ops::SquaredL2DistanceKernel); diff --git a/paddle/fluid/operators/squared_l2_norm_op.cc b/paddle/fluid/operators/squared_l2_norm_op.cc index a60c100948..4ce51259da 100644 --- a/paddle/fluid/operators/squared_l2_norm_op.cc +++ b/paddle/fluid/operators/squared_l2_norm_op.cc @@ -69,8 +69,8 @@ $$Out = \sum_{i} X_{i}^2$$ namespace ops = paddle::operators; REGISTER_OPERATOR(squared_l2_norm, ops::SquaredL2NormOp, ops::SquaredL2NormOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(squared_l2_norm_grad, ops::SquaredL2NormGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(squared_l2_norm_grad, ops::SquaredL2NormGradOp); REGISTER_OP_CPU_KERNEL( squared_l2_norm, ops::SquaredL2NormKernel); diff --git a/paddle/fluid/operators/transpose_op.cc b/paddle/fluid/operators/transpose_op.cc index 0f60dbf289..3555cb68ca 100644 --- a/paddle/fluid/operators/transpose_op.cc +++ b/paddle/fluid/operators/transpose_op.cc @@ -119,8 +119,8 @@ class TransposeOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(transpose, ops::TransposeOp, ops::TransposeOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(transpose_grad, ops::TransposeOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(transpose_grad, ops::TransposeOpGrad); REGISTER_OP_CPU_KERNEL( transpose, ops::TransposeKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/operators/unpool_op.cc b/paddle/fluid/operators/unpool_op.cc index 92a79269c2..b3cd87efa2 100644 --- a/paddle/fluid/operators/unpool_op.cc +++ b/paddle/fluid/operators/unpool_op.cc @@ -133,8 +133,8 @@ class UnpoolOpGrad : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(unpool, ops::UnpoolOp, ops::Unpool2dOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(unpool_grad, ops::UnpoolOpGrad) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(unpool_grad, ops::UnpoolOpGrad); REGISTER_OP_CPU_KERNEL( unpool, ops::UnpoolKernel, ops::UnpoolKernel); diff --git a/paddle/fluid/operators/warpctc_op.cc b/paddle/fluid/operators/warpctc_op.cc index ed81b5d266..6835a5dd62 100644 --- a/paddle/fluid/operators/warpctc_op.cc +++ b/paddle/fluid/operators/warpctc_op.cc @@ -133,8 +133,8 @@ class WarpCTCGradOp : public framework::OperatorWithKernel { namespace ops = paddle::operators; REGISTER_OPERATOR(warpctc, ops::WarpCTCOp, ops::WarpCTCOpMaker, - paddle::framework::DefaultGradOpDescMaker) -REGISTER_OPERATOR(warpctc_grad, ops::WarpCTCGradOp) + paddle::framework::DefaultGradOpDescMaker); +REGISTER_OPERATOR(warpctc_grad, ops::WarpCTCGradOp); REGISTER_OP_CPU_KERNEL( warpctc, ops::WarpCTCKernel); REGISTER_OP_CPU_KERNEL( diff --git a/paddle/fluid/pybind/protobuf.cc b/paddle/fluid/pybind/protobuf.cc index 93533e5c9d..7de7f84a3d 100644 --- a/paddle/fluid/pybind/protobuf.cc +++ b/paddle/fluid/pybind/protobuf.cc @@ -127,6 +127,8 @@ void BindProgramDesc(pybind11::module *m) { .def("block", &pd::ProgramDesc::MutableBlock, pybind11::return_value_policy::reference) .def("num_blocks", &pd::ProgramDesc::Size) + .def("get_feed_target_names", &pd::ProgramDesc::GetFeedTargetNames) + .def("get_fetch_target_names", &pd::ProgramDesc::GetFetchTargetNames) .def("serialize_to_string", SerializeMessage) .def("parse_from_string", [](pd::ProgramDesc &program_desc, const std::string &data) { @@ -299,6 +301,7 @@ void BindOpDesc(pybind11::module *m) { .def("check_attrs", &pd::OpDesc::CheckAttrs) .def("infer_shape", &pd::OpDesc::InferShape) .def("infer_var_type", &pd::OpDesc::InferVarType) + .def("set_is_target", &pd::OpDesc::SetIsTarget) .def("serialize_to_string", SerializeMessage) .def("block", &pd::OpDesc::Block, pybind11::return_value_policy::reference); diff --git