OnnxToTorch support for onnx.InstanceNormalization op

include/torch-mlir/Dialect/Torch/IR/GeneratedTorchOps.td

@@ -5973,6 +5973,37 @@ def Torch_AtenBatchNormOp : Torch_Op<"aten.batch_norm", [
   }];
 }
 
+def Torch_AtenInstanceNormOp : Torch_Op<"aten.instance_norm", [
+    AllowsTypeRefinement,
+    HasValueSemantics,
+    ReadOnly
+  ]> {
+  let summary = "Generated op for `aten::instance_norm : (Tensor, Tensor?, Tensor?, Tensor?, Tensor?, bool, float, float, bool) -> (Tensor)`";
+  let arguments = (ins
+    AnyTorchTensorType:$input,
+    AnyTorchOptionalTensorType:$weight,
+    AnyTorchOptionalTensorType:$bias,
+    AnyTorchOptionalTensorType:$running_mean,
+    AnyTorchOptionalTensorType:$running_var,
+    Torch_BoolType:$use_input_stats,
+    Torch_FloatType:$momentum,
+    Torch_FloatType:$eps,
+    Torch_BoolType:$cudnn_enabled
+  );
+  let results = (outs
+    AnyTorchTensorType:$result
+  );
+  let hasCustomAssemblyFormat = 1;
+  let extraClassDefinition = [{
+    ParseResult AtenInstanceNormOp::parse(OpAsmParser &parser, OperationState &result) {
+      return parseDefaultTorchOp(parser, result, 9, 1);
+    }
+    void AtenInstanceNormOp::print(OpAsmPrinter &printer) {
+      printDefaultTorchOp(printer, *this, 9, 1);
+    }
+  }];
+}
+
 def Torch_AtenNativeGroupNormOp : Torch_Op<"aten.native_group_norm", [
     AllowsTypeRefinement,
     HasValueSemantics,

lib/Conversion/TorchOnnxToTorch/DefaultDomainGtoP.cpp

@@ -335,6 +335,35 @@ void mlir::torch::onnx_c::populateDefaultDomainGtoP(
                       binder.op, resultType, lhs, rhs);
                   return success();
                 });
+  patterns.onOp(
+      "InstanceNormalization", 6,
+      [](OpBinder binder, ConversionPatternRewriter &rewriter) {
+        Torch::ValueTensorType resultType;
+        llvm::SmallVector<Value> operands;
+        float eps;
+
+        if (binder.tensorOperands(operands, 3) ||
+            binder.tensorResultType(resultType) || operands.size() != 3 ||
+            binder.f32FloatAttr(eps, "epsilon", 1e-05f)) {
+          return failure();
+        }
+        Value none = rewriter.create<Torch::ConstantNoneOp>(binder.getLoc());
+        Value boolFalse =
+            rewriter.create<Torch::ConstantBoolOp>(binder.getLoc(), false);
+        auto epsValue = rewriter.create<Torch::ConstantFloatOp>(
+            binder.getLoc(), rewriter.getF64FloatAttr(eps));
+
+        auto momentum = rewriter.create<Torch::ConstantFloatOp>(
+            binder.getLoc(), rewriter.getF64FloatAttr(0.0f));
+        rewriter.replaceOpWithNewOp<Torch::AtenInstanceNormOp>(
+            binder.op, resultType, /* input */ operands[0],
+            /* weight */ operands[1],
+            /* bias */ operands[2], /* running mean */ none,
+            /* running var */ none,
+            /* use input stats */ boolFalse, momentum, epsValue,
+            /* cudnn enabled */ boolFalse);
+        return success();
+      });
   patterns.onOp(
       "Max", 1, [](OpBinder binder, ConversionPatternRewriter &rewriter) {
         Torch::ValueTensorType resultType;

lib/Conversion/TorchToLinalg/Uncategorized.cpp

@@ -1826,6 +1826,199 @@ class ConvertAtenBatchNormOp : public OpConversionPattern<AtenBatchNormOp> {
 };
 } // namespace
 
