concrete.ml.quantization.quantized_ops.md
module concrete.ml.quantization.quantized_ops
concrete.ml.quantization.quantized_ops
Quantized versions of the ONNX operators for post training quantization.
class QuantizedSigmoid
QuantizedSigmoid
Quantized sigmoid op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedHardSigmoid
QuantizedHardSigmoid
Quantized HardSigmoid op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedRelu
QuantizedRelu
Quantized Relu op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedPRelu
QuantizedPRelu
Quantized PRelu op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedLeakyRelu
QuantizedLeakyRelu
Quantized LeakyRelu op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedHardSwish
QuantizedHardSwish
Quantized Hardswish op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedElu
QuantizedElu
Quantized Elu op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedSelu
QuantizedSelu
Quantized Selu op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedCelu
QuantizedCelu
Quantized Celu op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedClip
QuantizedClip
Quantized clip op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedRound
QuantizedRound
Quantized round op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedPow
QuantizedPow
Quantized pow op.
Only works for a float constant power. This operation will be fused to a (potentially larger) TLU.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedGemm
QuantizedGemm
Quantized Gemm op.
method __init__
__init__
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method q_impl
q_impl
class QuantizedMatMul
QuantizedMatMul
Quantized MatMul op.
method __init__
__init__
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method q_impl
q_impl
class QuantizedAdd
QuantizedAdd
Quantized Addition operator.
Can add either two variables (both encrypted) or a variable and a constant
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method can_fuse
can_fuse
Determine if this op can be fused.
Add operation can be computed in float and fused if it operates over inputs produced by a single integer tensor. For example the expression x + x * 1.75, where x is an encrypted tensor, can be computed with a single TLU.
Returns:
bool
: Whether the number of integer input tensors allows computing this op as a TLU
method q_impl
q_impl
class QuantizedTanh
QuantizedTanh
Quantized Tanh op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedSoftplus
QuantizedSoftplus
Quantized Softplus op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedExp
QuantizedExp
Quantized Exp op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedLog
QuantizedLog
Quantized Log op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedAbs
QuantizedAbs
Quantized Abs op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedIdentity
QuantizedIdentity
Quantized Identity op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method q_impl
q_impl
class QuantizedReshape
QuantizedReshape
Quantized Reshape op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method q_impl
q_impl
Reshape the input integer encrypted tensor.
Args:
q_inputs
: an encrypted integer tensor at index 0 and one constant shape at index 1attrs
: additional optional reshape options
Returns:
result
(QuantizedArray): reshaped encrypted integer tensor
class QuantizedConv
QuantizedConv
Quantized Conv op.
method __init__
__init__
Construct the quantized convolution operator and retrieve parameters.
Args:
n_bits_output
: number of bits for the quantization of the outputs of this operatorint_input_names
: names of integer tensors that are taken as input for this operationconstant_inputs
: the weights and activationsinput_quant_opts
: options for the input quantizerattrs
: convolution optionsdilations
(Tuple[int]): dilation of the kernel. Default to 1 on all dimensions.group
(int): number of convolution groups. Default to 1.kernel_shape
(Tuple[int]): shape of the kernel. Should have 2 elements for 2d convpads
(Tuple[int]): padding in ONNX format (begin, end) on each axisstrides
(Tuple[int]): stride of the convolution on each axis
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method q_impl
q_impl
Compute the quantized convolution between two quantized tensors.
Allows an optional quantized bias.
Args:
q_inputs
: input tuple, containsx
(numpy.ndarray): input data. Shape is N x C x H x W for 2dw
(numpy.ndarray): weights tensor. Shape is (O x I x Kh x Kw) for 2db
(numpy.ndarray, Optional): bias tensor, Shape is (O,)attrs
: convolution options handled in constructor
Returns:
res
(QuantizedArray): result of the quantized integer convolution
class QuantizedAvgPool
QuantizedAvgPool
Quantized Average Pooling op.
method __init__
__init__
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method q_impl
q_impl
class QuantizedMaxPool
QuantizedMaxPool
Quantized Max Pooling op.
method __init__
__init__
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method can_fuse
can_fuse
Determine if this op can be fused.
