| { |
| "usageCase": "default", |
| "fixtureArrays": { |
| "onnx_backend_hardsigmoid_input_x": [1.764052391052246, 0.40015721321105957, 0.978738009929657, 2.2408931255340576, 1.8675580024719238, -0.9772778749465942, 0.9500884413719177, -0.15135720372200012, -0.10321885347366333, 0.4105985164642334, 0.14404356479644775, 1.4542734622955322, 0.7610377073287964, 0.12167501449584961, 0.44386324286460876, 0.3336743414402008, 1.4940791130065918, -0.2051582634449005, 0.3130677044391632, -0.8540957570075989, -2.5529897212982178, 0.653618574142456, 0.8644362092018127, -0.7421650290489197, 2.269754648208618, -1.4543657302856445, 0.04575851559638977, -0.18718385696411133, 1.5327792167663574, 1.4693588018417358, 0.154947429895401, 0.37816253304481506, -0.8877857327461243, -1.980796456336975, -0.34791216254234314, 0.15634897351264954, 1.2302906513214111, 1.202379822731018, -0.38732680678367615, -0.302302747964859, -1.0485529899597168, -1.420017957687378, -1.7062702178955078, 1.950775384902954, -0.5096521973609924, -0.4380742907524109, -1.2527953386306763, 0.7774903774261475, -1.6138978004455566, -0.21274028718471527, -0.8954665660858154, 0.38690251111984253, -0.5108051300048828, -1.18063223361969, -0.02818222902715206, 0.4283318817615509, 0.06651721894741058, 0.30247190594673157, -0.6343221068382263, -0.3627411723136902] |
| }, |
| "cases": [ |
| { |
| "name": "default", |
| "inputs": { |
| "x": { "dtype": "float32", "shape": [5], "data": { "kind": "values", "values": [-4.0, -1.0, 0.0, 1.0, 4.0] } } |
| }, |
| "outputs": { "y": { "dtype": "float32", "shape": [5] } } |
| }, |
| { |
| "name": "nan_input_propagates", |
| "inputs": { |
| "x": { "dtype": "float32", "shape": [4], "data": { "kind": "values", "values": ["NaN", -10.0, 0.0, 10.0] } } |
| }, |
| "outputs": { "y": { "dtype": "float32", "shape": [4], "tolerance": 0.000001, "allowNaN": true } } |
| }, |
| { |
| "name": "f32_beta_zero_subnormal_clamp_gpu_gap", |
| "skipGpu": { |
| "category": "permanent", |
| "reason": "Portable WGSL floating-point semantics do not guarantee preservation of the subnormal values required by this fixture. Backend evidence: WebGPU/Metal flushes subnormals to zero (f32 and f16); the kernel cannot preserve denormal inputs/outputs bit-exactly." |
| }, |
| "provenance": { |
| "source": "onnxruntime/test/providers/cpu/activation/activation_op_test.cc", |
| "test": "ActivationOpTest.HardSigmoid", |
| "notes": "With beta=0 and alpha=1, positive subnormal inputs should pass through the lower clamp while negative subnormals clamp to zero." |
| }, |
| "attrs": { "alpha": 1, "beta": 0 }, |
| "inputs": { |
| "x": { "dtype": "float32", "shape": [4], "data": { "kind": "values", "values": [-1e-40, 0.0, 1e-40, 1e-39] } } |
| }, |
| "outputs": { "y": { "dtype": "float32", "shape": [4], "tolerance": 0 } } |
| }, |
| { |
| "name": "f32_beta_zero_subnormal_clamp_scalar_gpu_gap", |
| "skipGpu": { |
| "category": "permanent", |
| "reason": "Portable WGSL floating-point semantics do not guarantee preservation of the subnormal values required by this fixture. Backend evidence: WebGPU/Metal flushes subnormals to zero (f32 and f16); the kernel cannot preserve denormal inputs/outputs bit-exactly." |
| }, |
| "provenance": { |
| "source": "onnxruntime/test/providers/cpu/activation/activation_op_test.cc", |