+namespace {
+class ConvertAtenInstanceNormOp
+    : public OpConversionPattern<AtenInstanceNormOp> {
+public:
+  using OpConversionPattern::OpConversionPattern;
+  LogicalResult
+  matchAndRewrite(AtenInstanceNormOp op, OpAdaptor adaptor,
+                  ConversionPatternRewriter &rewriter) const override {
+    MLIRContext *context = op->getContext();
+    Location loc = op->getLoc();
+    Value input = adaptor.getInput();
+    Value scale = adaptor.getWeight();
+    Value bias = adaptor.getBias();
+    Value eps = adaptor.getEps();
+
+    auto inputType = input.getType().cast<RankedTensorType>();
+    auto inputRank = inputType.getRank();
+
+    SmallVector<AffineExpr, 2> ncExpr;
+    ncExpr.push_back(mlir::getAffineDimExpr(0, context));
+    ncExpr.push_back(mlir::getAffineDimExpr(1, context));
+
+    auto ncIndexingMap = AffineMap::get(
+        /*dimCount=*/inputRank,
+        /*symbolCount=*/0, ncExpr, context);
+
+    SmallVector<AffineExpr, 1> cExpr;
+    cExpr.push_back(mlir::getAffineDimExpr(1, context));
+
+    auto cIndexingMap = AffineMap::get(
+        /*dimCount=*/inputRank,
+        /*symbolCount=*/0, cExpr, context);
+
+    SmallVector<AffineMap, 2> indexingMaps = {
+        rewriter.getMultiDimIdentityMap(inputRank), // input
+        ncIndexingMap,                              // output
+    };
+
+    Type resultElementType = inputType.getElementType();
+    auto inputSize = getTensorSizes(rewriter, loc, input);
+    SmallVector<Value> ncSize({inputSize[0], inputSize[1]});
+
+    Value meanTensor =
+        createZeroInitTensor(rewriter, loc, ncSize, resultElementType);
+    Value varTensor =
+        createZeroInitTensor(rewriter, loc, ncSize, resultElementType);
+
+    SmallVector<utils::IteratorType> iteratorTypes = {
+        utils::IteratorType::parallel, utils::IteratorType::parallel,
+        utils::IteratorType::reduction, utils::IteratorType::reduction};
+
+    Value sumPool2d =
+        rewriter
+            .create<linalg::GenericOp>(
+                loc, meanTensor.getType(), ValueRange{input}, meanTensor,
+                /*indexingMaps=*/indexingMaps,
+                /*iteratorTypes=*/iteratorTypes,
+                [&](OpBuilder &b, Location loc, ValueRange args) {
+                  Value input = args[0], sum = args[1];
+                  Value result = b.create<arith::AddFOp>(loc, input, sum);
+                  b.create<linalg::YieldOp>(loc, result);
+                })
+            .getResult(0);
+
+    indexingMaps = {
+        rewriter.getMultiDimIdentityMap(2), // sumPool2d
+        rewriter.getMultiDimIdentityMap(2), // output
+    };
+
+    iteratorTypes = {utils::IteratorType::parallel,
+                     utils::IteratorType::parallel};
+    Value mean =
+        rewriter
+            .create<linalg::GenericOp>(
+                loc, meanTensor.getType(), ValueRange{sumPool2d}, meanTensor,
+                indexingMaps, iteratorTypes,
+                [&](OpBuilder &b, Location loc, ValueRange args) {
+                  Value input = args[0];
+                  Value hw = b.create<arith::ConstantOp>(
+                      loc, FloatAttr::get(resultElementType,
+                                          inputType.getShape()[2] *
+                                              inputType.getShape()[3]));
+                  Value result = b.create<arith::DivFOp>(loc, input, hw);
+                  b.create<linalg::YieldOp>(loc, result);
+                })
+            .getResult(0);
+
+    indexingMaps = {
+        rewriter.getMultiDimIdentityMap(inputRank), // input
+        ncIndexingMap,                              // mean
+        ncIndexingMap,                              // output
+    };
+
+    iteratorTypes = {
+        utils::IteratorType::parallel,
+        utils::IteratorType::parallel,
+        utils::IteratorType::reduction,
+        utils::IteratorType::reduction,
+    };
+    // (input - mean) ^ 2
+    Value varianceNumerator =
+        rewriter
+            .create<linalg::GenericOp>(
+                loc, varTensor.getType(), ValueRange{input, mean}, varTensor,
+                indexingMaps, iteratorTypes,
+                [&](OpBuilder &b, Location loc, ValueRange args) {
+                  Value input = args[0], mean = args[1], output = args[2];
