未验证 提交 65fd84a5 编写于 作者: K kavyasrinet 提交者: GitHub

Updating Var_desc.md with the updated typing system in Fluid (#8462)

* Updating Var_desc.md with the updated typing system in Fluid

* Added Channel to VarType
上级 bd58bf3e
## Background ## Background
PaddlePaddle divides the description of neural network computation into two stages: compile time and runtime. At compile time, the neural network computation is described as a `ProgramDesc` whereas at runtime an `Executor` interprets the `ProgramDesc` to compute the operations. PaddlePaddle divides the description of neural network computation into two stages: compile time and runtime. At compile time, the neural network computation is described as a `ProgramDesc` whereas at runtime an `Executor` interprets the `ProgramDesc` to compute the operations.
PaddlePaddle use proto message to describe compile time program because PaddlePaddle uses proto message to describe compile time program because :
1. The computation program description must be serializable and saved in a file. 1. The computation program description must be serializable and saved in a file.
1. During distributed training, the sreialized program will be sent to multiple workers. It should also be possible to break the program into different components, each of which can be executed on different workers. 1. During distributed training, the serialized program will be sent to multiple workers. It should also be possible to break the program into different components, each of which can be executed on a different worker.
The computation `Program` consists of nested `Blocks`. Each `Block` will consist of data(i.e. `Variable`) and `Operations`. The concept to represent them is in the table below. The computation `Program` consists of nested `Blocks`. Each `Block` will consist of data(i.e. `Variable`) and `Operations`. The concept to represent them is in the table below.
...@@ -14,28 +14,33 @@ The computation `Program` consists of nested `Blocks`. Each `Block` will consist ...@@ -14,28 +14,33 @@ The computation `Program` consists of nested `Blocks`. Each `Block` will consist
|Operation|OpDesc(proto)|Operator(cpp)| |Operation|OpDesc(proto)|Operator(cpp)|
## Definition of VarDesc ## Definition of VarType
A VarDesc should have a name, and value. The are two kinds of variable type in compile time, they are `LoDTensor` and `SelectedRows`. A VarDesc should have a name, type and whether or not it is persistable. The are different kinds of variable types supported in PaddlePaddle, apart from the POD_Types like: `LOD_TENSOR`, `SELECTED_ROWS`, `FEED_MINIBATCH`, `FETCH_LIST`, `STEP_SCOPES`, `LOD_RANK_TABLE`, `LOD_TENSOR_ARRAY`, `PLACE_LIST`, `READER` and `CHANNEL`. These are declared inside `VarType`. A `VarDesc` then looks as the following:
```proto ```proto
message VarDesc { message VarDesc {
required string name = 1; required string name = 1;
enum VarType {
LOD_TENSOR = 0;
SELECTED_ROWS = 1;
}
required VarType type = 2; required VarType type = 2;
optional LoDTensorDesc lod_desc = 3; optional bool persistable = 3 [ default = false ];
optional TensorDesc selected_rows_desc = 4;
optional bool persistable = 5 [ default = false ];
} }
``` ```
## Definition of TensorDesc ## Definition of TensorDesc
```proto ```proto
enum DataType { message TensorDesc {
// Should only be PODType. Is enforced in C++
required Type data_type = 1;
repeated int64 dims = 2; // [UNK, 640, 480] is saved as [-1, 640, 480]
}
```
The `Type` here comes from the enum defined inside of `VarType` :
```proto
enum Type {
// Pod Types
BOOL = 0; BOOL = 0;
INT16 = 1; INT16 = 1;
INT32 = 2; INT32 = 2;
...@@ -43,11 +48,18 @@ enum DataType { ...@@ -43,11 +48,18 @@ enum DataType {
FP16 = 4; FP16 = 4;
FP32 = 5; FP32 = 5;
FP64 = 6; FP64 = 6;
}
message TensorDesc { // Other types that may need additional descriptions
required DataType data_type = 1; LOD_TENSOR = 7;
repeated int64 dims = 2; // [UNK, 640, 480] is saved as [-1, 640, 480] SELECTED_ROWS = 8;
FEED_MINIBATCH = 9;
FETCH_LIST = 10;
STEP_SCOPES = 11;
LOD_RANK_TABLE = 12;
LOD_TENSOR_ARRAY = 13;
PLACE_LIST = 14;
READER = 15;
CHANNEL = 16;
} }
``` ```
...@@ -58,7 +70,7 @@ A TensorDesc describes `SelectedRows` and `LoDTensor`. For details of `SelectedR ...@@ -58,7 +70,7 @@ A TensorDesc describes `SelectedRows` and `LoDTensor`. For details of `SelectedR
```proto ```proto
message LoDTensorDesc { message LoDTensorDesc {
required TensorDesc tensor = 1; required TensorDesc tensor = 1;
optional int lod_level = 2; optional int32 lod_level = 2 [ default = 0 ];
} }
``` ```
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册