mirror of https://github.com/llvm/torch-mlir
merge indices and pooling computation into one linalg generic op
lingzhiz1998
parent
04740824ae
commit
ae1fa20290
|
|
@ -13,6 +13,7 @@
|
|||
#include "mlir/Dialect/Arith/IR/Arith.h"
|
||||
#include "mlir/Dialect/Arith/Utils/Utils.h"
|
||||
#include "mlir/Dialect/Linalg/IR/Linalg.h"
|
||||
#include "mlir/Dialect/SCF/IR/SCF.h"
|
||||
#include "mlir/IR/Matchers.h"
|
||||
#include "torch-mlir/Conversion/TorchToLinalg/Utils.h"
|
||||
#include "torch-mlir/Conversion/Utils/Utils.h"
|
||||
|
|
@ -208,8 +209,6 @@ static LogicalResult createPoolingOp(
|
|||
return success();
|
||||
}
|
||||
|
||||
namespace {
|
||||
|
||||
template <typename T> struct DimensionTraits {};
|
||||
|
||||
template <> struct DimensionTraits<AtenMaxPool1dOp> {
|
||||
|
|
@ -238,247 +237,156 @@ template <>
|
|||
struct DimensionTraits<AtenMaxPool3dWithIndicesOp>
|
||||
: DimensionTraits<AtenMaxPool3dOp> {};
|
||||
|
||||
template <typename OpTy>
|
||||
LogicalResult createCustomMaxPoolingOp(
|
||||
OpTy &op, typename OpTy::Adaptor adaptor,
|
||||
ConversionPatternRewriter &rewriter, const TypeConverter *typeConverter,
|
||||
SmallVectorImpl<Value> &kernelSizeIntValues,
|
||||
SmallVectorImpl<int64_t> &strideInts, SmallVectorImpl<int64_t> &paddingInts,
|
||||
SmallVectorImpl<int64_t> &dilationInts, bool ceilMode,
|
||||
SmallVectorImpl<Value> &outTensorShape, Value &paddedInput,
|
||||
ValueRange &results,
|
||||
std::function<Value(OpBuilder &builder, Location loc,
|
||||
ValueRange iteratorDims)> &&indicesComputation =
|
||||
nullptr) {
|
||||
constexpr bool withIndices =
|
||||
llvm::is_one_of<OpTy, AtenMaxPool2dWithIndicesOp,
|
||||
AtenMaxPool3dWithIndicesOp>::value;
|
||||
constexpr int64_t Dim = DimensionTraits<OpTy>::Dim;
|
||||
|
||||
if (withIndices && !indicesComputation) {
|
||||
return op->emitError("need to provide indices computation functor for "
|
||||
"lowering maxpool with indices op");
|
||||
}
|
||||
|
||||
Value self = adaptor.getSelf();
|
||||
Type elementType = cast<RankedTensorType>(self.getType()).getElementType();
|
||||
TypedAttr smallestFPValueAttr = rewriter.getFloatAttr(
|
||||
elementType,
|
||||
APFloat::getInf(cast<mlir::FloatType>(elementType).getFloatSemantics(),
|
||||
/*Negative=*/true));
|
||||
|
||||
Value initValue =
|
||||
rewriter.create<arith::ConstantOp>(op->getLoc(), smallestFPValueAttr);
|
||||
|
||||
paddedInput = padInputTensor(op, rewriter, self, ceilMode, Dim, strideInts,
|
||||
paddingInts, initValue);
|
||||
|
||||
auto maxOutputInitialized = computeOutputTensor(
|
||||
op, rewriter, self, Dim, ceilMode, strideInts, paddingInts, dilationInts,
|
||||
kernelSizeIntValues, outTensorShape, initValue);
|
||||
|
||||
auto shape =
|
||||
castIntVectorToIndexVector(rewriter, op->getLoc(), kernelSizeIntValues);
|
||||
Value windowTensor = rewriter.create<tensor::EmptyOp>(
|
||||
op->getLoc(), getAsOpFoldResult(shape), elementType);
|
||||
|
||||
MLIRContext *context = rewriter.getContext();
|
||||
|
||||
