SparseSoftmaxCrossEntropyWithLogits
Mantieni tutto organizzato con le raccolte
Salva e classifica i contenuti in base alle tue preferenze.
Calcola il costo dell'entropia incrociata softmax e i gradienti per la propagazione all'indietro.
A differenza di "SoftmaxCrossEntropyWithLogits", questa operazione non accetta una matrice di probabilità di etichette, ma piuttosto una singola etichetta per riga di caratteristiche. Si ritiene che questa etichetta abbia probabilità 1,0 per la riga specificata.
Gli input sono i logit, non le probabilità.
Costanti
| Corda | OP_NAME | Il nome di questa operazione, come noto al motore principale di TensorFlow |
Metodi ereditati
Dalla classe java.lang.Object | booleano | è uguale a (Oggetto arg0) |
| Classe finale<?> | getClass () |
| int | codice hash () |
| vuoto finale | notificare () |
| vuoto finale | notificaTutti () |
| Corda | accordare () |
| vuoto finale | attendere (lungo arg0, int arg1) |
| vuoto finale | aspetta (lungo arg0) |
| vuoto finale | Aspettare () |
Costanti
Stringa finale statica pubblica OP_NAME
Il nome di questa operazione, come noto al motore principale di TensorFlow
Valore costante: "SparseSoftmaxCrossEntropyWithLogits"
Metodi pubblici
output pubblico <T> backprop ()
gradienti retropropagati (matrice batch_size x num_classes).
Metodo factory per creare una classe che racchiude una nuova operazione SparseSoftmaxCrossEntropyWithLogits.
Parametri
| scopo | ambito attuale |
|---|
| caratteristiche | matrice batch_size x num_classes |
|---|
| etichette | vettore batch_size con valori in [0, num_classes). Questa è l'etichetta per la voce minibatch specificata. |
|---|
ritorna
- una nuova istanza di SparseSoftmaxCrossEntropyWithLogits
Uscita pubblica <T> perdita ()
Perdita per esempio (vettore batch_size).
Salvo quando diversamente specificato, i contenuti di questa pagina sono concessi in base alla licenza Creative Commons Attribution 4.0, mentre gli esempi di codice sono concessi in base alla licenza Apache 2.0. Per ulteriori dettagli, consulta le norme del sito di Google Developers. Java è un marchio registrato di Oracle e/o delle sue consociate.
Ultimo aggiornamento 2025-07-26 UTC.
[null,null,["Ultimo aggiornamento 2025-07-26 UTC."],[],[],null,["# SparseSoftmaxCrossEntropyWithLogits\n\npublic final class **SparseSoftmaxCrossEntropyWithLogits** \nComputes softmax cross entropy cost and gradients to backpropagate.\n\n\nUnlike \\`SoftmaxCrossEntropyWithLogits\\`, this operation does not accept\na matrix of label probabilities, but rather a single label per row\nof features. This label is considered to have probability 1.0 for the\ngiven row.\n\n\nInputs are the logits, not probabilities.\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n### Constants\n\n|--------|----------------------------------------------------------------------------------------------------|---------------------------------------------------------|\n| String | [OP_NAME](/jvm/api_docs/java/org/tensorflow/op/nn/raw/SparseSoftmaxCrossEntropyWithLogits#OP_NAME) | The name of this op, as known by TensorFlow core engine |\n\n### Public Methods\n\n|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| [Output](/jvm/api_docs/java/org/tensorflow/Output)\\\u003cT\\\u003e | [backprop](/jvm/api_docs/java/org/tensorflow/op/nn/raw/SparseSoftmaxCrossEntropyWithLogits#backprop())() backpropagated gradients (batch_size x num_classes matrix). |\n| static \\\u003cT extends [TNumber](/jvm/api_docs/java/org/tensorflow/types/family/TNumber)\\\u003e [SparseSoftmaxCrossEntropyWithLogits](/jvm/api_docs/java/org/tensorflow/op/nn/raw/SparseSoftmaxCrossEntropyWithLogits)\\\u003cT\\\u003e | [create](/jvm/api_docs/java/org/tensorflow/op/nn/raw/SparseSoftmaxCrossEntropyWithLogits#create(org.tensorflow.op.Scope, org.tensorflow.Operand\u003cT\u003e, org.tensorflow.Operand\u003c? extends