Index

A

AWQConfig (class in quark.torch.quantization.config.config)

AwqProcessor (class in quark.torch.algorithm.awq.awq)

B

backward() (quark.torch.kernel.DequantE4M3Function static method) (quark.torch.kernel.FakeQuantizePerChannelAffine static method) (quark.torch.kernel.FakeQuantizePerTensorAffine static method) (quark.torch.kernel.QuantDequantE4M3Function static method) (quark.torch.kernel.QuantE4M3Function static method)

BaseAlgoProcessor (class in quark.torch.algorithm.processor) BaseQuantModelForCausalLM (class in quark.torch.algorithm.models.base) block_sizes (quark.torch.quantization.tensor_quantize.FakeQuantize property)

C

calculate_qparams() (quark.torch.quantization.observer.observer.UniformScalingObserver method) check_min_max_valid() (in module quark.torch.quantization.nn.utils)

Config (class in quark.torch.quantization.config.config) CustomFormatter (class in quark.torch.utils.log)

D

DequantE4M3Function (class in quark.torch.kernel) DeviceType (class in quark.torch.quantization.config.type)

Dtype (class in quark.torch.quantization.config.type) DuplicateFilter (class in quark.torch.utils.log)

E

EmbeddingType (class in quark.torch.export.json_export.builder.llm_model_info) export_amax() (quark.torch.quantization.tensor_quantize.FakeQuantize method) export_model_info() (quark.torch.export.api.ModelExporter method) export_onnx_model() (quark.torch.export.api.ModelExporter method)

ExporterConfig (class in quark.torch.export.config.config) extra_repr() (quark.torch.quantization.observer.observer.PlaceholderObserver method) (quark.torch.quantization.observer.observer.UniformScalingObserver method) (quark.torch.quantization.tensor_quantize.FakeQuantize method)

F

FakeQuantize (class in quark.torch.quantization.tensor_quantize) FakeQuantizeBase (class in quark.torch.quantization.tensor_quantize) FakeQuantizePerChannelAffine (class in quark.torch.kernel) FakeQuantizePerTensorAffine (class in quark.torch.kernel) filter() (quark.torch.utils.log.DuplicateFilter method) format() (quark.torch.utils.log.CustomFormatter method) forward() (quark.torch.kernel.DequantE4M3Function static method) (quark.torch.kernel.FakeQuantizePerChannelAffine static method) (quark.torch.kernel.FakeQuantizePerTensorAffine static method) (quark.torch.kernel.QuantDequantE4M3Function static method) (quark.torch.kernel.QuantE4M3Function static method) (quark.torch.quantization.observer.observer.PerTensorHistogramObserver method) (quark.torch.quantization.observer.observer.PerTensorMinMaxObserver method)

freeze() (quark.torch.quantization.api.ModelQuantizer method) FreezedFakeQuantize (class in quark.torch.quantization.tensor_quantize)

G

get_min_max_by_mse() (quark.torch.quantization.observer.observer.PerTensorMSEObserver method) get_min_max_by_percentile() (quark.torch.quantization.observer.observer.PerTensorPercentileObserver method)

get_nested_attr_from_module() (in module quark.torch.algorithm.utils.module) GPTQConfig (class in quark.torch.quantization.config.config) GptqProcessor (class in quark.torch.algorithm.gptq.gptq)

J

JsonExporterConfig (class in quark.torch.export.config.config)

L

LayerNormType (class in quark.torch.export.json_export.builder.llm_model_info)

M

ModelExporter (class in quark.torch.export.api) ModelQuantizer (class in quark.torch.quantization.api) module quark quark.torch quark.torch.algorithm quark.torch.algorithm.awq quark.torch.algorithm.awq.awq quark.torch.algorithm.awq.modules quark.torch.algorithm.awq.modules.act quark.torch.algorithm.awq.scale quark.torch.algorithm.awq.smooth quark.torch.algorithm.gptq quark.torch.algorithm.gptq.gptq quark.torch.algorithm.gptq.quantizer quark.torch.algorithm.models quark.torch.algorithm.models.base quark.torch.algorithm.processor quark.torch.algorithm.utils quark.torch.algorithm.utils.module quark.torch.algorithm.utils.prepare quark.torch.algorithm.utils.utils quark.torch.export quark.torch.export.api quark.torch.export.config quark.torch.export.config.config quark.torch.export.config.custom_config quark.torch.export.json_export quark.torch.export.json_export.builder quark.torch.export.json_export.builder.llm_model_info quark.torch.export.json_export.builder.llm_model_info_builder quark.torch.export.json_export.builder.native_model_info_builder quark.torch.export.json_export.converter quark.torch.export.json_export.converter.ammo_converter quark.torch.export.json_export.utils quark.torch.export.json_export.utils.utils quark.torch.export.onnx quark.torch.kernel quark.torch.kernel.hw_emulation quark.torch.kernel.hw_emulation.extensions quark.torch.kernel.hw_emulation.hw_emulation_interface quark.torch.quantization quark.torch.quantization.api quark.torch.quantization.config quark.torch.quantization.config.config quark.torch.quantization.config.config_verification quark.torch.quantization.config.custom_config quark.torch.quantization.config.type quark.torch.quantization.nn quark.torch.quantization.nn.modules quark.torch.quantization.nn.modules.mixin quark.torch.quantization.nn.modules.quantize_linear quark.torch.quantization.nn.utils quark.torch.quantization.observer quark.torch.quantization.observer.observer quark.torch.quantization.tensor_quantize quark.torch.quantization.transformation quark.torch.quantization.transformation.model_transformation quark.torch.quantization.utils quark.torch.utils quark.torch.utils.log