SmallVector<mlir::AffineExpr> inputIndexing(
|
||||
{rewriter.getAffineDimExpr(0), rewriter.getAffineDimExpr(1)});
|
||||
SmallVector<mlir::AffineExpr> maxOutputIndexing = inputIndexing;
|
||||
SmallVector<mlir::AffineExpr> kernelIndexing;
|
||||
std::optional<SmallVector<mlir::AffineExpr>> indicesIndexing;
|
||||
for (int i = 0; i < Dim; i++) {
|
||||
mlir::AffineExpr poolingDim = rewriter.getAffineDimExpr(i + 2);
|
||||
mlir::AffineExpr kernelDim = rewriter.getAffineDimExpr(i + 2 + Dim);
|
||||
inputIndexing.push_back(
|
||||
poolingDim * getAffineConstantExpr(strideInts[i], context) +
|
||||
kernelDim * getAffineConstantExpr(dilationInts[i], context));
|
||||
maxOutputIndexing.push_back(poolingDim);
|
||||
kernelIndexing.push_back(kernelDim);
|
||||
}
|
||||
|
||||
auto iteratorTypes =
|
||||
SmallVector<utils::IteratorType>(2 + Dim, utils::IteratorType::parallel);
|
||||
iteratorTypes.append(Dim, utils::IteratorType::reduction);
|
||||
SmallVector<AffineMap> indexingMaps = {
|
||||
mlir::AffineMap::get(2 + Dim * 2, 0, inputIndexing, context),
|
||||
mlir::AffineMap::get(2 + Dim * 2, 0, kernelIndexing, context),
|
||||
mlir::AffineMap::get(2 + Dim * 2, 0, maxOutputIndexing, context)};
|
||||
SmallVector<mlir::Value> outputs({maxOutputInitialized});
|
||||
SmallVector<mlir::Type> outTypes({maxOutputInitialized.getType()});
|
||||
|
||||
if constexpr (withIndices) {
|
||||
// Indices tensor has same indexing/shape as max value tensor.
|
||||
indexingMaps.push_back(
|
||||
mlir::AffineMap::get(2 + Dim * 2, 0, maxOutputIndexing, context));
|
||||
RankedTensorType indicesRankedTensorType = cast<RankedTensorType>(
|
||||
typeConverter->convertType(op->getResult(1).getType()));
|
||||
Value cstMinusOne = rewriter.create<arith::ConstantOp>(
|
||||
op->getLoc(), rewriter.getI64IntegerAttr(-1));
|
||||
Value indicesOutputInitialized =
|
||||
createInitTensor(rewriter, op->getLoc(), outTensorShape,
|
||||
indicesRankedTensorType.getElementType(), cstMinusOne);
|
||||
outputs.push_back(indicesOutputInitialized);
|
||||
outTypes.push_back(indicesOutputInitialized.getType());
|
||||
}
|
||||
|
||||
results =
|
||||
rewriter
|
||||
.create<linalg::GenericOp>(
|
||||
op->getLoc(),
|
||||
/*result_types=*/outTypes,
|
||||
/*operands=*/ValueRange({paddedInput, windowTensor}),
|
||||
/*outputs=*/outputs,
|
||||
/*indexingMaps=*/indexingMaps,
|
||||
/*iteratorTypes=*/iteratorTypes,
|
||||
[&](OpBuilder &b, Location loc, ValueRange args) {
|
||||
Value currentVal = args[0], accMaxValue = args[2];
|
||||
if constexpr (withIndices) {
|
||||
Value curIndex = args[3];
|
||||
SmallVector<Value> iterators;
|
||||
for (int i = 0; i < Dim * 2; i++) {
|
||||
iterators.push_back(b.create<linalg::IndexOp>(loc, i + 2));
|
||||
}
|
||||
Value pred = b.create<arith::CmpFOp>(
|
||||
loc, arith::CmpFPredicate::UGT, currentVal, accMaxValue);
|
||||
// Consider the corner case: the max pooling result is same as
|
||||
// padding value, which is -inf. We should return the first
|
||||
// index of pooling window but not -1.
|
||||
pred = b.create<arith::OrIOp>(
|
||||
loc, pred,
|
||||
b.create<arith::CmpIOp>(
|
||||
loc, arith::CmpIPredicate::eq, curIndex,
|
||||
b.create<arith::ConstantOp>(
|
||||
loc, b.getI64IntegerAttr(-1))));
|
||||
ValueRange outResults =
|
||||
b.create<mlir::scf::IfOp>(
|
||||
loc, pred,
|
||||
[&](OpBuilder &b, Location loc) {
|
||||
SmallVector<Value> curResults{currentVal};
|
||||
if constexpr (withIndices) {
|
||||
curResults.push_back(
|
||||
indicesComputation(b, loc, iterators));
|
||||
}
|
||||
b.create<scf::YieldOp>(loc, curResults);
|
||||
},
|
||||
[&](OpBuilder &b, Location loc) {
|
||||
b.create<scf::YieldOp>(loc,
|
||||
args.drop_front(/*n=*/2));
|
||||
})
|
||||
->getResults();
|
||||
b.create<linalg::YieldOp>(loc, outResults);
|
||||
} else {
|
||||
Value max_result =
|
||||
b.create<arith::MaximumFOp>(loc, currentVal, accMaxValue);
|
||||
b.create<linalg::YieldOp>(loc, max_result);
|
||||
}
|
||||
})
|
||||
->getResults();
|
||||
|
||||
return success();
|
||||
}
|
||||
|
||||
namespace {
|
||||
template <typename OpTy>
|
||||
class ConvertAtenMaxPoolOp : public OpConversionPattern<OpTy> {
|
||||
using OpConversionPattern<OpTy>::OpConversionPattern;
|
||||
|
||||
static const bool withIndices =
|
||||
llvm::is_one_of<OpTy, AtenMaxPool2dWithIndicesOp,
|
||||
AtenMaxPool3dWithIndicesOp>::value;
|
||||
|
||||
private:
|
||||
static const int64_t Dim = DimensionTraits<OpTy>::Dim;
|
||||
|
||||
LogicalResult createPoolingMax3D(OpTy &op, typename OpTy::Adaptor adaptor,
|
||||
ConversionPatternRewriter &rewriter,
|
||||
SmallVectorImpl<Value> &kernelSizeIntValues,
|
||||
SmallVectorImpl<int64_t> &strideInts,
|
||||
SmallVectorImpl<int64_t> &paddingInts,
|
||||
SmallVectorImpl<int64_t> &dilationInts,
|
||||
bool ceilMode,
|
||||
SmallVectorImpl<Value> &outTensorShape,
|
||||
Value &paddedInput, Value &poolingOp) const {
|
||||
static_assert(Dim == 3, "op must be MaxPool3d or MaxPool3dWithIndices");
|
||||
Value self = adaptor.getSelf();
|
||||
Type elementType = cast<RankedTensorType>(self.getType()).getElementType();
|
||||
TypedAttr smallestFPValueAttr = rewriter.getFloatAttr(
|
||||
elementType,
|
||||
APFloat::getInf(cast<mlir::FloatType>(elementType).getFloatSemantics(),
|
||||
/*Negative=*/true));
|
||||
Value initValue =
|
||||
rewriter.create<arith::ConstantOp>(op->getLoc(), smallestFPValueAttr);
|
||||
|
||||
paddedInput = padInputTensor(op, rewriter, self, ceilMode, 3, strideInts,
|
||||
paddingInts, initValue);
|
||||
|
||||
auto outTensorInitialized = computeOutputTensor(
|
||||
op, rewriter, self, 3, ceilMode, strideInts, paddingInts, dilationInts,
|
||||
kernelSizeIntValues, outTensorShape, initValue);
