Input parameters for the backward split layer
More...
◆ addInputGradient()
virtual services::Status addInputGradient |
( |
const data_management::TensorPtr & |
igTensor, |
|
|
size_t |
index |
|
) |
| |
|
virtual |
Adds tensor with input gradient to the input object of the backward split layer
- Parameters
-
[in] | igTensor | Tensor with input gradient |
[in] | index | Index of the tensor with input gradient |
- Returns
- Status of computations
Reimplemented from Input.
◆ check()
Checks an input object of the backward split layer
- Parameters
-
[in] | par | Algorithm parameter |
[in] | method | Computation method |
- Returns
- Status of computations
Reimplemented from Input.
◆ get() [1/2]
Returns a tensor with a given index from the collection of input tensors
- Parameters
-
[in] | id | Identifier of the collection of input tensors |
[in] | index | Index of the tensor to be returned |
- Returns
- Pointer to the table with the input tensor
◆ get() [2/2]
Returns input Tensor of the layer algorithm
- Parameters
-
[in] | id | Identifier of the input tensor |
- Returns
- Input tensor that corresponds to the given identifier
◆ getLayout()
Returns the layout of the input object for the layer algorithm
- Returns
- Layout of the input object for the layer algorithm
Reimplemented from Input.
◆ set() [1/2]
void set |
( |
InputLayerDataId |
id, |
|
|
const data_management::TensorPtr & |
value, |
|
|
size_t |
index |
|
) |
| |
Sets an input object for the backward split layer
- Parameters
-
[in] | id | Identifier of the input object |
[in] | value | Pointer to the input object |
[in] | index | Index of the tensor to be set |
◆ set() [2/2]
Sets input for the layer algorithm
- Parameters
-
[in] | id | Identifier of the input object |
[in] | ptr | Pointer to the object |
◆ setInputFromForward()
virtual services::Status setInputFromForward |
( |
layers::forward::ResultPtr |
result | ) |
|
|
virtual |
Sets input structure retrieved from the result of the forward layer
- Parameters
-
[in] | result | Result of the forward layer |
Reimplemented from Input.
The documentation for this class was generated from the following file: