This greatly simplifies the implementation and removes custom parser functionality. The necessary methods are added to the DialectAsmParser.
PiperOrigin-RevId: 278015983
This allows for them to be used on other non-function, or even other function-like, operations. The algorithms are already generic, so this is simply changing the derived pass type. The majority of this change is just ensuring that the nesting of these passes remains the same, as the pass manager won't auto-nest them anymore.
PiperOrigin-RevId: 276573038
The first dim length of the axisStats attribute should equals to the slice size
of the input argument when splitted by the axis dimension.
PiperOrigin-RevId: 272798042
The CL adds a rounding mode flag to the class and changes the default to rmNearestTiesToAway from rmNearestTiesToEven because 1) Tensorflow QuantizeV2 ops uses rmNearestTiesToAway; 2) the specialization only implements rmNearestTiesToAway.
PiperOrigin-RevId: 270600739
A new converter with per axis quantization parameters is added to quantize a
dense elements attribute. For each slice along the quantization axis, it
creates an uniform quantized value converter, with different scale and zero
point, and quantizes the values in the slice.
The current implementation doesn't handle sparse elements attributes.
PiperOrigin-RevId: 270121986
Since we apply nudging for the zero point to make sure the nudged zerop points
can be in the range of [qmin, qmax], the constraint that rmin / rmax should
stride zero isn't necessary.
This also matches the documentation of tensorflow's FakeQuantWithMinMaxArgs op,
where min and max don't need to stride zero:
https://www.tensorflow.org/api_docs/python/tf/quantization/fake_quant_with_min_max_args
PiperOrigin-RevId: 268296285
Comparing to the existing quant.const_fake_quant op, the min and max attributes
of this new op is for each channel of last dimension of the input.
PiperOrigin-RevId: 268093722
Some TensorFlow simulated quantize ops such as QuantizeAndDequantizeV2Op have
attribute for the sign of the quantization, so quant_ConstFakeQuant should be
able to represent it with the new attribute is added.
The method for converting these attributes to an QuantizedType is updated to
handle this new argument.
PiperOrigin-RevId: 258810290
The current syntax separates the name and value with ':', but ':' is already overloaded by several other things(e.g. trailing types). This makes the syntax difficult to parse in some situtations:
Old:
"foo: 10 : i32"
New:
"foo = 10 : i32"
PiperOrigin-RevId: 255097928
This is the standard syntax for types on operations, and is also already used by IntegerAttr and FloatAttr.
Example:
dense<5> : tensor<i32>
dense<[3]> : tensor<1xi32>
PiperOrigin-RevId: 255069157
This name has caused some confusion because it suggests that it's running op verification (and that this verification isn't getting run by default).
PiperOrigin-RevId: 254035268
Adding the additional layer of directory was discussed offline and matches the Target/ tree. The names match the defacto convention we seem to be following where the C++ namespace is ^(.+)Ops/$ matched against the directory name.
This is in preparation for patching the Quantizer into this tree, which would have been confusing without moving the Quantization dialect to its more proper home. It is left to others to move other dialects if desired.
Tested:
ninja check-mlir
--
PiperOrigin-RevId: 248171982
River Riddle