| "test": "ActivationOpTest.HardSigmoid", |
| "notes": "On an unaligned scalar path with beta=0 and alpha=1, positive subnormals pass through the lower clamp." |
| }, |
| "attrs": { "alpha": 1, "beta": 0 }, |
| "inputs": { |
| "x": { "dtype": "float32", "shape": [3], "data": { "kind": "values", "values": [-1e-40, 0.0, 1e-40] } } |
| }, |
| "outputs": { "y": { "dtype": "float32", "shape": [3], "tolerance": 0 } } |
| }, |
| { |
| "name": "f32_nonfinite_endpoints", |
| "provenance": { |
| "source": "onnxruntime/test/providers/cpu/activation/activation_op_test.cc", |
| "test": "ActivationOpTest.HardSigmoid", |
| "notes": "Additional ORT-validated endpoint coverage for explicit +/-Infinity and NaN behavior." |
| }, |
| "attrs": { "alpha": 0.2, "beta": 0.5 }, |
| "inputs": { |
| "x": { |
| "dtype": "float32", |
| "shape": [8], |
| "data": { "kind": "values", "values": ["-Infinity", -100.0, -3.0, 0.0, 3.0, 100.0, "Infinity", "NaN"] } |
| } |
| }, |
| "outputs": { "y": { "dtype": "float32", "shape": [8], "tolerance": 0.000001, "allowNaN": true } } |
| }, |
| { |
| "name": "onnx_example_nondefault_alpha_beta", |
| "provenance": { "source": "cmake/external/onnx/onnx/backend/test/data/node/test_hardsigmoid_example" }, |
| "attrs": { "alpha": 0.5, "beta": 0.6 }, |
| "inputs": { "x": { "dtype": "float32", "shape": [3], "data": { "kind": "values", "values": [-1.0, 0.0, 1.0] } } }, |
| "outputs": { "y": { "dtype": "float32", "shape": [3], "tolerance": 0.000001 } } |
| }, |
| { |
| "name": "f16_nondefault_clamp_boundaries", |
| "attrs": { "alpha": 0.5, "beta": 0.6 }, |
| "inputs": { |
| "x": { |
| "dtype": "float16", |
| "shape": [6], |
| "data": { "kind": "values", "values": [-4.0, -1.2, -1.0, 0.8, 1.0, 4.0] } |
| } |
| }, |
| "outputs": { "y": { "dtype": "float16", "shape": [6], "tolerance": 0.001 } } |
| }, |
| { |
| "name": "ort_alpha_0_2_beta_0_5_activation_extremes", |
| "provenance": { |
| "source": "onnxruntime/test/providers/cpu/activation/activation_op_test.cc", |
| "test": "ActivationOpTest.HardSigmoid", |
| "notes": "ORT shared activation vector with alpha=0.2 and beta=0.5." |
| }, |
| "attrs": { "alpha": 0.2, "beta": 0.5 }, |
| "inputs": { |
| "x": { |
| "dtype": "float32", |
| "shape": [13], |
| "data": { |
| "kind": "values", |
| "values": [-1.0, 0.0, 1.0, 100.0, -100.0, 1000.0, -1000.0, 1.1754943508222875e-38, 1.1754943508222876e-39, -1.1754943508222876e-39, 3.4028234663852886e+38, -3.4028234663852886e+38, "Infinity"] |
| } |
| } |
| }, |
| "outputs": { "y": { "dtype": "float32", "shape": [13], "tolerance": 0.000001 } } |
| }, |
| { |
| "name": "ort_fp16_alpha_0_2_beta_0_5_activation_extremes", |
| "provenance": { |
| "source": "onnxruntime/test/providers/cpu/activation/activation_op_test.cc", |
| "test": "ActivationOpTest.HardSigmoid", |
| "notes": "Float16 projection of ORT's shared activation vector with alpha=0.2 and beta=0.5." |
| }, |
| "attrs": { "alpha": 0.2, "beta": 0.5 }, |
| "inputs": { |
| "x": { |
| "dtype": "float16", |
| "shape": [13], |
| "data": { |
| "kind": "values", |
| "values": [-1.0, 0.0, 1.0, 100.0, -100.0, 1000.0, -1000.0, 1.1754943508222875e-38, 1.1754943508222876e-39, -1.1754943508222876e-39, 65504.0, -65504.0, "Infinity"] |
| } |
| } |
| }, |
| "outputs": { "y": { "dtype": "float16", "shape": [13], "tolerance": 0.001 } } |
| }, |
| { |