|
||||
|
||||
auto shape =
|
||||
castIntVectorToIndexVector(rewriter, op->getLoc(), kernelSizeIntValues);
|
||||
Value windowTensor = rewriter.create<tensor::EmptyOp>(
|
||||
op->getLoc(), getAsOpFoldResult(shape), elementType);
|
||||
|
||||
MLIRContext *context = rewriter.getContext();
|
||||
|
||||
auto mapInput = mlir::AffineMap::get(
|
||||
8, 0,
|
||||
{
|
||||
rewriter.getAffineDimExpr(0), // n
|
||||
rewriter.getAffineDimExpr(1), // c
|
||||
// dim_d * stride_d + kernal_d * dilation_d
|
||||
rewriter.getAffineDimExpr(2) *
|
||||
getAffineConstantExpr(strideInts[0], context) +
|
||||
rewriter.getAffineDimExpr(5) *
|
||||
getAffineConstantExpr(dilationInts[0], context),
|
||||
// dim_h * stride_h + kernal_h * dilation_h
|
||||
rewriter.getAffineDimExpr(3) *
|
||||
getAffineConstantExpr(strideInts[1], context) +
|
||||
rewriter.getAffineDimExpr(6) *
|
||||
getAffineConstantExpr(dilationInts[1], context),
|
||||
// dim_w * stride_w + kernal_w * dilation_w
|
||||
rewriter.getAffineDimExpr(4) *
|
||||
getAffineConstantExpr(strideInts[2], context) +
|
||||
rewriter.getAffineDimExpr(7) *
|
||||
getAffineConstantExpr(dilationInts[2], context),
|
||||
},
|
||||
context);
|
||||
auto mapKernel =
|
||||
mlir::AffineMap::get(8, 0,
|
||||
{
|
||||
rewriter.getAffineDimExpr(5), // kd
|
||||
rewriter.getAffineDimExpr(6), // kh
|
||||
rewriter.getAffineDimExpr(7) // kw
|
||||
},
|
||||
context);
|
||||
auto mapOutput = mlir::AffineMap::get(
|
||||
8, 0,
|
||||
{rewriter.getAffineDimExpr(0), rewriter.getAffineDimExpr(1),
|
||||
rewriter.getAffineDimExpr(2), rewriter.getAffineDimExpr(3),
|
||||
rewriter.getAffineDimExpr(4)},
|
||||
context);
|
||||
auto iteratorTypes =
|
||||
SmallVector<utils::IteratorType>(5, utils::IteratorType::parallel);
|
||||
iteratorTypes.append(3, utils::IteratorType::reduction);
|
||||
SmallVector<AffineMap> indexingMaps = {mapInput, mapKernel, mapOutput};
|
||||
poolingOp = rewriter
|
||||
.create<linalg::GenericOp>(
|
||||
op->getLoc(),
|
||||
/* result types */ outTensorInitialized.getType(),
|
||||
/* operands */ ValueRange({paddedInput, windowTensor}),
|
||||
/* outputs */ outTensorInitialized,
|
||||
/*indexingMaps=*/indexingMaps,
|
||||
/*iteratorTypes=*/iteratorTypes,
|
||||
[&](OpBuilder &b, Location loc, ValueRange args) {
|
||||
Value currentVal = args[0], accMaxValue = args[2];
|
||||
Value max_result = b.create<arith::MaximumFOp>(
|
||||
loc, currentVal, accMaxValue);
|
||||
b.create<linalg::YieldOp>(loc, max_result);
|
||||
})
|
||||
.getResult(0);
|
||||
|
||||
return success();
|
||||
}
|
||||
|
||||
// Returns the corresponding indices of the input tensor for the max pooling
|
||||
// result tensor.