O

ObserverBase (class in quark.torch.quantization.observer.observer)

OnnxExporterConfig (class in quark.torch.export.config.config)

P

PerChannelMinMaxObserver (class in quark.torch.quantization.observer.observer) PerTensorHistogramObserver (class in quark.torch.quantization.observer.observer) PerTensorMinMaxObserver (class in quark.torch.quantization.observer.observer)

PerTensorMSEObserver (class in quark.torch.quantization.observer.observer) PerTensorPercentileObserver (class in quark.torch.quantization.observer.observer) PlaceholderObserver (class in quark.torch.quantization.observer.observer)

Q

QSchemeType (class in quark.torch.quantization.config.type) QuantDequantE4M3Function (class in quark.torch.kernel) QuantE4M3Function (class in quark.torch.kernel) QuantizationConfig (class in quark.torch.quantization.config.config) QuantizationSpec (class in quark.torch.quantization.config.config) quantize_model() (quark.torch.quantization.api.ModelQuantizer method) Quantizer (class in quark.torch.algorithm.gptq.quantizer) QuantLinear (class in quark.torch.quantization.nn.modules.quantize_linear) quark module quark.torch module quark.torch.algorithm module quark.torch.algorithm.awq module quark.torch.algorithm.awq.awq module quark.torch.algorithm.awq.modules module quark.torch.algorithm.awq.modules.act module quark.torch.algorithm.awq.scale module quark.torch.algorithm.awq.smooth module quark.torch.algorithm.gptq module quark.torch.algorithm.gptq.gptq module quark.torch.algorithm.gptq.quantizer module quark.torch.algorithm.models module quark.torch.algorithm.models.base module quark.torch.algorithm.processor module quark.torch.algorithm.utils module quark.torch.algorithm.utils.module module quark.torch.algorithm.utils.prepare module quark.torch.algorithm.utils.utils module quark.torch.export module quark.torch.export.api module quark.torch.export.config module quark.torch.export.config.config module quark.torch.export.config.custom_config module quark.torch.export.json_export module quark.torch.export.json_export.builder module quark.torch.export.json_export.builder.llm_model_info module

quark.torch.export.json_export.builder.llm_model_info_builder module quark.torch.export.json_export.builder.native_model_info_builder module quark.torch.export.json_export.converter module quark.torch.export.json_export.converter.ammo_converter module quark.torch.export.json_export.utils module quark.torch.export.json_export.utils.utils module quark.torch.export.onnx module quark.torch.kernel module quark.torch.kernel.hw_emulation module quark.torch.kernel.hw_emulation.extensions module quark.torch.kernel.hw_emulation.hw_emulation_interface module quark.torch.quantization module quark.torch.quantization.api module quark.torch.quantization.config module quark.torch.quantization.config.config module quark.torch.quantization.config.config_verification module quark.torch.quantization.config.custom_config module quark.torch.quantization.config.type module quark.torch.quantization.nn module quark.torch.quantization.nn.modules module quark.torch.quantization.nn.modules.mixin module quark.torch.quantization.nn.modules.quantize_linear module quark.torch.quantization.nn.utils module quark.torch.quantization.observer module quark.torch.quantization.observer.observer module quark.torch.quantization.tensor_quantize module quark.torch.quantization.transformation module quark.torch.quantization.transformation.model_transformation module quark.torch.quantization.utils module quark.torch.utils module quark.torch.utils.log module

R

reset_min_max_vals() (quark.torch.quantization.observer.observer.UniformScalingObserver method)

RoundType (class in quark.torch.quantization.config.type)

S

ScaledActivation (class in quark.torch.algorithm.awq.modules.act) ScaleType (class in quark.torch.quantization.config.type) SequentialFakeQuantize (class in quark.torch.quantization.tensor_quantize)

SmoothQuantConfig (class in quark.torch.quantization.config.config) SmoothQuantProcessor (class in quark.torch.algorithm.awq.smooth) split_model_info() (in module quark.torch.export.json_export.utils.utils)

U

UniformScalingObserver (class in quark.torch.quantization.observer.observer)