alexa_v0.1.xml

<?xml version="1.0"?>
<net name="torch_jit" version="11">
	<layers>
		<layer id="0" name="onnx::Flatten_0" type="Parameter" version="opset1">
			<data shape="1,16,96" element_type="f32" />
			<output>
				<port id="0" precision="FP32" names="onnx::Flatten_0">
					<dim>1</dim>
					<dim>16</dim>
					<dim>96</dim>
				</port>
			</output>
		</layer>
		<layer id="1" name="Constant_8" type="Const" version="opset1">
			<data element_type="i64" shape="2" offset="0" size="16" />
			<output>
				<port id="0" precision="I64">
					<dim>2</dim>
				</port>
			</output>
		</layer>
		<layer id="2" name="Flatten_0" type="Reshape" version="opset1">
			<data special_zero="true" />
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>16</dim>
					<dim>96</dim>
				</port>
				<port id="1" precision="I64">
					<dim>2</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP32" names="onnx::Gemm_7">
					<dim>1</dim>
					<dim>1536</dim>
				</port>
			</output>
		</layer>
		<layer id="3" name="1.weight" type="Const" version="opset1">
			<data element_type="u8" shape="128, 1536" offset="16" size="196608" />
			<output>
				<port id="0" precision="U8">
					<dim>128</dim>
					<dim>1536</dim>
				</port>
			</output>
		</layer>
		<layer id="4" name="Convert_4011" type="Convert" version="opset1">
			<data destination_type="f16" />
			<input>
				<port id="0" precision="U8">
					<dim>128</dim>
					<dim>1536</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP16">
					<dim>128</dim>
					<dim>1536</dim>
				</port>
			</output>
		</layer>
		<layer id="5" name="1.weight/zero_point" type="Const" version="opset1">
			<data element_type="u8" shape="128, 1" offset="196624" size="128" />
			<output>
				<port id="0" precision="U8">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="6" name="Convert_4014" type="Convert" version="opset1">
			<data destination_type="f16" />
			<input>
				<port id="0" precision="U8">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP16">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="7" name="1.weight/zero_point/subtract" type="Subtract" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP16">
					<dim>128</dim>
					<dim>1536</dim>
				</port>
				<port id="1" precision="FP16">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP16">
					<dim>128</dim>
					<dim>1536</dim>
				</port>
			</output>
		</layer>
		<layer id="8" name="1.weight/scale" type="Const" version="opset1">
			<data element_type="f16" shape="128, 1" offset="196752" size="256" />
			<output>
				<port id="0" precision="FP16">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="9" name="1.weight/fq_weights_1" type="Multiply" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP16">
					<dim>128</dim>
					<dim>1536</dim>
				</port>
				<port id="1" precision="FP16">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP16">
					<dim>128</dim>
					<dim>1536</dim>
				</port>
			</output>
		</layer>
		<layer id="10" name="1.weight/fq_weights_1/convert" type="Convert" version="opset1">
			<data destination_type="f32" />
			<input>
				<port id="0" precision="FP16">
					<dim>128</dim>
					<dim>1536</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP32">
					<dim>128</dim>
					<dim>1536</dim>
				</port>
			</output>
		</layer>
		<layer id="11" name="Gemm_1/WithoutBiases" type="MatMul" version="opset1">
			<data transpose_a="false" transpose_b="true" />
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>1536</dim>
				</port>
				<port id="1" precision="FP32">
					<dim>128</dim>
					<dim>1536</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="12" name="Constant_10" type="Const" version="opset1">
			<data element_type="f32" shape="" offset="197008" size="4" />
			<output>
				<port id="0" precision="FP32" />
			</output>
		</layer>
		<layer id="13" name="Multiply_13" type="Multiply" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
				<port id="1" precision="FP32" />
			</input>
			<output>
				<port id="2" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="14" name="Constant_11" type="Const" version="opset1">
			<data element_type="f32" shape="" offset="197008" size="4" />
			<output>
				<port id="0" precision="FP32" />
			</output>
		</layer>
		<layer id="15" name="1.bias" type="Const" version="opset1">
			<data element_type="f32" shape="128" offset="197012" size="512" />
			<output>
				<port id="0" precision="FP32" names="1.bias">
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="16" name="Multiply_14" type="Multiply" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP32" />
