Skip to content

Latest commit

 

History

History
311 lines (166 loc) · 8.29 KB

output.md

File metadata and controls

311 lines (166 loc) · 8.29 KB
Raw

模型输出

所有模型的输出都是[~utils.ModelOutput]子类的实例。这些数据结构包含模型返回的所有信息，但也可以用作元组或字典。

让我们看一个示例：

from transformers import BertTokenizer, BertForSequenceClassification
import torch

tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
model = BertForSequenceClassification.from_pretrained("bert-base-uncased")

inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
labels = torch.tensor([1]).unsqueeze(0)  # 批量大小为1
outputs = model(**inputs, labels=labels)

outputs对象是[~modeling_outputs.SequenceClassifierOutput]，正如我们在下面该类的文档中可以看到的， 它意味着它有一个可选的loss，一个logits，一个可选的hidden_states和一个可选的attentions属性。 这里我们有一个loss，因为我们传递了labels，但我们没有hidden_states和attentions，因为我们没有传递 output_hidden_states=True或output_attentions=True。

在传递output_hidden_states=True时，你可以期望outputs.hidden_states[-1]与outputs.last_hidden_states完全匹配。 然而，情况并非一直如此。有些模型在返回最后的隐藏状态时对其应用归一化或后续处理。

你可以像通常一样访问每个属性，如果该属性未被模型返回，你将得到None。 在这里，例如，outputs.loss是模型计算的损失，outputs.attentions是None。

当将我们的outputs对象视为元组时，它只考虑那些没有None值的属性。 这里，例如，它有两个元素，先是loss，然后是logits，所以

outputs[:2]