|
||||
//
|
||||
// For finding the indices, we follow the below method:
|
||||
//
|
||||
// Take maxpool2d as an example to illustrate. Let's say the input tensor is a
|
||||
// 4-d tensor. The maxpool2d and indices will also be a 4-d tensor. Then:
|
||||
// for i in range(N):
|
||||
// for j in range(C):
|
||||
// for m in range(Hout):
|
||||
// for n in range(Wout):
|
||||
// for p in range(kH):
|
||||
// for r in range(kW):
|
||||
// indexH = m * stride[0] + p * dilation[0]
|
||||
// indexW = n * stride[0] + r * dilation[0]
|
||||
// if paddedInput[i, j, indexH, indexW] ==
|
||||
// maxPool2d[i, j, m, n]:
|
||||
// indices[i, j, m, n] =
|
||||
// (indexH - padding[0]) * W +
|
||||
// (indexW - padding[1])
|
||||
//
|
||||
LogicalResult
|
||||
computeMaxPoolingIndices(Value maxPool, Value paddedInput, OpTy &op,
|
||||
typename OpTy::Adaptor adaptor,
|
||||
ConversionPatternRewriter &rewriter,
|
||||
SmallVectorImpl<Value> &outTensorShape,
|
||||
SmallVectorImpl<Value> &kernelSizeIntValues,
|
||||
SmallVectorImpl<int64_t> &strideInts,
|
||||
SmallVectorImpl<int64_t> &paddingInts,
|
||||
SmallVectorImpl<int64_t> &dilationInts, int64_t rank,
|
||||
Value &indicesResult) const {
|
||||
Location loc = op->getLoc();
|
||||
RankedTensorType indicesRankedTensorType = cast<RankedTensorType>(
|
||||
this->getTypeConverter()->convertType(op->getResult(1).getType()));
|
||||
Value cstMinusOne =
|
||||
rewriter.create<arith::ConstantOp>(loc, rewriter.getI64IntegerAttr(-1));
|
||||
Value indicesTensor =
|
||||
createInitTensor(rewriter, loc, outTensorShape,
|
||||
indicesRankedTensorType.getElementType(), cstMinusOne);
|
||||
|
||||
SmallVector<Value> kernelSize =
|
||||
castIntVectorToIndexVector(rewriter, loc, kernelSizeIntValues);
|
||||
SmallVector<Value> padding =
|
||||
getAsConstantIndexValues(rewriter, loc, paddingInts);
|
||||
SmallVector<Value> dilation =
|
||||
getAsConstantIndexValues(rewriter, loc, dilationInts);
|
||||
SmallVector<Value> kernelStride =
|
||||
getAsConstantIndexValues(rewriter, loc, strideInts);
|
||||
|
||||
Value windowTensor = rewriter.create<tensor::EmptyOp>(
|
||||
loc, getAsOpFoldResult(kernelSize),
|
||||
indicesRankedTensorType.getElementType());
|
||||
|
||||
SmallVector<AffineExpr> inputExprs, outputExprs, kernelExprs;
|
||||
for (unsigned i = 0; i < rank; i++) {
|
||||
inputExprs.push_back(rewriter.getAffineDimExpr(i));
|
||||
outputExprs.push_back(rewriter.getAffineDimExpr(i));
|
||||
}
|
||||
for (unsigned i = 0; i < rank - 2; i++) {
|
||||
kernelExprs.push_back(rewriter.getAffineDimExpr(i + rank));
|
||||
}
|
||||
|
||||
// If computing indices for maxpool2d, we have six dimensions here. Each
|
||||
// corresponding to N, C, Hout, Wout, kH, and kW, respectively, as described
|
||||
// in the algorithm above.
|
||||
SmallVector<AffineMap> indexingMaps = AffineMap::inferFromExprList(
|
||||
{inputExprs, kernelExprs, outputExprs}, rewriter.getContext());
|
||||
SmallVector<utils::IteratorType> iteratorTypes(
|
||||
rank, utils::IteratorType::parallel);
|
||||
iteratorTypes.append(rank - 2, utils::IteratorType::reduction);
|
||||
|
||||
// Extract pooling dimensions of input shape.