				<port id="1" precision="FP32">
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP32">
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="17" name="Gemm_1" type="Add" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
				<port id="1" precision="FP32">
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP32" names="input">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="18" name="Relu_2" type="ReLU" version="opset1">
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP32" names="onnx::Gemm_9">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="19" name="3.weight" type="Const" version="opset1">
			<data element_type="u8" shape="128, 128" offset="197524" size="16384" />
			<output>
				<port id="0" precision="U8">
					<dim>128</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="20" name="Convert_7999" type="Convert" version="opset1">
			<data destination_type="f16" />
			<input>
				<port id="0" precision="U8">
					<dim>128</dim>
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP16">
					<dim>128</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="21" name="3.weight/zero_point" type="Const" version="opset1">
			<data element_type="u8" shape="128, 1" offset="213908" size="128" />
			<output>
				<port id="0" precision="U8">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="22" name="Convert_8002" type="Convert" version="opset1">
			<data destination_type="f16" />
			<input>
				<port id="0" precision="U8">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP16">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="23" name="3.weight/zero_point/subtract" type="Subtract" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP16">
					<dim>128</dim>
					<dim>128</dim>
				</port>
				<port id="1" precision="FP16">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP16">
					<dim>128</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="24" name="3.weight/scale" type="Const" version="opset1">
			<data element_type="f16" shape="128, 1" offset="214036" size="256" />
			<output>
				<port id="0" precision="FP16">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="25" name="3.weight/fq_weights_1" type="Multiply" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP16">
					<dim>128</dim>
					<dim>128</dim>
				</port>
				<port id="1" precision="FP16">
					<dim>128</dim>
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP16">
					<dim>128</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="26" name="3.weight/fq_weights_1/convert" type="Convert" version="opset1">
			<data destination_type="f32" />
			<input>
				<port id="0" precision="FP16">
					<dim>128</dim>
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP32">
					<dim>128</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="27" name="Gemm_3/WithoutBiases" type="MatMul" version="opset1">
			<data transpose_a="false" transpose_b="true" />
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
				<port id="1" precision="FP32">
					<dim>128</dim>
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="28" name="Constant_17" type="Const" version="opset1">
			<data element_type="f32" shape="" offset="197008" size="4" />
			<output>
				<port id="0" precision="FP32" />
			</output>
		</layer>
		<layer id="29" name="Multiply_20" type="Multiply" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
				<port id="1" precision="FP32" />
			</input>
			<output>
				<port id="2" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="30" name="Constant_18" type="Const" version="opset1">
			<data element_type="f32" shape="" offset="197008" size="4" />
			<output>
				<port id="0" precision="FP32" />
			</output>
		</layer>
		<layer id="31" name="3.bias" type="Const" version="opset1">
			<data element_type="f32" shape="128" offset="214292" size="512" />
			<output>
				<port id="0" precision="FP32" names="3.bias">
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="32" name="Multiply_21" type="Multiply" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP32" />
				<port id="1" precision="FP32">
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP32">
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="33" name="Gemm_3" type="Add" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
				<port id="1" precision="FP32">
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP32" names="input.3">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="34" name="Relu_4" type="ReLU" version="opset1">
