Update supported Op documentation

* Updated the following to show Opset 19 support based on the merged PRs * Cast: onnx#2322 * Constant: onnx#2315 * Equal: onnx#2401 * Pad: onnx#2482 * QuantizeLinear: onnx#2480 * Resize: onnx#2490 * Slice: There is no Opset 19 Slice so I'm unsure why 19 was marked unsupported. * Split There is no Opset 19 Split so I'm unsure why 19 was marked unsupported. * Updated the following to show as unsupported for Opset 20 by onnx-mlir * ConstantOfShape adds support for bfloat16, * ReduceMin adds support for bool * ReduceMax adds support for bool * DFT promotes axis as input * The following were also upated in Opset 20 but not marked as unsupported as they wil be covered in onnx-mlir before 0.4.2 * IsInf: onnx#2721 * IsNaN: onnx#2721 Signed-off-by: Charles Volzka <[email protected]>
cjvolzka · Feb 20, 2024 · 940a71c · 940a71c
1 parent 0124472
commit 940a71c
Show file tree

Hide file tree

Showing 3 changed files with 21 additions and 24 deletions.
diff --git a/docs/SupportedONNXOps-NNPA.md b/docs/SupportedONNXOps-NNPA.md
@@ -3,11 +3,11 @@
 
 # Supported ONNX Operation for Target *NNPA*.
 
-Onnx-mlir currently supports ONNX operations targeting up to opset 19. Limitations are listed when applicable. This documentation highlights the minimum and maximum opset versions that are fully supported by onnx-mlir and not the version changes.
+Onnx-mlir currently supports ONNX operations targeting up to opset 20. Limitations are listed when applicable. This documentation highlights the minimum and maximum opset versions that are fully supported by onnx-mlir and not the version changes.
 
 * Operations are defined by the [ONNX Standard](https://github.com/onnx/onnx/blob/main/docs/Operators.md).
-* **Supported Opsets** indicates the lowest and highest opset a model may have for onnx-mlir to support compiling a model with the operator. 
-   * A * indicates onnx-mlir is compatible with the latest version of that operator available as of opset 19.
+* **Supported Opsets** indicates the lowest and highest opset a model may have for onnx-mlir to support compiling a model with the operator.
+   * A * indicates onnx-mlir is compatible with the latest version of that operator available as of opset 20.
 
 
 NNPA has hardware limitations in dimension index size and tensor size, which are described in [NNPALimit.h](../src/Accelerators/NNPA/Support/NNPALimit.h). They are large enough for normal use cases, but if your model exceeds the limitations, CPU is used instead of NNPA.

diff --git a/docs/SupportedONNXOps-cpu.md b/docs/SupportedONNXOps-cpu.md
@@ -6,7 +6,7 @@
 Onnx-mlir currently supports ONNX operations targeting up to opset 20. Limitations are listed when applicable. This documentation highlights the minimum and maximum opset versions that are fully supported by onnx-mlir and not the version changes.
 
 * Operations are defined by the [ONNX Standard](https://github.com/onnx/onnx/blob/main/docs/Operators.md).
-* **Supported Opsets** indicates the lowest and highest opset a model may have for onnx-mlir to support compiling a model with the operator. 
+* **Supported Opsets** indicates the lowest and highest opset a model may have for onnx-mlir to support compiling a model with the operator.
    * A * indicates onnx-mlir is compatible with the latest version of that operator available as of opset 20.
 