|
||||
SmallVector<Value> inputSubShape;
|
||||
for (unsigned i = 0; i < rank - 2; i++) {
|
||||
inputSubShape.push_back(
|
||||
getDimOp(rewriter, loc, adaptor.getSelf(), i + 2));
|
||||
}
|
||||
|
||||
indicesResult =
|
||||
rewriter
|
||||
.create<linalg::GenericOp>(
|
||||
loc, /*resultTensorTypes=*/indicesTensor.getType(),
|
||||
/*inputs=*/ValueRange({maxPool, windowTensor}),
|
||||
/*outputs=*/indicesTensor,
|
||||
/*indexingMaps=*/indexingMaps,
|
||||
/*iteratorTypes=*/iteratorTypes,
|
||||
[&](OpBuilder &b, Location loc, ValueRange args) {
|
||||
Value maxVal = args[0], res = args[2];
|
||||
|
||||
SmallVector<Value> inputDims;
|
||||
inputDims.append({b.create<linalg::IndexOp>(loc, 0),
|
||||
b.create<linalg::IndexOp>(loc, 1)});
|
||||
for (unsigned i = 2; i < rank; i++) {
|
||||
Value mainIndex = b.create<linalg::IndexOp>(loc, i);
|
||||
Value subIndex =
|
||||
b.create<linalg::IndexOp>(loc, i + rank - 2);
|
||||
Value origin = b.create<arith::MulIOp>(loc, mainIndex,
|
||||
kernelStride[i - 2]);
|
||||
Value offset =
|
||||
b.create<arith::MulIOp>(loc, subIndex, dilation[i - 2]);
|
||||
inputDims.push_back(
|
||||
b.create<arith::AddIOp>(loc, origin, offset));
|
||||
}
|
||||
|
||||
Value input =
|
||||
b.create<tensor::ExtractOp>(loc, paddedInput, inputDims);
|
||||
Value pred = b.create<arith::CmpFOp>(
|
||||
loc, arith::CmpFPredicate::OEQ, input, maxVal);
|
||||
|
||||
Value outIndex =
|
||||
b.create<arith::ConstantOp>(loc, b.getIndexAttr(0));
|
||||
Value curInputStride =
|
||||
b.create<arith::ConstantOp>(loc, b.getIndexAttr(1));
|
||||
for (unsigned i = 0; i < rank - 2; i++) {
|
||||
Value minusPadding = b.create<arith::SubIOp>(
|
||||
loc, inputDims[rank - 1 - i], padding[rank - 3 - i]);
|
||||
Value timesStride = b.create<arith::MulIOp>(
|
||||
loc, minusPadding, curInputStride);
|
||||
outIndex =
|
||||
b.create<arith::AddIOp>(loc, outIndex, timesStride);
|
||||
curInputStride = b.create<arith::MulIOp>(
|
||||
loc, curInputStride, inputSubShape[rank - 3 - i]);
|
||||
}
|
||||
Value result = b.create<arith::SelectOp>(
|
||||
loc, pred, castIndexToInt64(b, loc, outIndex), res);
|
||||
|
||||
Value predInvalidIndex = b.create<arith::CmpIOp>(
|
||||
loc, arith::CmpIPredicate::eq, res, cstMinusOne);
|
||||
Value out = b.create<arith::SelectOp>(loc, predInvalidIndex,
|
||||
result, res);
|
||||
|
||||
b.create<linalg::YieldOp>(loc, out);
|
||||
})
|
||||
.getResult(0);
|
||||
|
||||
return success();
|
||||
}
|
||||
|
||||
public:
|
||||
LogicalResult
|
||||
matchAndRewrite(OpTy op, typename OpTy::Adaptor adaptor,
|
||||
|
|
@ -546,32 +454,124 @@ public:
|
|||
paddedInput, maxPool)))
|
||||
return rewriter.notifyMatchFailure(op, "unable to compute maxpool2d");
|
||||
} else {
|
||||
if (failed(createPoolingMax3D(op, adaptor, rewriter, kernelSizeIntValues,
|
||||
strideInts, paddingInts, dilationInts,
|
||||
ceilMode, outTensorShape, paddedInput,
|
||||
maxPool)))
|
||||
ValueRange poolingResults;
|
||||
if (failed(createCustomMaxPoolingOp(
|
||||
op, adaptor, rewriter, typeConverter, kernelSizeIntValues,