a/paddle/fluid/pybind/pybind.cc b/paddle/fluid/pybind/pybind.cc index 19bd30d966..64d92cac7e 100644 --- a/paddle/fluid/pybind/pybind.cc +++ b/paddle/fluid/pybind/pybind.cc @@ -294,7 +294,7 @@ All parameter, weight, gradient are variables in Paddle. const std::vector> &targets) { ProgramDesc prog_with_targets(origin); for (const auto &t : targets) { - prog_with_targets.MutableBlock(t[0])->Op(t[1])->MarkAsTarget(); + prog_with_targets.MutableBlock(t[0])->Op(t[1])->SetIsTarget(true); } proto::ProgramDesc pruned_desc; Prune(*prog_with_targets.Proto(), &pruned_desc); diff --git a/python/paddle/fluid/framework.py b/python/paddle/fluid/framework.py index 4b841ef31d..5e6c6204c5 100644 --- a/python/paddle/fluid/framework.py +++ b/python/paddle/fluid/framework.py @@ -1070,6 +1070,12 @@ class Program(object): for t in targets: if not isinstance(t, Operator): if isinstance(t, Variable): + if t.op is None: + global_block = self.global_block() + for op in global_block.ops: + if t.name in op.output_arg_names: + t.op = op + break t = t.op else: raise ValueError(("All targets of prune() can only be " diff --git a/python/paddle/fluid/io.py b/python/paddle/fluid/io.py index 1c0f1f6eb4..bf4d81233d 100644 --- a/python/paddle/fluid/io.py +++ b/python/paddle/fluid/io.py @@ -340,6 +340,13 @@ def save_inference_model(dirname, if not os.path.isdir(dirname): os.makedirs(dirname) + # Clear the is_target information and remove the existed feed and fetch op + global_block = main_program.global_block() + for i, op in enumerate(global_block.ops): + op.desc.set_is_target(False) + if op.type == "feed" or op.type == "fetch": + global_block.remove_op(i) + pruned_program = main_program.prune(targets=target_vars) inference_program = pruned_program.inference_optimize() fetch_var_names = [v.name for v in target_vars] @@ -362,24 +369,6 @@ def save_inference_model(dirname, save_persistables(executor, dirname, inference_program, params_filename) -def get_feed_targets_names(program): - feed_targets_names = [] - global_block = program.global_block() - for op in global_block.ops: - if op.desc.type() == 'feed': - feed_targets_names.insert(0, op.desc.output('Out')[0]) - return feed_targets_names - - -def get_fetch_targets_names(program): - fetch_targets_names = [] - global_block = program.global_block() - for op in global_block.ops: - if op.desc.type() == 'fetch': - fetch_targets_names.append(op.desc.input('X')[0]) - return fetch_targets_names - - def load_inference_model(dirname, executor, model_filename=None, @@ -418,8 +407,8 @@ def load_inference_model(dirname, program = Program.parse_from_string(program_desc_str) load_persistables(executor, dirname, program, params_filename) - feed_target_names = get_feed_targets_names(program) - fetch_target_names = get_fetch_targets_names(program) + feed_target_names = program.desc.get_feed_target_names() + fetch_target_names = program.desc.get_fetch_target_names() fetch_targets = [ program.global_block().var(name) for name in fetch_target_names ] diff --git a/python/paddle/fluid/tests/book/test_image_classification.py b/python/paddle/fluid/tests/book/test_image_classification.py index 0027b651e8..d3c14b83fa 100644 --- a/python/paddle/fluid/tests/book/test_image_classification.py +++ b/python/paddle/fluid/tests/book/test_image_classification.py @@ -248,6 +248,10 @@ def infer(use_cuda, save_dirname=None): print("infer results: ", results[0]) + fluid.io.save_inference_model(save_dirname, feed_target_names, + fetch_targets, exe, + inference_transpiler_program) + def main(net_type, use_cuda, is_local=True): if use_cuda and not fluid.core.is_compiled_with_cuda():