 
@@ -36,7 +36,7 @@ Onnx-mlir currently supports ONNX operations targeting up to opset 20. Limitatio
 | **BitwiseOr** |18 - * | | |
 | **BitwiseXor** |18 - * | | |
 | **BlackmanWindow** |none | | | |
-| **Cast** |6 - 18 |Cast only between float and double types. Only ppc64le and MacOS platforms support float16. | |
+| **Cast** |6 - * |Cast only between float and double types. Only ppc64le and MacOS platforms support float16. | |
 | **CastLike** |19 - * |CastLike only between float and double types. Only ppc64le and MacOS platforms support float16. | |
 | **CastMap** |none | | | |
 | **CategoryMapper** |none | | | |
@@ -48,15 +48,15 @@ Onnx-mlir currently supports ONNX operations targeting up to opset 20. Limitatio
 | **Compress** |9 - * | | |
 | **Concat** |6 - * | | |
 | **ConcatFromSequence** |none | | | |
-| **Constant** |6 - 18 | | |
-| **ConstantOfShape** |9 - * | | |
+| **Constant** |6 - * | | |
+| **ConstantOfShape** |9 - 19 | | |
 | **Conv** |6 - * | | |
 | **ConvInteger** |none | | | |
 | **ConvTranspose** |6 - * |Unknown dimension in spatial dimensions (such as H and W) not supported. | |
 | **Cos** |7 - * | | |
 | **Cosh** |9 - * | | |
 | **CumSum** |11 - * | | |
-| **DFT** |none | | | |
+| **DFT** |17 - 19 | | |
 | **DeformConv** |none | | | |
 | **DepthToSpace** |13 - * | | |
 | **DequantizeLinear** |10 - * |Only support for per-tensor or layer dequantization. No support for per-axis dequantization. | |
@@ -67,7 +67,7 @@ Onnx-mlir currently supports ONNX operations targeting up to opset 20. Limitatio
 | **DynamicQuantizeLinear** |11 - * | | |
 | **Einsum** |12 - * |Limited to the types supported by ReduceSum and MatMul (which we decompose to in most cases) which exclude integers with width < 32. | |
 | **Elu** |6 - * | | |
-| **Equal** |7 - 18 | | |
+| **Equal** |7 - * | | |
 | **Erf** |9 - * | | |
 | **Exp** |6 - * | | |
 | **Expand** |8 - * | | |
@@ -142,11 +142,11 @@ Onnx-mlir currently supports ONNX operations targeting up to opset 20. Limitatio
 | **OptionalHasElement** |none | | | |
 | **Or** |7 - * | | |
 | **PRelu** |6 - * | | |
-| **Pad** |6 - 18 |axes input not supported. | |
+| **Pad** |6 - * |axes input not supported. | |
 | **Pow** |7 - * |No support for power with integer types. | |
 | **QLinearConv** |none | | | |
 | **QLinearMatMul** |none | | | |
-| **QuantizeLinear** |10 - 18 |Do not support per-axis and i8 quantization. | |
+| **QuantizeLinear** |10 - * |Do not support per-axis and i8 quantization. | |
 | **RNN** |7 - * | | |
 | **RandomNormal** |none | | | |
 | **RandomNormalLike** |none | | | |
@@ -158,15 +158,15 @@ Onnx-mlir currently supports ONNX operations targeting up to opset 20. Limitatio
 | **ReduceL2** |13 - * |do_not_keep_dim not supported. | |
 | **ReduceLogSum** |13 - * |do_not_keep_dim not supported. | |
 | **ReduceLogSumExp** |13 - * |do_not_keep_dim not supported. | |
-| **ReduceMax** |6 - * |do_not_keep_dim not supported. | |
+| **ReduceMax** |6 - 19 |do_not_keep_dim not supported. | |
 | **ReduceMean** |6 - * |do_not_keep_dim not supported. | |
-| **ReduceMin** |6 - * |do_not_keep_dim not supported. | |
+| **ReduceMin** |6 - 19 |do_not_keep_dim not supported. | |
 | **ReduceProd** |13 - * |do_not_keep_dim not supported. | |
 | **ReduceSum** |6 - * |Default axis and do_not_keep_dim not supported. |Default axis and do_not_keep_dim temporarily removed due to changes in onnx 1.8.1. |
 | **ReduceSumSquare** |13 - * |Default axis and do_not_keep_dim not supported. | |
 | **Relu** |6 - * | | |
 | **Reshape** |6 - * |allowzero not supported. | |
-| **Resize** |10 - 18 |Missing support for linear, cubic, crop, pytorch_half_pixel, and floor. Attributes antialias, axes and keep_aspect_ratio_policy are not supported. | |
+| **Resize** |10 - * |Missing support for linear, cubic, crop, pytorch_half_pixel, and floor. Attributes antialias, axes and keep_aspect_ratio_policy are not supported. | |
 | **ReverseSequence** |10 - * | | |
 | **RoiAlign** |none | | | |
 | **Round** |11 - * | | |
@@ -193,13 +193,13 @@ Onnx-mlir currently supports ONNX operations targeting up to opset 20. Limitatio
 | **Sin** |7 - * | | |
 | **Sinh** |9 - * | | |
 | **Size** |13 - * | | |
-| **Slice** |13 - 18 |Axis must be a constant argument. |Add tests to slices, currently have none. |
+| **Slice** |13 - * |Axis must be a constant argument. |Add tests to slices, currently have none. |
 | **Softmax** |6 - * | | |
 | **SoftmaxCrossEntropyLoss** |none | | | |
 | **Softplus** |6 - * | | |
 | **Softsign** |6 - * | | |
 | **SpaceToDepth** |13 - * | |Example works, the other is imprecise. To investigate. |
-| **Split** |6 - 18 |Does not support static and dynamic shape, zero size splits. |Temporally removed due to changes in onnx 1.8.1. |
+| **Split** |6 - * |Does not support static and dynamic shape, zero size splits. |Temporally removed due to changes in onnx 1.8.1. |
 | **SplitToSequence** |none | | | |
 | **Sqrt** |6 - * | | |
 | **Squeeze** |6 - * |Does not support static and dynamic shape. |Temporally removed due to changes in onnx 1.8.1. |

