Code
Hub
Workspaces
Connect
Indexed graphs
Engine
MCP
copy
hub
/
github.com/NVIDIA/TensorRT-LLM
/ forward
Method
forward
tensorrt_llm/quantization/layers.py:96–97 ·
view source on GitHub ↗
(self, input)
Source
from the content-addressed store, hash-verified
94
self.axis = axis
95
96
def
forward(self, input):
97
return
dequantize(input, self.scaling_factor.value, self.axis)
98
99
100
class
SmoothQuantLinear(Linear):
Callers
1
test_dequantize
Method · 0.95
Calls
1
dequantize
Function · 0.85
Tested by
1
test_dequantize
Method · 0.76