Max Pooling operation can not be fused since it must be performed over integer tensors and it combines different elements of the input tensors.
Returns:
bool
: False, this operation can not be fused as it adds different encrypted integers
method q_impl
q_impl
class QuantizedPad
QuantizedPad
Quantized Padding op.
method __init__
__init__
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method can_fuse
can_fuse
Determine if this op can be fused.
Pad operation cannot be fused since it must be performed over integer tensors.
Returns:
bool
: False, this operation cannot be fused as it is manipulates integer tensors
class QuantizedWhere
QuantizedWhere
Where operator on quantized arrays.
Supports only constants for the results produced on the True/False branches.
method __init__
__init__
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedCast
QuantizedCast
Cast the input to the required data type.
In FHE we only support a limited number of output types. Booleans are cast to integers.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedGreater
QuantizedGreater
Comparison operator >.
Only supports comparison with a constant.
method __init__
__init__
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedGreaterOrEqual
QuantizedGreaterOrEqual
Comparison operator >=.
Only supports comparison with a constant.
method __init__
__init__
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedLess
QuantizedLess
Comparison operator <.
Only supports comparison with a constant.
method __init__
__init__
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedLessOrEqual
QuantizedLessOrEqual
Comparison operator <=.
Only supports comparison with a constant.
method __init__
__init__
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedOr
QuantizedOr
Or operator ||.
This operation is not really working as a quantized operation. It just works when things got fused, as in e.g. Act(x) = x || (x + 42))
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedDiv
QuantizedDiv
Div operator /.
This operation is not really working as a quantized operation. It just works when things got fused, as in e.g. Act(x) = 1000 / (x + 42))
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedMul
QuantizedMul
Multiplication operator.
Only multiplies an encrypted tensor with a float constant for now. This operation will be fused to a (potentially larger) TLU.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedSub
QuantizedSub
Subtraction operator.
This works the same as addition on both encrypted - encrypted and on encrypted - constant.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method can_fuse
can_fuse
Determine if this op can be fused.
Add operation can be computed in float and fused if it operates over inputs produced by a single integer tensor. For example the expression x + x * 1.75, where x is an encrypted tensor, can be computed with a single TLU.
Returns:
bool
: Whether the number of integer input tensors allows computing this op as a TLU
method q_impl
q_impl
class QuantizedBatchNormalization
QuantizedBatchNormalization
Quantized Batch normalization with encrypted input and in-the-clear normalization params.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedFlatten
QuantizedFlatten
Quantized flatten for encrypted inputs.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method can_fuse
can_fuse
Determine if this op can be fused.
Flatten operation cannot be fused since it must be performed over integer tensors.
Returns:
bool
: False, this operation cannot be fused as it is manipulates integer tensors.
method q_impl
q_impl
Flatten the input integer encrypted tensor.
Args:
q_inputs
: an encrypted integer tensor at index 0attrs
: contains axis attribute
Returns:
result
(QuantizedArray): reshaped encrypted integer tensor
class QuantizedReduceSum
QuantizedReduceSum
ReduceSum with encrypted input.
method __init__
__init__
Construct the quantized ReduceSum operator and retrieve parameters.
Args:
n_bits_output
(int): Number of bits for the operator's quantization of outputs.int_input_names
(Optional[Set[str]]): Names of input integer tensors. Default to None.constant_inputs
(Optional[Dict]): Input constant tensor.axes
(Optional[numpy.ndarray]): Array of integers along which to reduce. The default is to reduce over all the dimensions of the input tensor if 'noop_with_empty_axes' is false, else act as an Identity op when 'noop_with_empty_axes' is true. Accepted range is [-r, r-1] where r = rank(data). Default to None.input_quant_opts
(Optional[QuantizationOptions]): Options for the input quantizer. Default to None.attrs
(dict): RecuseSum options.keepdims
(int): Keep the reduced dimension or not, 1 means keeping the input dimension, 0 will reduce it along the given axis. Default to 1.noop_with_empty_axes
(int): Defines behavior if 'axes' is empty or set to None. Default behavior with 0 is to reduce all axes. When axes is empty and this attribute is set to true 1, input tensor will not be reduced, and the output tensor would be equivalent to input tensor. Default to 0.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method calibrate
calibrate
Create corresponding QuantizedArray for the output of the activation function.
