Retrieve FTRL embedding parameters with debug support.
An op that retrieves optimization parameters from embedding to host memory. Must be preceded by a ConfigureTPUEmbeddingHost op that sets up the correct embedding table configuration. For example, this op is used to retrieve updated parameters before saving a checkpoint.
Nested Classes
class | RetrieveTPUEmbeddingFTRLParametersGradAccumDebug.Options | Optional attributes for RetrieveTPUEmbeddingFTRLParametersGradAccumDebug
|
Public Methods
Output<Float> |
accumulators()
Parameter accumulators updated by the FTRL optimization algorithm.
|
static RetrieveTPUEmbeddingFTRLParametersGradAccumDebug.Options |
config(String config)
|
static RetrieveTPUEmbeddingFTRLParametersGradAccumDebug |
create(Scope scope, Long numShards, Long shardId, Options... options)
Factory method to create a class wrapping a new RetrieveTPUEmbeddingFTRLParametersGradAccumDebug operation.
|
Output<Float> |
gradientAccumulators()
Parameter gradient_accumulators updated by the FTRL optimization algorithm.
|
Output<Float> |
linears()
Parameter linears updated by the FTRL optimization algorithm.
|
Output<Float> |
parameters()
Parameter parameters updated by the FTRL optimization algorithm.
|
static RetrieveTPUEmbeddingFTRLParametersGradAccumDebug.Options |
tableId(Long tableId)
|
static RetrieveTPUEmbeddingFTRLParametersGradAccumDebug.Options |
tableName(String tableName)
|
Inherited Methods
Public Methods
public Output<Float> accumulators ()
Parameter accumulators updated by the FTRL optimization algorithm.
public static RetrieveTPUEmbeddingFTRLParametersGradAccumDebug create (Scope scope, Long numShards, Long shardId, Options... options)
Factory method to create a class wrapping a new RetrieveTPUEmbeddingFTRLParametersGradAccumDebug operation.
Parameters
scope | current scope |
---|---|
options | carries optional attributes values |
Returns
- a new instance of RetrieveTPUEmbeddingFTRLParametersGradAccumDebug
public Output<Float> gradientAccumulators ()
Parameter gradient_accumulators updated by the FTRL optimization algorithm.