| "name": "onnx_backend_hardsigmoid", |
| "provenance": { "source": "cmake/external/onnx/onnx/backend/test/data/node/test_hardsigmoid" }, |
| "attrs": { "alpha": 0.5, "beta": 0.6000000238418579 }, |
| "inputs": { |
| "x": { |
| "dtype": "float32", |
| "shape": [3, 4, 5], |
| "data": { "kind": "values", "values": { "$ref": "#/fixtureArrays/onnx_backend_hardsigmoid_input_x" } } |
| } |
| }, |
| "outputs": { "y": { "dtype": "float32", "shape": [3, 4, 5], "tolerance": 0.00001 } } |
| }, |
| { |
| "name": "onnx_backend_hardsigmoid_default", |
| "provenance": { "source": "cmake/external/onnx/onnx/backend/test/data/node/test_hardsigmoid_default" }, |
| "inputs": { |
| "x": { |
| "dtype": "float32", |
| "shape": [3, 4, 5], |
| "data": { "kind": "values", "values": { "$ref": "#/fixtureArrays/onnx_backend_hardsigmoid_input_x" } } |
| } |
| }, |
| "outputs": { "y": { "dtype": "float32", "shape": [3, 4, 5], "tolerance": 0.00001 } } |
| }, |
| { |
| "name": "onnx_backend_hardsigmoid_example", |
| "provenance": { "source": "cmake/external/onnx/onnx/backend/test/data/node/test_hardsigmoid_example" }, |
| "attrs": { "alpha": 0.5, "beta": 0.6000000238418579 }, |
| "inputs": { "x": { "dtype": "float32", "shape": [3], "data": { "kind": "values", "values": [-1.0, 0.0, 1.0] } } }, |
| "outputs": { "y": { "dtype": "float32", "shape": [3], "tolerance": 0.00001 } } |
| }, |
| { |
| "name": "vec4_f16_lanes", |
| "inputs": { |
| "x": { |
| "dtype": "float16", |
| "shape": [16], |
| "data": { |
| "kind": "values", |
| "values": [-6.0, -4.0, -3.0, -2.0, -1.5, -1.0, -0.5, -0.25, 0.0, 0.25, 0.5, 1.0, 1.5, 2.0, 4.0, 6.0] |
| } |
| } |
| }, |
| "outputs": { "y": { "dtype": "float16", "shape": [16], "tolerance": 0.001, "relTolerance": 0.002 } } |
| }, |
| { |
| "name": "vec4_f32_nan_passthrough", |
| "inputs": { |
| "x": { "dtype": "float32", "shape": [4], "data": { "kind": "values", "values": ["NaN", -10.0, 0.25, 10.0] } } |
| }, |
| "outputs": { "y": { "dtype": "float32", "shape": [4], "tolerance": 0.000001, "allowNaN": true } } |
| }, |
| { |
| "name": "empty_input_zero_dim", |
| "inputs": { "x": { "dtype": "float32", "shape": [0], "data": { "kind": "values", "values": [] } } }, |
| "outputs": { "y": { "dtype": "float32", "shape": [0], "tolerance": 0 } } |
| }, |
| { |
| "name": "f32_exact_clamp_boundary_values", |
| "attrs": { "alpha": 0.2, "beta": 0.5 }, |
| "inputs": { |
| "x": { |
| "dtype": "float32", |
| "shape": [6], |
| "data": { "kind": "values", "values": [-3.0, -2.5, -2.0, 2.0, 2.5, 3.0] } |
| } |
| }, |
| "outputs": { |
| "y": { |
| "dtype": "float32", |
| "shape": [6], |
| "tolerance": 0, |
| "data": { "kind": "values", "values": [0.0, 0.0, 0.1, 0.9, 1.0, 1.0] } |
| } |
| } |
| }, |
| { |
| "name": "alpha_zero_constant_beta", |
| "provenance": { |
| "notes": "With alpha=0, every finite input produces beta, while an infinite input produces NaN through 0*Infinity. Finite values, infinities, and NaN make the per-lane behavior observable." |
| }, |
| "attrs": { "alpha": 0, "beta": 0.5 }, |
| "inputs": { |
| "x": { |
| "dtype": "float32", |
| "shape": [8], |
| "data": { "kind": "values", "values": [0.0, 1.0, "Infinity", "-Infinity", "NaN", 3.5, -2.25, 0.75] } |
| } |
| }, |
| "outputs": { "y": { "dtype": "float32", "shape": [8], "tolerance": 0, "allowNaN": true } } |
| } |
| ] |
| } |
|
|