Args:
*inputs (numpy.ndarray)
: Calibration sample inputs.
Returns:
numpy.ndarray
: The output values for the provided calibration samples.
method q_impl
q_impl
Sum the encrypted tensor's values along the given axes.
Args:
q_inputs
(QuantizedArray): An encrypted integer tensor at index 0.attrs
(Dict): Options are handled in constructor.
Returns:
(QuantizedArray)
: The sum of all values along the given axes.
class QuantizedErf
QuantizedErf
Quantized erf op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedNot
QuantizedNot
Quantized Not op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedBrevitasQuant
QuantizedBrevitasQuant
Brevitas uniform quantization with encrypted input.
method __init__
__init__
Construct the Brevitas quantization operator.
Args:
n_bits_output
(int): Number of bits for the operator's quantization of outputs. Not used, will be overridden by the bit_width in ONNXint_input_names
(Optional[Set[str]]): Names of input integer tensors. Default to None.constant_inputs
(Optional[Dict]): Input constant tensor.scale
(float): Quantizer scalezero_point
(float): Quantizer zero-pointbit_width
(int): Number of bits of the integer representationinput_quant_opts
(Optional[QuantizationOptions]): Options for the input quantizer. Default to None. attrs (dict):rounding_mode
(str): Rounding mode (default and only accepted option is "ROUND")signed
(int): Whether this op quantizes to signed integers (default 1),narrow
(int): Whether this op quantizes to a narrow range of integers e.g. [-2n_bits-1 .. 2n_bits-1] (default 0),
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method calibrate
calibrate
Create corresponding QuantizedArray for the output of Quantization function.
Args:
*inputs (numpy.ndarray)
: Calibration sample inputs.
Returns:
numpy.ndarray
: the output values for the provided calibration samples.
method q_impl
q_impl
Quantize values.
Args:
q_inputs
: an encrypted integer tensor at index 0 and one constant shape at index 1attrs
: additional optional reshape options
Returns:
result
(QuantizedArray): reshaped encrypted integer tensor
class QuantizedTranspose
QuantizedTranspose
Transpose operator for quantized inputs.
This operator performs quantization, transposes the encrypted data, then dequantizes again.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method q_impl
q_impl
Reshape the input integer encrypted tensor.
Args:
q_inputs
: an encrypted integer tensor at index 0 and one constant shape at index 1attrs
: additional optional reshape options
Returns:
result
(QuantizedArray): reshaped encrypted integer tensor
class QuantizedFloor
QuantizedFloor
Quantized Floor op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedMax
QuantizedMax
Quantized Max op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedMin
QuantizedMin
Quantized Min op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedNeg
QuantizedNeg
Quantized Neg op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedSign
QuantizedSign
Quantized Neg op.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
class QuantizedUnsqueeze
QuantizedUnsqueeze
Unsqueeze operator.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method q_impl
q_impl
Unsqueeze the input tensors on a given axis.
Args:
q_inputs
: an encrypted integer tensorattrs
: additional optional unsqueeze options
Returns:
result
(QuantizedArray): unsqueezed encrypted integer tensor
class QuantizedConcat
QuantizedConcat
Concatenate operator.
property int_input_names
Get the names of encrypted integer tensors that are used by this op.
Returns:
List[str]
: the names of the tensors
method q_impl
q_impl
Concatenate the input tensors on a giver axis.
Args:
q_inputs
: an encrypted integer tensorattrs
: additional optional concatenate options
Returns:
result
(QuantizedArray): concatenated encrypted integer tensor
Last updated