diff --git a/test/backend/inference_backend.py b/test/backend/inference_backend.py
@@ -394,7 +394,6 @@ def get_test_models():
         },
         # ==OP== Cast
         # ==MIN== 6
-        # ==UNSUPPORTED== 19
         # ==LIM== Cast only between float and double types. Only ppc64le and MacOS platforms support float16.
         "test_cast_FLOAT_to_DOUBLE_cpu": {
             STATIC_SHAPE: {},
@@ -615,11 +614,11 @@ def get_test_models():
         },
         # ==OP== Constant
         # ==MIN== 1
-        # ==UNSUPPORTED== 19
         # By def, no dynamic shapes.
         "test_constant_cpu": {STATIC_SHAPE: {}},
         # ==OP== ConstantOfShape
         # ==MIN== 9
+        # ==UNSUPPORTED== 20
         # By def, no dynamic shapes.
         "test_constantofshape_float_ones_cpu": {STATIC_SHAPE: {}},
         "test_constantofshape_int_zeros_cpu": {STATIC_SHAPE: {}},
@@ -769,6 +768,8 @@ def get_test_models():
             CONSTANT_INPUT: {-1},
         },
         # ==OP== DFT
+        # ==MIN== 17
+        # ==UNSUPPORTED== 20
         # "test_dft_axis_cpu": {STATIC_SHAPE:{}, DYNAMIC_SHAPE:{-1:{-1}}, CONSTANT_INPUT:{-1}},
         # "test_dft_cpu": {STATIC_SHAPE:{}, DYNAMIC_SHAPE:{-1:{-1}}, CONSTANT_INPUT:{-1}},
         # "test_dft_inverse_cpu": {STATIC_SHAPE:{}, DYNAMIC_SHAPE:{-1:{-1}}, CONSTANT_INPUT:{-1}},
@@ -917,7 +918,6 @@ def get_test_models():
         },
         # ==OP== Equal
         # ==MIN== 7
-        # ==UNSUPPORTED== 19
         "test_equal_cpu": {
             STATIC_SHAPE: {},
             DYNAMIC_SHAPE: {-1: {-1}},
@@ -2244,7 +2244,6 @@ def get_test_models():
         },
         # ==OP== Pad
         # ==MIN== 2
-        # ==UNSUPPORTED== 19
         # ==LIM== axes input not supported
         "test_constant_pad_cpu": {
             STATIC_SHAPE: {},
@@ -2298,7 +2297,6 @@ def get_test_models():
         },
         # ==OP== QuantizeLinear
         # ==MIN== 10
-        # ==UNSUPPORTED== 19
         # ==LIM== Do not support per-axis and i8 quantization.
         # "test_quantizelinear_axis_cpu": {STATIC_SHAPE:{}, DYNAMIC_SHAPE:{-1:{-1}}, CONSTANT_INPUT:{-1}},
         "test_quantizelinear_cpu": {
@@ -2430,6 +2428,7 @@ def get_test_models():
         },
         # ==OP== ReduceMax
         # ==MIN== 1
+        # ==UNSUPPORTED== 20
         # ==LIM== do_not_keep_dim not supported.
         # "test_reduce_max_default_axes_keepdim_example_cpu": {STATIC_SHAPE:{}, DYNAMIC_SHAPE:{-1:{-1}}, CONSTANT_INPUT:{-1}},
         # "test_reduce_max_default_axes_keepdims_random_cpu": {STATIC_SHAPE:{}, DYNAMIC_SHAPE:{-1:{-1}}, CONSTANT_INPUT:{-1}},
@@ -2484,6 +2483,7 @@ def get_test_models():
         },
         # ==OP== ReduceMin
         # ==MIN== 1
+        # ==UNSUPPORTED== 20
         # ==LIM== do_not_keep_dim not supported.
         # "test_reduce_min_default_axes_keepdims_example_cpu": {STATIC_SHAPE:{}, DYNAMIC_SHAPE:{-1:{-1}}, CONSTANT_INPUT:{-1}},
         # "test_reduce_min_default_axes_keepdims_random_cpu": {STATIC_SHAPE:{}, DYNAMIC_SHAPE:{-1:{-1}}, CONSTANT_INPUT:{-1}},
@@ -2653,7 +2653,6 @@ def get_test_models():
         },
         # ==OP== Resize
         # ==MIN== 10
-        # ==UNSUPPORTED== 19
         # ==LIM== Missing support for linear, cubic, crop, pytorch_half_pixel, and floor. Attributes antialias, axes and keep_aspect_ratio_policy are not supported.
         # Resize
         # All test cases in onnx v1.11.0. yes for currently supported
@@ -2908,7 +2907,6 @@ def get_test_models():
         },
         # ==OP== Slice
         # ==MIN== 13
-        # ==UNSUPPORTED== 19
         # ==LIM== Axis must be a constant argument.
         # ==TODO== Add tests to slices, currently have none.
         # (makes Axis a runtime argument, which is not supported).
@@ -2992,7 +2990,6 @@ def get_test_models():
         },
         # ==OP== Split
         # ==MIN== 2
-        # ==UNSUPPORTED== 19
         # ==LIM== Does not support static and dynamic shape, zero size splits.
         # ==TODO== Temporally removed due to changes in onnx 1.8.1
         # "test_split_equal_parts_1d_cpu": {STATIC_SHAPE:{}, DYNAMIC_SHAPE:{-1:{-1}}, CONSTANT_INPUT:{-1}},