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP32" names="onnx::Gemm_11">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="35" name="5.weight" type="Const" version="opset1">
			<data element_type="u8" shape="1, 128" offset="214804" size="128" />
			<output>
				<port id="0" precision="U8">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="36" name="Convert_8010" type="Convert" version="opset1">
			<data destination_type="f16" />
			<input>
				<port id="0" precision="U8">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP16">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="37" name="5.weight/zero_point" type="Const" version="opset1">
			<data element_type="u8" shape="1, 1" offset="214932" size="1" />
			<output>
				<port id="0" precision="U8">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="38" name="Convert_8013" type="Convert" version="opset1">
			<data destination_type="f16" />
			<input>
				<port id="0" precision="U8">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP16">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="39" name="5.weight/zero_point/subtract" type="Subtract" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP16">
					<dim>1</dim>
					<dim>128</dim>
				</port>
				<port id="1" precision="FP16">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP16">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="40" name="5.weight/scale" type="Const" version="opset1">
			<data element_type="f16" shape="1, 1" offset="214933" size="2" />
			<output>
				<port id="0" precision="FP16">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="41" name="5.weight/fq_weights_1" type="Multiply" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP16">
					<dim>1</dim>
					<dim>128</dim>
				</port>
				<port id="1" precision="FP16">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP16">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="42" name="5.weight/fq_weights_1/convert" type="Convert" version="opset1">
			<data destination_type="f32" />
			<input>
				<port id="0" precision="FP16">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</output>
		</layer>
		<layer id="43" name="Gemm_5/WithoutBiases" type="MatMul" version="opset1">
			<data transpose_a="false" transpose_b="true" />
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
				<port id="1" precision="FP32">
					<dim>1</dim>
					<dim>128</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP32">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="44" name="Constant_24" type="Const" version="opset1">
			<data element_type="f32" shape="" offset="197008" size="4" />
			<output>
				<port id="0" precision="FP32" />
			</output>
		</layer>
		<layer id="45" name="Multiply_27" type="Multiply" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>1</dim>
				</port>
				<port id="1" precision="FP32" />
			</input>
			<output>
				<port id="2" precision="FP32">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="46" name="Constant_25" type="Const" version="opset1">
			<data element_type="f32" shape="" offset="197008" size="4" />
			<output>
				<port id="0" precision="FP32" />
			</output>
		</layer>
		<layer id="47" name="5.bias" type="Const" version="opset1">
			<data element_type="f32" shape="1" offset="214935" size="4" />
			<output>
				<port id="0" precision="FP32" names="5.bias">
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="48" name="Multiply_28" type="Multiply" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP32" />
				<port id="1" precision="FP32">
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP32">
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="49" name="Gemm_5" type="Add" version="opset1">
			<data auto_broadcast="numpy" />
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>1</dim>
				</port>
				<port id="1" precision="FP32">
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="2" precision="FP32" names="onnx::Sigmoid_12">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="50" name="13" type="Sigmoid" version="opset1">
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</input>
			<output>
				<port id="1" precision="FP32" names="13">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</output>
		</layer>
		<layer id="51" name="13/sink_port_0" type="Result" version="opset1" output_names="13">
			<input>
				<port id="0" precision="FP32">
					<dim>1</dim>
					<dim>1</dim>
				</port>
			</input>
		</layer>
	</layers>
	<edges>
		<edge from-layer="0" from-port="0" to-layer="2" to-port="0" />
		<edge from-layer="1" from-port="0" to-layer="2" to-port="1" />