|
||||
strideInts, paddingInts, dilationInts, ceilMode, outTensorShape,
|
||||
paddedInput, poolingResults)))
|
||||
return rewriter.notifyMatchFailure(op, "unable to compute maxpool3d");
|
||||
maxPool = poolingResults.front();
|
||||
}
|
||||
|
||||
Value outMaxPool = rewriter.create<tensor::CastOp>(
|
||||
op->getLoc(), maxPoolResultType, maxPool);
|
||||
SmallVector<Value> outResult({outMaxPool});
|
||||
if (withIndices) {
|
||||
Value indicesResult;
|
||||
if (failed(computeMaxPoolingIndices(
|
||||
maxPool, paddedInput, op, adaptor, rewriter, outTensorShape,
|
||||
kernelSizeIntValues, strideInts, paddingInts, dilationInts,
|
||||
selfRank, indicesResult)))
|
||||
return rewriter.notifyMatchFailure(op,
|
||||
"unable to compute maxpool indices");
|
||||
Type indicesResultType =
|
||||
typeConverter->convertType(op->getResult(1).getType());
|
||||
Value outIndices = rewriter.create<tensor::CastOp>(
|
||||
op->getLoc(), indicesResultType, indicesResult);
|
||||
outResult.push_back(outIndices);
|
||||
}
|
||||
rewriter.replaceOp(op, outResult);
|
||||
rewriter.replaceOpWithNewOp<tensor::CastOp>(op, maxPoolResultType, maxPool);
|
||||
return success();
|
||||
}
|
||||
};
|
||||
} // namespace
|
||||
|
||||
namespace {
|
||||
template <typename OpTy>
|
||||
class ConvertAtenMaxPoolWithIndicesOp : public OpConversionPattern<OpTy> {
|
||||
using OpConversionPattern<OpTy>::OpConversionPattern;
|
||||
|
||||
private:
|
||||
static const int64_t Dim = DimensionTraits<OpTy>::Dim;
|
||||
|
||||
public:
|
||||
LogicalResult
|
||||
matchAndRewrite(OpTy op, typename OpTy::Adaptor adaptor,
|
||||
ConversionPatternRewriter &rewriter) const override {
|
||||
if (failed(verifyLinalgCompatibleTypes(op, rewriter)))
|
||||
return failure();
|
||||
|
||||
Location loc = op->getLoc();
|
||||
const TypeConverter *typeConverter = this->getTypeConverter();
|
||||
|
||||
bool ceilMode;
|
||||
SmallVector<Value, Dim> kernelSizeIntValues;
|
||||
SmallVector<int64_t, Dim> strideInts, paddingInts, dilationInts;
|
||||
if (!matchPattern(op.getDilation(),
|
||||
m_TorchListOfConstantInts(dilationInts)))
|
||||
return rewriter.notifyMatchFailure(op,
|
||||
"only support constant int dilations");
|
||||
|
||||
if (failed(checkAndGetPoolingParameters<OpTy>(op, rewriter, typeConverter,
|
||||
ceilMode, kernelSizeIntValues,
|
||||
strideInts, paddingInts)))
|
||||
return rewriter.notifyMatchFailure(op, "invalid pooling parameters");
|
||||
|
||||
// Initialize padding/dilation/kernelStride to help computing indices
|
||||
// correspond to max pooling values.
|
||||
SmallVector<Value> kernelSize =
|
||||
castIntVectorToIndexVector(rewriter, loc, kernelSizeIntValues);
|
||||
SmallVector<Value> padding =
|
||||
getAsConstantIndexValues(rewriter, loc, paddingInts);
|
||||
SmallVector<Value> dilation =
|
||||
getAsConstantIndexValues(rewriter, loc, dilationInts);
|
||||
SmallVector<Value> kernelStride =
|
||||
getAsConstantIndexValues(rewriter, loc, strideInts);
|
||||
// Extract pooling dimensions of input shape.
