neural_compressor.experimental
Intel® Neural Compressor: An open-source Python library supporting popular model compression techniques.
Subpackages
neural_compressor.experimental.commonneural_compressor.experimental.common.criterionneural_compressor.experimental.common.dataloaderneural_compressor.experimental.common.metricneural_compressor.experimental.common.modelneural_compressor.experimental.common.optimizerneural_compressor.experimental.common.postprocessneural_compressor.experimental.common.torch_utils
neural_compressor.experimental.compressionneural_compressor.experimental.contribneural_compressor.experimental.dataneural_compressor.experimental.data.dataloadersneural_compressor.experimental.data.dataloaders.base_dataloaderneural_compressor.experimental.data.dataloaders.dataloaderneural_compressor.experimental.data.dataloaders.default_dataloaderneural_compressor.experimental.data.dataloaders.fetcherneural_compressor.experimental.data.dataloaders.mxnet_dataloaderneural_compressor.experimental.data.dataloaders.onnxrt_dataloaderneural_compressor.experimental.data.dataloaders.pytorch_dataloaderneural_compressor.experimental.data.dataloaders.samplerneural_compressor.experimental.data.dataloaders.tensorflow_dataloader
neural_compressor.experimental.data.datasetsneural_compressor.experimental.data.datasets.bert_datasetneural_compressor.experimental.data.datasets.coco_datasetneural_compressor.experimental.data.datasets.datasetneural_compressor.experimental.data.datasets.dummy_datasetneural_compressor.experimental.data.datasets.dummy_dataset_v2neural_compressor.experimental.data.datasets.imagenet_datasetneural_compressor.experimental.data.datasets.style_transfer_dataset
neural_compressor.experimental.data.filtersneural_compressor.experimental.data.transforms
neural_compressor.experimental.exportneural_compressor.experimental.metricneural_compressor.experimental.metric.bleuneural_compressor.experimental.metric.bleu_utilneural_compressor.experimental.metric.coco_label_mapneural_compressor.experimental.metric.coco_toolsneural_compressor.experimental.metric.evaluate_squadneural_compressor.experimental.metric.f1neural_compressor.experimental.metric.metric
neural_compressor.experimental.nasneural_compressor.experimental.pruner_legacyneural_compressor.experimental.pruning_recipesneural_compressor.experimental.pytorch_prunerneural_compressor.experimental.pytorch_pruner.loggerneural_compressor.experimental.pytorch_pruner.patternsneural_compressor.experimental.pytorch_pruner.prune_utilsneural_compressor.experimental.pytorch_pruner.prunerneural_compressor.experimental.pytorch_pruner.pruningneural_compressor.experimental.pytorch_pruner.scheduler
neural_compressor.experimental.strategyneural_compressor.experimental.strategy.utilsneural_compressor.experimental.strategy.auto_mixed_precisionneural_compressor.experimental.strategy.basicneural_compressor.experimental.strategy.bayesianneural_compressor.experimental.strategy.exhaustiveneural_compressor.experimental.strategy.mseneural_compressor.experimental.strategy.mse_v2neural_compressor.experimental.strategy.randomneural_compressor.experimental.strategy.strategy
Submodules
neural_compressor.experimental.benchmarkneural_compressor.experimental.componentneural_compressor.experimental.distillationneural_compressor.experimental.graph_optimizationneural_compressor.experimental.mixed_precisionneural_compressor.experimental.model_conversionneural_compressor.experimental.pruningneural_compressor.experimental.pruning_v2neural_compressor.experimental.quantizationneural_compressor.experimental.scheduler
Package Contents
Classes
|
This is base class of Neural Compressor Component. |
|
This class provides easy use API for quantization. |
|
This is base class of pruning object. |
|
Benchmark class is used to evaluate the model performance with the objective settings. |
|
Graph_Optimization class. |
|
Class used for generating low precision model. |
|
ModelConversion class is used to convert one model format to another. |
|
Distillation class derived from Component class. |
|
Create object of different NAS approaches. |
Attributes
|