MCPcopy
hub / github.com/NVIDIA/TensorRT-LLM / has_per_group_scaling

Method has_per_group_scaling

tensorrt_llm/quantization/mode.py:161–162  ·  view source on GitHub ↗
(self)

Source from the content-addressed store, hash-verified

159 return self._any(self.PER_CHANNEL)
160
161 def has_per_group_scaling(self):
162 return self._any(self.PER_GROUP)
163
164 def has_int8_kv_cache(self):
165 return self._any(self.INT8_KV_CACHE)

Callers 15

to_dictMethod · 0.95
build_gptFunction · 0.80
buildFunction · 0.80
get_quant_methodFunction · 0.80
has_int8_woq_per_channelMethod · 0.80
__init__Method · 0.80
__init__Method · 0.80
forwardMethod · 0.80
forwardMethod · 0.80
forwardMethod · 0.80
forwardMethod · 0.80
quantizeFunction · 0.80

Calls 1

_anyMethod · 0.95

Tested by

no test coverage detected