		<edge from-layer="2" from-port="2" to-layer="11" to-port="0" />
		<edge from-layer="3" from-port="0" to-layer="4" to-port="0" />
		<edge from-layer="4" from-port="1" to-layer="7" to-port="0" />
		<edge from-layer="5" from-port="0" to-layer="6" to-port="0" />
		<edge from-layer="6" from-port="1" to-layer="7" to-port="1" />
		<edge from-layer="7" from-port="2" to-layer="9" to-port="0" />
		<edge from-layer="8" from-port="0" to-layer="9" to-port="1" />
		<edge from-layer="9" from-port="2" to-layer="10" to-port="0" />
		<edge from-layer="10" from-port="1" to-layer="11" to-port="1" />
		<edge from-layer="11" from-port="2" to-layer="13" to-port="0" />
		<edge from-layer="12" from-port="0" to-layer="13" to-port="1" />
		<edge from-layer="13" from-port="2" to-layer="17" to-port="0" />
		<edge from-layer="14" from-port="0" to-layer="16" to-port="0" />
		<edge from-layer="15" from-port="0" to-layer="16" to-port="1" />
		<edge from-layer="16" from-port="2" to-layer="17" to-port="1" />
		<edge from-layer="17" from-port="2" to-layer="18" to-port="0" />
		<edge from-layer="18" from-port="1" to-layer="27" to-port="0" />
		<edge from-layer="19" from-port="0" to-layer="20" to-port="0" />
		<edge from-layer="20" from-port="1" to-layer="23" to-port="0" />
		<edge from-layer="21" from-port="0" to-layer="22" to-port="0" />
		<edge from-layer="22" from-port="1" to-layer="23" to-port="1" />
		<edge from-layer="23" from-port="2" to-layer="25" to-port="0" />
		<edge from-layer="24" from-port="0" to-layer="25" to-port="1" />
		<edge from-layer="25" from-port="2" to-layer="26" to-port="0" />
		<edge from-layer="26" from-port="1" to-layer="27" to-port="1" />
		<edge from-layer="27" from-port="2" to-layer="29" to-port="0" />
		<edge from-layer="28" from-port="0" to-layer="29" to-port="1" />
		<edge from-layer="29" from-port="2" to-layer="33" to-port="0" />
		<edge from-layer="30" from-port="0" to-layer="32" to-port="0" />
		<edge from-layer="31" from-port="0" to-layer="32" to-port="1" />
		<edge from-layer="32" from-port="2" to-layer="33" to-port="1" />
		<edge from-layer="33" from-port="2" to-layer="34" to-port="0" />
		<edge from-layer="34" from-port="1" to-layer="43" to-port="0" />
		<edge from-layer="35" from-port="0" to-layer="36" to-port="0" />
		<edge from-layer="36" from-port="1" to-layer="39" to-port="0" />
		<edge from-layer="37" from-port="0" to-layer="38" to-port="0" />
		<edge from-layer="38" from-port="1" to-layer="39" to-port="1" />
		<edge from-layer="39" from-port="2" to-layer="41" to-port="0" />
		<edge from-layer="40" from-port="0" to-layer="41" to-port="1" />
		<edge from-layer="41" from-port="2" to-layer="42" to-port="0" />
		<edge from-layer="42" from-port="1" to-layer="43" to-port="1" />
		<edge from-layer="43" from-port="2" to-layer="45" to-port="0" />
		<edge from-layer="44" from-port="0" to-layer="45" to-port="1" />
		<edge from-layer="45" from-port="2" to-layer="49" to-port="0" />
		<edge from-layer="46" from-port="0" to-layer="48" to-port="0" />
		<edge from-layer="47" from-port="0" to-layer="48" to-port="1" />
		<edge from-layer="48" from-port="2" to-layer="49" to-port="1" />
		<edge from-layer="49" from-port="2" to-layer="50" to-port="0" />
		<edge from-layer="50" from-port="1" to-layer="51" to-port="0" />
	</edges>
	<rt_info>
		<nncf>
			<friendly_names_were_updated value="True" />
			<version value="2.18.0" />
			<weight_compression>
				<advanced_parameters value="{'statistics_path': None, 'lora_adapter_rank': 256, 'group_size_fallback_mode': 'ignore', 'min_adjusted_group_size': 16, 'awq_params': {'subset_size': 32, 'percent_to_apply': 0.002, 'alpha_min': 0.0, 'alpha_max': 1.0, 'steps': 100, 'prefer_data_aware_scaling': True}, 'scale_estimation_params': {'subset_size': 64, 'initial_steps': 5, 'scale_steps': 5, 'weight_penalty': -1.0}, 'gptq_params': {'damp_percent': 0.1, 'block_size': 128, 'subset_size': 128}, 'lora_correction_params': {'adapter_rank': 8, 'num_iterations': 3, 'apply_regularization': True, 'subset_size': 128, 'use_int8_adapters': True}, 'backend_params': {}, 'codebook': None}" />
				<all_layers value="False" />
				<awq value="False" />
				<backup_mode value="int8_asym" />
				<compression_format value="dequantize" />
				<gptq value="False" />
				<group_size value="-1" />
				<ignored_scope value="[]" />
				<lora_correction value="False" />
				<mode value="int8_asym" />
				<ratio value="1" />
				<scale_estimation value="False" />
				<sensitivity_metric value="weight_quantization_error" />
			</weight_compression>
		</nncf>
	</rt_info>
</net>