+                  Value two = b.create<arith::ConstantOp>(
+                      loc, FloatAttr::get(resultElementType, 2));
+                  Value inputSubMean =
+                      b.create<arith::SubFOp>(loc, input, mean);
+                  Value squared =
+                      b.create<math::PowFOp>(loc, inputSubMean, two);
+                  Value sum = b.create<arith::AddFOp>(loc, squared, output);
+                  b.create<linalg::YieldOp>(loc, sum);
+                })
+            .getResult(0);
+
+    indexingMaps = {
+        rewriter.getMultiDimIdentityMap(2), // sumPool2d
+        rewriter.getMultiDimIdentityMap(2), // output
+    };
+
+    iteratorTypes = {
+        utils::IteratorType::parallel,
+        utils::IteratorType::parallel,
+    };
+
+    Value variance =
+        rewriter
+            .create<linalg::GenericOp>(
+                loc, varTensor.getType(), ValueRange{varianceNumerator},
+                varTensor, indexingMaps, iteratorTypes,
+                [&](OpBuilder &b, Location loc, ValueRange args) {
+                  Value numerator = args[0];
+                  Value hw = b.create<arith::ConstantOp>(
+                      loc, FloatAttr::get(resultElementType,
+                                          inputType.getShape()[2] *
+                                              inputType.getShape()[3]));
+                  Value sum = b.create<arith::DivFOp>(loc, numerator, hw);
+                  b.create<linalg::YieldOp>(loc, sum);
+                })
+            .getResult(0);
+
+    iteratorTypes = {
+        utils::IteratorType::parallel,
+        utils::IteratorType::parallel,
+        utils::IteratorType::parallel,
+        utils::IteratorType::parallel,
+    };
+    indexingMaps = {
+        rewriter.getMultiDimIdentityMap(inputRank), // input
+        ncIndexingMap,                              // mean
+        ncIndexingMap,                              // variance
+        cIndexingMap,                               // scale
+        cIndexingMap,                               // bias
+        rewriter.getMultiDimIdentityMap(inputRank), // output
+    };
+
+    Value outTensor =
+        createZeroInitTensor(rewriter, loc, inputSize, resultElementType);
+
+    Value instNorm =
+        rewriter
+            .create<linalg::GenericOp>(
+                loc, outTensor.getType(),
+                ValueRange{input, mean, variance, scale, bias}, outTensor,
+                indexingMaps, iteratorTypes,
+                [&](OpBuilder &b, Location loc, ValueRange args) {
+                  Value input = args[0], mean = args[1], var = args[2],
+                        scale = args[3], bias = args[4];
+                  Value inputSubMean =
+                      b.create<arith::SubFOp>(loc, input, mean);
+                  Value truncatedEps =
+                      b.create<arith::TruncFOp>(loc, var.getType(), eps);
+                  Value varPlusEps =
+                      b.create<arith::AddFOp>(loc, var, truncatedEps);
+                  Value rSTD = b.create<math::RsqrtOp>(loc, varPlusEps);
+                  Value temp = b.create<arith::MulFOp>(loc, inputSubMean, rSTD);
+                  Value timesScale = b.create<arith::MulFOp>(loc, temp, scale);
+                  Value plusBias =
+                      b.create<arith::AddFOp>(loc, timesScale, bias);
+                  b.create<linalg::YieldOp>(loc, plusBias);
+                })
+            .getResult(0);
+    Type newResultType = getTypeConverter()->convertType(op.getType());
+    rewriter.replaceOpWithNewOp<tensor::CastOp>(op, newResultType, instNorm);
+
+    return success();
+  }
+};
+} // namespace
+
 namespace {
 class ConvertAtenNllLossBackwardOp
     : public OpConversionPattern<AtenNllLossBackwardOp> {
@@ -2367,6 +2560,8 @@ void mlir::torch::torch_to_linalg::populateUncategorizedPatternsAndLegality(
   patterns.add<ConvertAtenBatchNormOp>(typeConverter, context);
   target.addIllegalOp<AtenLogitOp>();
   patterns.add<ConvertLogitOp>(typeConverter, context);
+  target.addIllegalOp<AtenInstanceNormOp>();
+  patterns.add<ConvertAtenInstanceNormOp>(typeConverter, context);
   target.addIllegalOp<PrimsCollapseOp>();
   patterns.add<ConvertPrimsCollapseOp>(typeConverter, context);
   target.addIllegalOp<PrimsSplitDimOp>();

