Skip to content

Latest commit

History

History
150 lines (83 loc) 路 3.42 KB

modeling_ort.mdx

File metadata and controls

150 lines (83 loc) 路 3.42 KB

Models

Generic model classes

The following ORT classes are available for instantiating a base model class without a specific head.

ORTModel

[[autodoc]] onnxruntime.ORTModel

Natural Language Processing

The following ORT classes are available for the following natural language processing tasks.

ORTModelForCausalLM

[[autodoc]] onnxruntime.ORTModelForCausalLM - forward

ORTModelForMaskedLM

[[autodoc]] onnxruntime.ORTModelForMaskedLM

ORTModelForSeq2SeqLM

[[autodoc]] onnxruntime.ORTModelForSeq2SeqLM - forward

ORTModelForSequenceClassification

[[autodoc]] onnxruntime.ORTModelForSequenceClassification

ORTModelForTokenClassification

[[autodoc]] onnxruntime.ORTModelForTokenClassification

ORTModelForMultipleChoice

[[autodoc]] onnxruntime.ORTModelForMultipleChoice

ORTModelForQuestionAnswering

[[autodoc]] onnxruntime.ORTModelForQuestionAnswering

Computer vision

The following ORT classes are available for the following computer vision tasks.

ORTModelForImageClassification

[[autodoc]] onnxruntime.ORTModelForImageClassification

ORTModelForSemanticSegmentation

[[autodoc]] onnxruntime.ORTModelForSemanticSegmentation

Audio

The following ORT classes are available for the following audio tasks.

ORTModelForAudioClassification

[[autodoc]] onnxruntime.ORTModelForAudioClassification

ORTModelForAudioFrameClassification

[[autodoc]] onnxruntime.ORTModelForAudioFrameClassification

ORTModelForCTC

[[autodoc]] onnxruntime.ORTModelForCTC

ORTModelForSpeechSeq2Seq

[[autodoc]] onnxruntime.ORTModelForSpeechSeq2Seq - forward

ORTModelForAudioXVector

[[autodoc]] onnxruntime.ORTModelForAudioXVector

Multimodal

The following ORT classes are available for the following multimodal tasks.

ORTModelForVision2Seq

[[autodoc]] onnxruntime.ORTModelForVision2Seq - forward

ORTModelForPix2Struct

[[autodoc]] onnxruntime.ORTModelForPix2Struct - forward

Custom Tasks

The following ORT classes are available for the following custom tasks.

ORTModelForCustomTasks

[[autodoc]] onnxruntime.ORTModelForCustomTasks

ORTModelForFeatureExtraction

[[autodoc]] onnxruntime.ORTModelForFeatureExtraction

Stable Diffusion

ORTStableDiffusionPipeline

[[autodoc]] onnxruntime.ORTStableDiffusionPipeline - call

ORTStableDiffusionImg2ImgPipeline

[[autodoc]] onnxruntime.ORTStableDiffusionImg2ImgPipeline - call

ORTStableDiffusionInpaintPipeline

[[autodoc]] onnxruntime.ORTStableDiffusionInpaintPipeline - call

ORTStableDiffusionXLPipeline

[[autodoc]] onnxruntime.ORTStableDiffusionXLPipeline - call

ORTStableDiffusionXLImg2ImgPipeline

[[autodoc]] onnxruntime.ORTStableDiffusionXLImg2ImgPipeline - call

ORTLatentConsistencyModelPipeline

[[autodoc]] onnxruntime.ORTLatentConsistencyModelPipeline - call