org.tensorflow.types.family.TNumber\u003e))([Scope](/jvm/api_docs/java/org/tensorflow/op/Scope) scope, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e features, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003c? extends [TNumber](/jvm/api_docs/java/org/tensorflow/types/family/TNumber)\\\u003e labels) Factory method to create a class wrapping a new SparseSoftmaxCrossEntropyWithLogits operation. |\n| [Output](/jvm/api_docs/java/org/tensorflow/Output)\\\u003cT\\\u003e | [loss](/jvm/api_docs/java/org/tensorflow/op/nn/raw/SparseSoftmaxCrossEntropyWithLogits#loss())() Per example loss (batch_size vector). |\n\n### Inherited Methods\n\nFrom class [org.tensorflow.op.RawOp](/jvm/api_docs/java/org/tensorflow/op/RawOp) \n\n|----------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| final boolean | [equals](/jvm/api_docs/java/org/tensorflow/op/RawOp#equals(java.lang.Object))(Object obj) |\n| final int | [hashCode](/jvm/api_docs/java/org/tensorflow/op/RawOp#hashCode())() |\n| [Operation](/jvm/api_docs/java/org/tensorflow/Operation) | [op](/jvm/api_docs/java/org/tensorflow/op/RawOp#op())() Return this unit of computation as a single [Operation](/jvm/api_docs/java/org/tensorflow/Operation). |\n| final String | [toString](/jvm/api_docs/java/org/tensorflow/op/RawOp#toString())() |\n\nFrom class java.lang.Object \n\n|------------------|---------------------------|\n| boolean | equals(Object arg0) |\n| final Class\\\u003c?\\\u003e | getClass() |\n| int | hashCode() |\n| final void | notify() |\n| final void | notifyAll() |\n| String | toString() |\n| final void | wait(long arg0, int arg1) |\n| final void | wait(long arg0) |\n| final void | wait() |\n\nFrom interface [org.tensorflow.op.Op](/jvm/api_docs/java/org/tensorflow/op/Op) \n\n|-----------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| abstract [ExecutionEnvironment](/jvm/api_docs/java/org/tensorflow/ExecutionEnvironment) | [env](/jvm/api_docs/java/org/tensorflow/op/Op#env())() Return the execution environment this op was created in. |\n| abstract [Operation](/jvm/api_docs/java/org/tensorflow/Operation) | [op](/jvm/api_docs/java/org/tensorflow/op/Op#op())() Return this unit of computation as a single [Operation](/jvm/api_docs/java/org/tensorflow/Operation). |\n\nConstants\n---------\n\n#### public static final String\n**OP_NAME**\n\nThe name of this op, as known by TensorFlow core engine \nConstant Value: \"SparseSoftmaxCrossEntropyWithLogits\"\n\nPublic Methods\n--------------\n\n#### public [Output](/jvm/api_docs/java/org/tensorflow/Output)\\\u003cT\\\u003e\n**backprop**\n()\n\nbackpropagated gradients (batch_size x num_classes matrix). \n\n#### public static [SparseSoftmaxCrossEntropyWithLogits](/jvm/api_docs/java/org/tensorflow/op/nn/raw/SparseSoftmaxCrossEntropyWithLogits)\\\u003cT\\\u003e\n**create**\n([Scope](/jvm/api_docs/java/org/tensorflow/op/Scope) scope, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e features, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003c? extends [TNumber](/jvm/api_docs/java/org/tensorflow/types/family/TNumber)\\\u003e labels)\n\nFactory method to create a class wrapping a new SparseSoftmaxCrossEntropyWithLogits operation. \n\n##### Parameters\n\n| scope | current scope |\n| features | batch_size x num_classes matrix |\n| labels | batch_size vector with values in \\[0, num_classes). This is the label for the given minibatch entry. |\n|----------|------------------------------------------------------------------------------------------------------|\n\n##### Returns\n\n- a new instance of SparseSoftmaxCrossEntropyWithLogits \n\n#### public [Output](/jvm/api_docs/java/org/tensorflow/Output)\\\u003cT\\\u003e\n**loss**\n()\n\nPer example loss (batch_size vector)."]]
