saymrwulf/onnxruntime
1
0
Fork 0
mirror of https://github.com/saymrwulf/onnxruntime.git synced 2026-07-25 19:48:11 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

remove deprecated quantize api (#11263)

Browse source
This commit is contained in:
Yufeng Li 2022-04-19 19:41:55 -07:00 committed by GitHub
parent acb555c4c7
commit 2e6c2177af
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 1 additions and 80 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
onnxruntime/python/tools/quantization/__init__.py
View file

@ -1,4 +1,4 @@
from .quantize import quantize, quantize_static, quantize_dynamic
from .quantize import quantize_static, quantize_dynamic
from .quantize import QuantizationMode
from .calibrate import CalibrationDataReader, CalibraterBase, MinMaxCalibrater, create_calibrator, CalibrationMethod
from .quant_utils import QuantType, QuantFormat, write_calibration_table

79
onnxruntime/python/tools/quantization/quantize.py
View file

@ -34,85 +34,6 @@ def check_static_quant_arguments(quant_format : QuantFormat,
"Or it will lead to bad performance on x64.")
def quantize(model,
per_channel=False,
nbits=8,
quantization_mode=QuantizationMode.IntegerOps,
static=False,
symmetric_activation=False,
symmetric_weight=False,
quantization_params=None,
nodes_to_quantize=None,
nodes_to_exclude=None,
op_types_to_quantize=[]):
'''
Given an onnx model, create a quantized onnx model and save it into a file
:param model: ModelProto to quantize
:param per_channel: quantize weights per channel
:param nbits: number of bits to represent quantized data. Currently only supporting 8-bit types
:param quantization_mode: Can be one of the QuantizationMode types.
IntegerOps:
the function will use integer ops. Only ConvInteger and MatMulInteger ops are supported now.
QLinearOps:
the function will use QLinear ops. Only QLinearConv and QLinearMatMul ops are supported now.
:param static:
True: The inputs/activations are quantized using static scale and zero point values
specified through quantization_params.
False: The inputs/activations are quantized using dynamic scale and zero point values
computed while running the model.
:param symmetric_activation:
True: activations are quantized into signed integers.
False: activations are quantized into unsigned integers.
:param symmetric_weight:
True: weights are quantized into signed integers.
False: weights are quantized into unsigned integers.
:param quantization_params:
Dictionary to specify the zero point and scale values for inputs to conv and matmul nodes.
Should be specified when static is set to True.
The quantization_params should be specified in the following format:
{
"input_name": [zero_point, scale]
}.
zero_point should be of type np.uint8 and scale should be of type np.float32.
example:
{
'resnet_model/Relu_1:0': [np.uint8(0), np.float32(0.019539741799235344)],
'resnet_model/Relu_2:0': [np.uint8(0), np.float32(0.011359662748873234)]
}
:param nodes_to_quantize:
List of nodes names to quantize. When this list is not None only the nodes in this list
are quantized.
example:
[
'Conv__224',
'Conv__252'
]
:param nodes_to_exclude:
List of nodes names to exclude. The nodes in this list will be excluded from quantization
when it is not None.
:param op_types_to_quantize: specify the types of operators to quantize, like ['Conv'] to quantize Conv only. It quantizes all supported operators by default.
:return: ModelProto with quantization
'''
logging.warning("onnxruntime.quantization.quantize is deprecated.\n\
Please use quantize_static for static quantization, quantize_dynamic for dynamic quantization.")
if nbits == 8 or nbits == 7:
mode = quantization_mode
copy_model = onnx_proto.ModelProto()
copy_model.CopyFrom(model)
if not op_types_to_quantize or len(op_types_to_quantize) == 0:
op_types_to_quantize = list(QLinearOpsRegistry.keys()) if static else list(IntegerOpsRegistry.keys())
quantizer = ONNXQuantizer(copy_model, per_channel, nbits == 7, mode, static, symmetric_weight,
symmetric_activation, quantization_params, nodes_to_quantize, nodes_to_exclude,
op_types_to_quantize)
quantizer.quantize_model()
return quantizer.model.model
else:
raise ValueError('Only 8 and 7 bit quantization is currently supported')
def quantize_static(model_input,
model_output,
calibration_data_reader: CalibrationDataReader,