lib/Dialect/Torch/Transforms/AbstractInterpLibrary.cpp

@@ -8744,6 +8744,10 @@ StringRef mlir::torch::Torch::getAbstractInterpLibrary() {
 "    %3 = torch.prim.TupleConstruct %0, %1, %2 : !torch.list<int>, !torch.list<int>, !torch.list<int> -> !torch.tuple<list<int>, list<int>, list<int>>\n"
 "    return %3 : !torch.tuple<list<int>, list<int>, list<int>>\n"
 "  }\n"
+"  func.func @\"__torch_mlir_shape_fn.aten.instance_norm\"(%arg0: !torch.list<int>, %arg1: !torch.optional<list<int>>, %arg2: !torch.optional<list<int>>, %arg3: !torch.optional<list<int>>, %arg4: !torch.optional<list<int>>, %arg5: !torch.bool, %arg6: !torch.float, %arg7: !torch.float, %arg8: !torch.bool) -> !torch.list<int> {\n"
+"    %0 = call @__torch__.torch.jit._shape_functions.unary(%arg0) : (!torch.list<int>) -> !torch.list<int>\n"
+"    return %0 : !torch.list<int>\n"
+"  }\n"
 "  func.func @\"__torch_mlir_shape_fn.aten.slice.Tensor\"(%arg0: !torch.list<int>, %arg1: !torch.int, %arg2: !torch.optional<int>, %arg3: !torch.optional<int>, %arg4: !torch.int) -> !torch.list<int> {\n"
 "    %0 = call @__torch__.torch.jit._shape_functions.slice(%arg0, %arg1, %arg2, %arg3, %arg4) : (!torch.list<int>, !torch.int, !torch.optional<int>, !torch.optional<int>, !torch.int) -> !torch.list<int>\n"
 "    return %0 : !torch.list<int>\n"
@@ -9588,6 +9592,10 @@ StringRef mlir::torch::Torch::getAbstractInterpLibrary() {
 "    %3 = torch.prim.TupleConstruct %0#1, %0#1, %0#1 : !torch.int, !torch.int, !torch.int -> !torch.tuple<int, int, int>\n"
 "    return %3 : !torch.tuple<int, int, int>\n"
 "  }\n"
+"  func.func @\"__torch_mlir_dtype_fn.aten.instance_norm\"(%arg0: !torch.tuple<int, int>, %arg1: !torch.optional<tuple<int, int>>, %arg2: !torch.optional<tuple<int, int>>, %arg3: !torch.optional<tuple<int, int>>, %arg4: !torch.optional<tuple<int, int>>, %arg5: !torch.bool, %arg6: !torch.float, %arg7: !torch.float, %arg8: !torch.bool) -> !torch.int {\n"
+"    %0:2 = torch.prim.TupleUnpack %arg0 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
+"    return %0#1 : !torch.int\n"
+"  }\n"
 "  func.func @\"__torch_mlir_dtype_fn.aten.bernoulli_.float\"(%arg0: !torch.tuple<int, int>, %arg1: !torch.float, %arg2: !torch.any) -> !torch.int {\n"
 "    %0:2 = torch.prim.TupleUnpack %arg0 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
 "    return %0#1 : !torch.int\n"

projects/pt1/python/torch_mlir/jit_ir_importer/build_tools/abstract_interp_lib_gen.py

@@ -1388,6 +1388,9 @@ def aten〇group_norm〡shape(input: List[int], num_groups: int, weight: Optiona
 def aten〇native_group_norm〡shape(input: List[int], weight: Optional[List[int]], bias: Optional[List[int]], N: int, C: int, HxW: int, group: int, eps: float) -> Tuple[List[int], List[int], List[int]]:
     return upstream_shape_functions.unary(input), [N, group], [N, group]
 
+def aten〇instance_norm〡shape(input: List[int], weight: Optional[List[int]], bias: Optional[List[int]], running_mean: Optional[List[int]], running_var: Optional[List[int]], use_input_stats: bool, momentum: float, eps: float, cudnn_enabled: bool) -> List[int]:
+    return upstream_shape_functions.unary(input)
+
 def aten〇slice〇Tensor〡shape(self: List[int], dim: int = 0, start: Optional[int] = None, end: Optional[int] = None, step: int = 1) -> List[int]:
     return upstream_shape_functions.slice(self, dim, start, end, step)
 
@@ -2006,6 +2009,11 @@ def aten〇native_group_norm〡dtype(input_rank_dtype: Tuple[int, int], weight_r
     assert not is_integer_dtype(input_dtype)
     return input_dtype, input_dtype, input_dtype
 