|
||||
SmallVector<Value> inputSubShape;
|
||||
for (int i = 0; i < Dim; i++) {
|
||||
inputSubShape.push_back(
|
||||
getDimOp(rewriter, loc, adaptor.getSelf(), i + 2));
|
||||
}
|
||||
|
||||
auto indicesComputation = [&](OpBuilder &b, Location loc,
|
||||
ValueRange iteratorDims) -> Value {
|
||||
SmallVector<Value> inputDims;
|
||||
for (int i = 0; i < Dim; i++) {
|
||||
Value origin =
|
||||
b.create<arith::MulIOp>(loc, iteratorDims[i], kernelStride[i]);
|
||||
Value offset =
|
||||
b.create<arith::MulIOp>(loc, iteratorDims[i + Dim], dilation[i]);
|
||||
inputDims.push_back(b.create<arith::AddIOp>(loc, origin, offset));
|
||||
}
|
||||
|
||||
Value outIndex = b.create<arith::ConstantOp>(loc, b.getIndexAttr(0));
|
||||
Value curInputStride =
|
||||
b.create<arith::ConstantOp>(loc, b.getIndexAttr(1));
|
||||
Value validIndex =
|
||||
b.create<arith::ConstantOp>(loc, b.getIntegerAttr(b.getI1Type(), 1));
|
||||
Value cstZero = b.create<arith::ConstantOp>(loc, b.getIndexAttr(0));
|
||||
for (int i = 0; i < Dim; i++) {
|
||||
Value minusPadding = b.create<arith::SubIOp>(
|
||||
loc, inputDims[Dim - 1 - i], padding[Dim - 1 - i]);
|
||||
validIndex = b.create<arith::AndIOp>(
|
||||
loc, validIndex,
|
||||
b.create<arith::CmpIOp>(loc, arith::CmpIPredicate::sge,
|
||||
minusPadding, cstZero));
|
||||
Value timesStride =
|
||||
b.create<arith::MulIOp>(loc, minusPadding, curInputStride);
|
||||
outIndex = b.create<arith::AddIOp>(loc, outIndex, timesStride);
|
||||
curInputStride = b.create<arith::MulIOp>(loc, curInputStride,
|
||||
inputSubShape[Dim - 1 - i]);
|
||||
}
|
||||
return b.create<arith::SelectOp>(
|
||||
loc, validIndex, castIndexToInt64(b, loc, outIndex),
|
||||
b.create<arith::ConstantOp>(loc, b.getI64IntegerAttr(-1)));
|
||||
};
|
||||
|
||||
Value paddedInput;
|
||||
SmallVector<Value, 4> outTensorShape;
|
||||
ValueRange results;
|
||||
if (failed(createCustomMaxPoolingOp(
|
||||
op, adaptor, rewriter, typeConverter, kernelSizeIntValues,
|
||||
strideInts, paddingInts, dilationInts, ceilMode, outTensorShape,
|
||||
paddedInput, results, std::move(indicesComputation))))
|
||||
return rewriter.notifyMatchFailure(
|
||||
op, "unable to compute maxpool with indices");
|
||||
|
||||
Type maxPoolResultType =
|
||||
typeConverter->convertType(op->getResult(0).getType());
|
||||
Type indicesResultType =
|
||||
typeConverter->convertType(op->getResult(1).getType());
|
||||
Value outMaxpool = rewriter.create<tensor::CastOp>(loc, maxPoolResultType,
|
||||
results.front());
|
||||
Value outIndices =
|
||||
rewriter.create<tensor::CastOp>(loc, indicesResultType, results.back());
|
||||
|
||||
rewriter.replaceOp(op, {outMaxpool, outIndices});
|
||||
return success();
|
||||
}
|
||||
};
|
||||
|
|
@ -1521,10 +1521,10 @@ void mlir::torch::torch_to_linalg::populatePoolingPatternsAndLegality(
|
|||
|
||||
target.addIllegalOp<AtenMaxPool2dWithIndicesOp>();
|
||||
target.addIllegalOp<AtenMaxPool3dWithIndicesOp>();
|
||||
patterns.add<ConvertAtenMaxPoolOp<AtenMaxPool2dWithIndicesOp>>(typeConverter,
|
||||
context);
|
||||
patterns.add<ConvertAtenMaxPoolOp<AtenMaxPool3dWithIndicesOp>>(typeConverter,
|
||||
context);
|
||||
patterns.add<ConvertAtenMaxPoolWithIndicesOp<AtenMaxPool2dWithIndicesOp>>(
|
||||
typeConverter, context);
|
||||
patterns.add<ConvertAtenMaxPoolWithIndicesOp<AtenMaxPool3dWithIndicesOp>>(
|
||||
typeConverter, context);
|
||||
|
||||
target.addIllegalOp<AtenMaxUnpool3dOp>();
|
||||
patterns.add<ConvertAtenMaxUnpool3dOp>(typeConverter, context);
|
||||
|
|
|
|||
Loading…
Reference in New Issue