例如，将返回元组(outputs.loss, outputs.logits)。

当将我们的outputs对象视为字典时，它只考虑那些没有None值的属性。 这里，例如，它有两个键，分别是loss和logits。

这里我们记录了使用多个模型类型的通用模型输出。特定输出类型在相应的模型页面中有记录。

ModelOutput

[[autodoc]] utils.ModelOutput - to_tuple

BaseModelOutput

[[autodoc]] modeling_outputs.BaseModelOutput

BaseModelOutputWithPooling

[[autodoc]] modeling_outputs.BaseModelOutputWithPooling

BaseModelOutputWithCrossAttentions

[[autodoc]] modeling_outputs.BaseModelOutputWithCrossAttentions

BaseModelOutputWithPoolingAndCrossAttentions

[[autodoc]] modeling_outputs.BaseModelOutputWithPoolingAndCrossAttentions

BaseModelOutputWithPast

[[autodoc]] modeling_outputs.BaseModelOutputWithPast

BaseModelOutputWithPastAndCrossAttentions

[[autodoc]] modeling_outputs.BaseModelOutputWithPastAndCrossAttentions

Seq2SeqModelOutput

[[autodoc]] modeling_outputs.Seq2SeqModelOutput

CausalLMOutput

[[autodoc]] modeling_outputs.CausalLMOutput

CausalLMOutputWithCrossAttentions

[[autodoc]] modeling_outputs.CausalLMOutputWithCrossAttentions

CausalLMOutputWithPast

[[autodoc]] modeling_outputs.CausalLMOutputWithPast

MaskedLMOutput

[[autodoc]] modeling_outputs.MaskedLMOutput

Seq2SeqLMOutput

[[autodoc]] modeling_outputs.Seq2SeqLMOutput

NextSentencePredictorOutput

[[autodoc]] modeling_outputs.NextSentencePredictorOutput

SequenceClassifierOutput

[[autodoc]] modeling_outputs.SequenceClassifierOutput

Seq2SeqSequenceClassifierOutput

[[autodoc]] modeling_outputs.Seq2SeqSequenceClassifierOutput

MultipleChoiceModelOutput

[[autodoc]] modeling_outputs.MultipleChoiceModelOutput

TokenClassifierOutput

[[autodoc]] modeling_outputs.TokenClassifierOutput

QuestionAnsweringModelOutput

[[autodoc]] modeling_outputs.QuestionAnsweringModelOutput

Seq2SeqQuestionAnsweringModelOutput

[[autodoc]] modeling_outputs.Seq2SeqQuestionAnsweringModelOutput

Seq2SeqSpectrogramOutput

[[autodoc]] modeling_outputs.Seq2SeqSpectrogramOutput

SemanticSegmenterOutput

[[autodoc]] modeling_outputs.SemanticSegmenterOutput

ImageClassifierOutput

[[autodoc]] modeling_outputs.ImageClassifierOutput

ImageClassifierOutputWithNoAttention

[[autodoc]] modeling_outputs.ImageClassifierOutputWithNoAttention

DepthEstimatorOutput

[[autodoc]] modeling_outputs.DepthEstimatorOutput

Wav2Vec2BaseModelOutput

[[autodoc]] modeling_outputs.Wav2Vec2BaseModelOutput

XVectorOutput

[[autodoc]] modeling_outputs.XVectorOutput

Seq2SeqTSModelOutput

[[autodoc]] modeling_outputs.Seq2SeqTSModelOutput

Seq2SeqTSPredictionOutput

[[autodoc]] modeling_outputs.Seq2SeqTSPredictionOutput

SampleTSPredictionOutput

[[autodoc]] modeling_outputs.SampleTSPredictionOutput

TFBaseModelOutput

[[autodoc]] modeling_tf_outputs.TFBaseModelOutput

TFBaseModelOutputWithPooling

[[autodoc]] modeling_tf_outputs.TFBaseModelOutputWithPooling

TFBaseModelOutputWithPoolingAndCrossAttentions

[[autodoc]] modeling_tf_outputs.TFBaseModelOutputWithPoolingAndCrossAttentions

TFBaseModelOutputWithPast

[[autodoc]] modeling_tf_outputs.TFBaseModelOutputWithPast

TFBaseModelOutputWithPastAndCrossAttentions

[[autodoc]] modeling_tf_outputs.TFBaseModelOutputWithPastAndCrossAttentions

TFSeq2SeqModelOutput

[[autodoc]] modeling_tf_outputs.TFSeq2SeqModelOutput

TFCausalLMOutput

[[autodoc]] modeling_tf_outputs.TFCausalLMOutput

TFCausalLMOutputWithCrossAttentions

[[autodoc]] modeling_tf_outputs.TFCausalLMOutputWithCrossAttentions

TFCausalLMOutputWithPast

[[autodoc]] modeling_tf_outputs.TFCausalLMOutputWithPast

TFMaskedLMOutput

[[autodoc]] modeling_tf_outputs.TFMaskedLMOutput

TFSeq2SeqLMOutput

[[autodoc]] modeling_tf_outputs.TFSeq2SeqLMOutput

TFNextSentencePredictorOutput

[[autodoc]] modeling_tf_outputs.TFNextSentencePredictorOutput

TFSequenceClassifierOutput

[[autodoc]] modeling_tf_outputs.TFSequenceClassifierOutput

TFSeq2SeqSequenceClassifierOutput

[[autodoc]] modeling_tf_outputs.TFSeq2SeqSequenceClassifierOutput

TFMultipleChoiceModelOutput

[[autodoc]] modeling_tf_outputs.TFMultipleChoiceModelOutput

TFTokenClassifierOutput

[[autodoc]] modeling_tf_outputs.TFTokenClassifierOutput

TFQuestionAnsweringModelOutput

[[autodoc]] modeling_tf_outputs.TFQuestionAnsweringModelOutput

TFSeq2SeqQuestionAnsweringModelOutput

[[autodoc]] modeling_tf_outputs.TFSeq2SeqQuestionAnsweringModelOutput

FlaxBaseModelOutput

[[autodoc]] modeling_flax_outputs.FlaxBaseModelOutput

FlaxBaseModelOutputWithPast

[[autodoc]] modeling_flax_outputs.FlaxBaseModelOutputWithPast

FlaxBaseModelOutputWithPooling

[[autodoc]] modeling_flax_outputs.FlaxBaseModelOutputWithPooling

FlaxBaseModelOutputWithPastAndCrossAttentions

[[autodoc]] modeling_flax_outputs.FlaxBaseModelOutputWithPastAndCrossAttentions

FlaxSeq2SeqModelOutput

[[autodoc]] modeling_flax_outputs.FlaxSeq2SeqModelOutput

FlaxCausalLMOutputWithCrossAttentions

[[autodoc]] modeling_flax_outputs.FlaxCausalLMOutputWithCrossAttentions

FlaxMaskedLMOutput

[[autodoc]] modeling_flax_outputs.FlaxMaskedLMOutput

FlaxSeq2SeqLMOutput

[[autodoc]] modeling_flax_outputs.FlaxSeq2SeqLMOutput

FlaxNextSentencePredictorOutput

[[autodoc]] modeling_flax_outputs.FlaxNextSentencePredictorOutput

FlaxSequenceClassifierOutput

[[autodoc]] modeling_flax_outputs.FlaxSequenceClassifierOutput

FlaxSeq2SeqSequenceClassifierOutput

[[autodoc]] modeling_flax_outputs.FlaxSeq2SeqSequenceClassifierOutput

FlaxMultipleChoiceModelOutput

[[autodoc]] modeling_flax_outputs.FlaxMultipleChoiceModelOutput

FlaxTokenClassifierOutput

[[autodoc]] modeling_flax_outputs.FlaxTokenClassifierOutput

FlaxQuestionAnsweringModelOutput

[[autodoc]] modeling_flax_outputs.FlaxQuestionAnsweringModelOutput

FlaxSeq2SeqQuestionAnsweringModelOutput

[[autodoc]] modeling_flax_outputs.FlaxSeq2SeqQuestionAnsweringModelOutput