1b97cdf885
This better matches the rest of the infrastructure, is much simpler, and makes it easier to move these types to being declaratively specified. Differential Revision: https://reviews.llvm.org/D93432
88 lines
3.3 KiB
C++
88 lines
3.3 KiB
C++
//===- UniformSupport.cpp - Support utilities for uniform quant -----------===//
|
|
//
|
|
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include "mlir/Dialect/Quant/UniformSupport.h"
|
|
#include "mlir/IR/BuiltinTypes.h"
|
|
#include <numeric>
|
|
|
|
using namespace mlir;
|
|
using namespace mlir::quant;
|
|
|
|
static bool isQuantizablePrimitiveType(Type inputType) {
|
|
return inputType.isa<FloatType>();
|
|
}
|
|
|
|
const ExpressedToQuantizedConverter
|
|
ExpressedToQuantizedConverter::forInputType(Type inputType) {
|
|
if (inputType.isa<TensorType, VectorType>()) {
|
|
Type elementType = inputType.cast<ShapedType>().getElementType();
|
|
if (!isQuantizablePrimitiveType(elementType))
|
|
return ExpressedToQuantizedConverter{inputType, nullptr};
|
|
return ExpressedToQuantizedConverter{inputType, elementType};
|
|
}
|
|
// Supported primitive type (which just is the expressed type).
|
|
if (isQuantizablePrimitiveType(inputType))
|
|
return ExpressedToQuantizedConverter{inputType, inputType};
|
|
// Unsupported.
|
|
return ExpressedToQuantizedConverter{inputType, nullptr};
|
|
}
|
|
|
|
Type ExpressedToQuantizedConverter::convert(QuantizedType elementalType) const {
|
|
assert(expressedType && "convert() on unsupported conversion");
|
|
if (auto tensorType = inputType.dyn_cast<RankedTensorType>())
|
|
return RankedTensorType::get(tensorType.getShape(), elementalType);
|
|
if (auto tensorType = inputType.dyn_cast<UnrankedTensorType>())
|
|
return UnrankedTensorType::get(elementalType);
|
|
if (auto vectorType = inputType.dyn_cast<VectorType>())
|
|
return VectorType::get(vectorType.getShape(), elementalType);
|
|
|
|
// If the expressed types match, just use the new elemental type.
|
|
if (elementalType.getExpressedType() == expressedType)
|
|
return elementalType;
|
|
// Unsupported.
|
|
return nullptr;
|
|
}
|
|
|
|
ElementsAttr
|
|
UniformQuantizedPerAxisValueConverter::convert(Attribute realValue) {
|
|
if (auto attr = realValue.dyn_cast<DenseFPElementsAttr>()) {
|
|
return convert(attr);
|
|
}
|
|
// TODO: handles sparse elements attribute
|
|
return nullptr;
|
|
}
|
|
|
|
DenseElementsAttr
|
|
UniformQuantizedPerAxisValueConverter::convert(DenseFPElementsAttr attr) {
|
|
// Creates the converter for each chunk. Normally the size of the
|
|
// quantization dim is 3, so we can cache all the converters.
|
|
ShapedType type = attr.getType();
|
|
size_t dimSize = type.getDimSize(quantizationDim);
|
|
if (dimSize != scales.size()) {
|
|
return {};
|
|
}
|
|
SmallVector<UniformQuantizedValueConverter, 4> converters;
|
|
converters.reserve(dimSize);
|
|
for (int i = 0, e = dimSize; i != e; ++i) {
|
|
converters.push_back(getPerChunkConverter(i));
|
|
}
|
|
|
|
// Scan the elements of the dense elements attributes and quantize them by
|
|
// using the right quantization parameters.
|
|
int64_t flattenIndex = 0;
|
|
auto shape = type.getShape();
|
|
int64_t chunkSize =
|
|
std::accumulate(std::next(shape.begin(), quantizationDim + 1),
|
|
shape.end(), 1, std::multiplies<int64_t>());
|
|
Type newElementType = IntegerType::get(attr.getContext(), storageBitWidth);
|
|
return attr.mapValues(newElementType, [&](const APFloat &old) {
|
|
int chunkIndex = (flattenIndex++) / chunkSize;
|
|
return converters[chunkIndex % dimSize].quantizeFloatToInt(old);
|
|
});
|
|
}
|