+# device is not supported hence unable to check the dtype function
+def aten〇instance_norm〡dtype(input_rank_dtype: Tuple[int, int], weight_rank_dtype: Optional[Tuple[int, int]], bias_rank_dtype: Optional[Tuple[int, int]], running_mean_rank_dtype: Optional[Tuple[int, int]], running_var_rank_dtype: Optional[Tuple[int, int]], use_input_stats: bool, momentum: float, eps: float, cudnn_enabled: bool) -> int:
+    input_rank, input_dtype = input_rank_dtype
+    return input_dtype
+
 @check_dtype_function(_check_tensors_with_the_same_dtype(num_of_tensors=1))
 def aten〇bernoulli_〇float〡dtype(self_rank_dtype: Tuple[int, int], p: float = 0.5, generator: Any = None) -> int:
     self_rank, self_dtype = self_rank_dtype

projects/pt1/python/torch_mlir/jit_ir_importer/build_tools/torch_ods_gen.py

@@ -432,6 +432,9 @@ def emit_with_mutating_variants(key, **kwargs):
     emit(
         "aten::batch_norm : (Tensor, Tensor?, Tensor?, Tensor?, Tensor?, bool, float, float, bool) -> (Tensor)"
     )
+    emit(
+        "aten::instance_norm : (Tensor, Tensor?, Tensor?, Tensor?, Tensor?, bool, float, float, bool) -> (Tensor)"
+    )
     emit(
         "aten::native_group_norm : (Tensor, Tensor?, Tensor?, int, int, int, int, float) -> (Tensor, Tensor, Tensor)"
     )

projects/pt1/python/torch_mlir_e2e_test/test_suite/norm_like.py

@@ -489,3 +489,19 @@ def forward(self, x):
 def LayerNormNormalizeOverAllDimsModule_basic(module, tu: TestUtils):
     module.forward(tu.rand(2, 2, 3))
 
+class InstanceNormModule(torch.nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.inorm = torch.nn.InstanceNorm2d(100)
+
+    @export
+    @annotate_args([
+        None,
+        ([20, 100, 35, 45], torch.float32, True),
+    ])
+    def forward(self, x):
+        return self.inorm(x)
+
+@register_test_case(module_factory=lambda: InstanceNormModule())
+def InstanceNormModule_basic(module, tu: TestUtils):
+    module.forward(tu.rand(20, 100, 35, 45))

test/Conversion/TorchOnnxToTorch/simple_ops_g_to_p.mlir

@@ -555,6 +555,15 @@ func.func @test_globalaveragepool_precomputed(%arg0: !torch.vtensor<[1,1,3,3],f3
 
 // -----
 
+// CHECK-LABEL: func.func @test_instancenorm
+  func.func @test_instancenorm(%arg0: !torch.vtensor<[1,2,1,3],f32>, %arg1: !torch.vtensor<[2],f32>, %arg2: !torch.vtensor<[2],f32>) -> !torch.vtensor<[1,2,1,3],f32> attributes {torch.onnx_meta.ir_version = 3 : si64, torch.onnx_meta.opset_version = 6 : si64, torch.onnx_meta.producer_name = "backend-test", torch.onnx_meta.producer_version = ""} {
+    // CHECK:  torch.aten.instance_norm %arg0, %arg1, %arg2, %none, %none, %false, %float0.000000e00, %float9.999990e-06, %false : !torch.vtensor<[1,2,1,3],f32>, !torch.vtensor<[2],f32>, !torch.vtensor<[2],f32>, !torch.none, !torch.none, !torch.bool, !torch.float, !torch.float, !torch.bool -> !torch.vtensor<[1,2,1,3],f32>
+    %0 = torch.operator "onnx.InstanceNormalization"(%arg0, %arg1, %arg2) : (!torch.vtensor<[1,2,1,3],f32>, !torch.vtensor<[2],f32>, !torch.vtensor<[2],f32>) -> !torch.vtensor<[1,2,1,3],f32>
+    return %0 : !torch.vtensor<[1,2,1,3],f32>
+  }
+
+// -----
+
 // CHECK-LABEL: func.func @test_not_2d
 func.func @test_not_2d(%arg0: !torch.vtensor<[3,4],i1>) -> !torch.vtensor<[3,4],i1> attributes {torch.onnx_meta.ir_version = 3 : si64, torch.onnx_meta.opset_version = 1 : si64, torch.onnx_meta.producer_name = "backend-test", torch.onnx_meta.producer_version = ""} {
     // CHECK: torch.aten.bitwise_not %arg0 : !torch.vtensor<[3,4],i1> -> !torch.vtensor<[3,4],i1>