TensorRT  7.2.1.6
NVIDIA TensorRT
Looking for a C++ dev who knows TensorRT?
I'm looking for work. Hire me!
All Classes Namespaces Functions Variables Typedefs Enumerations Enumerator Friends Pages
tests.quant_linear_test.TestQuantLinear Class Reference

Public Member Functions

def test_raise (self)
 
def test_weight_fake_per_tensor (self)
 
def test_weight_fake_per_channel (self)
 
def test_test_input_fake_per_tensor (self)
 
def test_fake_quant_per_tensor (self)
 
def test_fake_quant_per_tensor_with_bias (self)
 
def test_fake_quant_per_channel (self)
 
def test_fake_quant_per_channel_other_precs (self)
 
def test_fake_quant_against_unquantized (self)
 
def test_set_default_quant_desc (self)
 
def test_unused_kwargs (self)
 

Member Function Documentation

◆ test_raise()

def tests.quant_linear_test.TestQuantLinear.test_raise (   self)

◆ test_weight_fake_per_tensor()

def tests.quant_linear_test.TestQuantLinear.test_weight_fake_per_tensor (   self)

◆ test_weight_fake_per_channel()

def tests.quant_linear_test.TestQuantLinear.test_weight_fake_per_channel (   self)

◆ test_test_input_fake_per_tensor()

def tests.quant_linear_test.TestQuantLinear.test_test_input_fake_per_tensor (   self)

◆ test_fake_quant_per_tensor()

def tests.quant_linear_test.TestQuantLinear.test_fake_quant_per_tensor (   self)
quantize everything, activations will scaled per tensor in ALL cases

◆ test_fake_quant_per_tensor_with_bias()

def tests.quant_linear_test.TestQuantLinear.test_fake_quant_per_tensor_with_bias (   self)
quantize everything, activations will scaled per tensor in ALL cases

◆ test_fake_quant_per_channel()

def tests.quant_linear_test.TestQuantLinear.test_fake_quant_per_channel (   self)
quantize everything, activations will scaled per tensor in ALL cases

◆ test_fake_quant_per_channel_other_precs()

def tests.quant_linear_test.TestQuantLinear.test_fake_quant_per_channel_other_precs (   self)
Test some precisions other than 8bit.

◆ test_fake_quant_against_unquantized()

def tests.quant_linear_test.TestQuantLinear.test_fake_quant_against_unquantized (   self)
Quantized Linear should introduce bounded error compare to Linear

◆ test_set_default_quant_desc()

def tests.quant_linear_test.TestQuantLinear.test_set_default_quant_desc (   self)

◆ test_unused_kwargs()

def tests.quant_linear_test.TestQuantLinear.test_unused_kwargs (   self)

The documentation for this class was generated from the following file: