Add FP8 ONNX tests

test/onnx/add_fp8_test.onnx

@@ -0,0 +1,16 @@
+	add_fp8_test:Q
+
+
0
+
1
2"Addadd_fp8_testZ
+
0
+
+
+
Z
+
1
+
+
+
b
+
2
+
+
+
B

test/onnx/conv_1d_fp8_test.onnx

@@ -0,0 +1,19 @@
+	conv_1d_fp8_test:n
+
+
0
+
1
2"Convconv_1d_fp8_testZ
+
0
+
+

+
+Z
+
1
+
+

+
+b
+
2
+
+

+

+B

test/onnx/cos_fp8_test.onnx

@@ -0,0 +1,13 @@
+	cos_fp8_test:=
+
+
x
y"Coscos_fp8_testZ
+
x
+
+
+
+b
+
y
+
+
+
+B

test/onnx/div_fp8_test.onnx

@@ -0,0 +1,16 @@
+	div_fp8_test:a
+
+
0
+
1out"Divdiv_fp8_testZ
+
0
+
+
+Z
+
1
+
+
+b
+out
+
+
+B

test/onnx/gen_onnx.py

@@ -124,6 +124,21 @@ def add_fp16_test():
         ])
 
 
+@onnx_test()
+def add_fp8_test():
+    x = helper.make_tensor_value_info('0', TensorProto.FLOAT8E4M3FNUZ, [1])
+    y = helper.make_tensor_value_info('1', TensorProto.FLOAT8E4M3FNUZ, [1])
+    z = helper.make_tensor_value_info('2', TensorProto.FLOAT8E4M3FNUZ, [1])
+
+    node = onnx.helper.make_node(
+        'Add',
+        inputs=['0', '1'],
+        outputs=['2'],
+    )
+
+    return ([node], [x, y], [z])
+
+
 @onnx_test()
 def add_scalar_test():
     x = helper.make_tensor_value_info('0', TensorProto.UINT8, [2, 3, 4, 5])
@@ -618,6 +633,42 @@ def binary_dyn_brcst_mul_test():
     return ([node], [arg0, arg1], [arg_out])
 
 
+@onnx_test()
+def binary_dyn_brcst_mul_fp8_test():
+    arg0 = helper.make_tensor_value_info('0', TensorProto.FLOAT8E4M3FNUZ,
+                                         [None, 3, 4, 5])
+    arg1 = helper.make_tensor_value_info('1', TensorProto.FLOAT8E4M3FNUZ,
+                                         [4, 1])
+    arg_out = helper.make_tensor_value_info('out', TensorProto.FLOAT8E4M3FNUZ,
+                                            [None, 3, 4, 5])
+
+    node = onnx.helper.make_node(
+        'Mul',
+        inputs=['0', '1'],
+        outputs=['out'],
+    )
+
+    return ([node], [arg0, arg1], [arg_out])
+
+
+@onnx_test()
+def div_fp8_test():
+    arg0 = helper.make_tensor_value_info('0', TensorProto.FLOAT8E4M3FNUZ,
+                                         [2, 3])
+    arg1 = helper.make_tensor_value_info('1', TensorProto.FLOAT8E4M3FNUZ,
+                                         [2, 3])
+    arg_out = helper.make_tensor_value_info('out', TensorProto.FLOAT8E4M3FNUZ,
+                                            [2, 3])
+
+    node = onnx.helper.make_node(
+        'Div',
+        inputs=['0', '1'],
+        outputs=['out'],
+    )
+
+    return ([node], [arg0, arg1], [arg_out])
+
+
 @onnx_test()
 def cast_test():
     x = helper.make_tensor_value_info('x', TensorProto.FLOAT16, [10])
@@ -1252,6 +1303,20 @@ def conv_3d_test():
     return ([node], [x, y], [out])
 
 
+@onnx_test()
+def conv_1d_fp8_test():
+    x = helper.make_tensor_value_info('0', TensorProto.FLOAT8E4M3FNUZ,
+                                      [1, 3, 5])
+    y = helper.make_tensor_value_info('1', TensorProto.FLOAT8E4M3FNUZ,
+                                      [1, 3, 3])
+    out = helper.make_tensor_value_info('2', TensorProto.FLOAT8E4M3FNUZ,
+                                        [1, 1, 3])
+
+    node = onnx.helper.make_node('Conv', inputs=['0', '1'], outputs=['2'])
+
+    return ([node], [x, y], [out])
+
+
 @onnx_test()
 def conv_attr_fail_test():
     x = helper.make_tensor_value_info('0', TensorProto.FLOAT, [1, 3, 5])
@@ -1688,6 +1753,20 @@ def cos_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def cos_fp8_test():
+    x = helper.make_tensor_value_info('x', TensorProto.FLOAT8E4M3FNUZ, [10])
+    y = helper.make_tensor_value_info('y', TensorProto.FLOAT8E4M3FNUZ, [10])
+
+    node = onnx.helper.make_node(
+        'Cos',
+        inputs=['x'],
+        outputs=['y'],
+    )
+
+    return ([node], [x], [y])
+
+
 @onnx_test()
 def cosh_test():
     x = helper.make_tensor_value_info('x', TensorProto.FLOAT, [1])
@@ -3626,6 +3705,23 @@ def gemm_half_test():
     return ([node], [A, B, C], [Y])
 
 
+@onnx_test()
+def gemm_fp8_test():
+    A = helper.make_tensor_value_info('A', TensorProto.FLOAT8E4M3FNUZ, [8, 6])
+    B = helper.make_tensor_value_info('B', TensorProto.FLOAT8E4M3FNUZ, [8, 7])
+    C = helper.make_tensor_value_info('C', TensorProto.FLOAT8E4M3FNUZ, [6, 1])
+    Y = helper.make_tensor_value_info('Y', TensorProto.FLOAT8E4M3FNUZ, [6, 7])
+
+    node = onnx.helper.make_node('Gemm',
+                                 inputs=['A', 'B', 'C'],
+                                 outputs=['Y'],
+                                 alpha=0.5,
+                                 beta=0.8,
+                                 transA=1)
+
+    return ([node], [A, B, C], [Y])
+
+
 @onnx_test()
 def gemm_dyn_inner_test():
     A = helper.make_tensor_value_info('A', TensorProto.FLOAT, [None, 6])
@@ -3705,6 +3801,22 @@ def globalavgpool_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def globalavgpool_fp8_test():
+    x = helper.make_tensor_value_info('0', TensorProto.FLOAT8E4M3FNUZ,
+                                      [1, 3, 16, 16])
+    y = helper.make_tensor_value_info('1', TensorProto.FLOAT8E4M3FNUZ,
+                                      [1, 3, 1, 1])
+
+    node = onnx.helper.make_node(
+        'GlobalAveragePool',
+        inputs=['0'],
+        outputs=['1'],
+    )
+
+    return ([node], [x], [y])
+
+
 @onnx_test()
 def globalavgpool_dyn_test():
     x = helper.make_tensor_value_info('0', TensorProto.FLOAT,
@@ -3763,6 +3875,22 @@ def globalmaxpool_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def globalmaxpool_fp8_test():
+    x = helper.make_tensor_value_info('0', TensorProto.FLOAT8E4M3FNUZ,
+                                      [1, 3, 16, 16])
+    y = helper.make_tensor_value_info('1', TensorProto.FLOAT8E4M3FNUZ,
+                                      [1, 3, 1, 1])
+
+    node = onnx.helper.make_node(
+        'GlobalMaxPool',
+        inputs=['0'],
+        outputs=['1'],
+    )
+
+    return ([node], [x], [y])
+
+
 @onnx_test()
 def globalmaxpool_dyn_test():
     x = helper.make_tensor_value_info('0', TensorProto.FLOAT,
@@ -8610,6 +8738,24 @@ def reducemax_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def reducemax_fp8_test():
+    x = helper.make_tensor_value_info('x', TensorProto.FLOAT8E4M3FNUZ,
+                                      [3, 4, 5, 6])
+    y = helper.make_tensor_value_info('y', TensorProto.FLOAT8E4M3FNUZ,
+                                      [3, 4, 6])
+
+    axes = [2]
+
+    node = onnx.helper.make_node('ReduceMax',
+                                 inputs=['x'],
+                                 outputs=['y'],
+                                 axes=axes,
+                                 keepdims=0)
+
+    return ([node], [x], [y])
+
+
 @onnx_test
 def reducemax_dyn_test():
     x = helper.make_tensor_value_info('x', TensorProto.FLOAT, [None, 4, 5, 6])
@@ -8699,6 +8845,22 @@ def reducesum_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def reducesum_fp8_test():
+    x = helper.make_tensor_value_info('x', TensorProto.FLOAT8E4M3FNUZ,
+                                      [3, 4, 5, 6])
+    y = helper.make_tensor_value_info('y', TensorProto.FLOAT8E4M3FNUZ,
+                                      [3, 4, 1, 6])
+
+    node = onnx.helper.make_node('ReduceSum',
+                                 inputs=['x'],
+                                 outputs=['y'],
+                                 axes=[2],
+                                 keepdims=0)
+
+    return ([node], [x], [y])
+
+
 @onnx_test()
 def reducesum_empty_axes_test():
     x = helper.make_tensor_value_info('x', TensorProto.FLOAT, [3, 4, 5, 6])
@@ -9962,6 +10124,22 @@ def shrink_int8_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def shrink_fp8_test():
+    x = helper.make_tensor_value_info('x', TensorProto.FLOAT8E4M3FNUZ, [3, 3])
+    y = helper.make_tensor_value_info('y', TensorProto.FLOAT8E4M3FNUZ, [3, 3])
+
+    node = onnx.helper.make_node(
+        "Shrink",
+        inputs=["x"],
+        outputs=["y"],
+        lambd=1.5,
+        bias=1.5,
+    )
+
+    return ([node], [x], [y])
+
+
 @onnx_test()
 def shrink_uint8_test():
     x = helper.make_tensor_value_info('x', TensorProto.UINT8, [3, 3])
@@ -10006,6 +10184,20 @@ def sin_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def sin_fp8_test():
+    x = helper.make_tensor_value_info('x', TensorProto.FLOAT8E4M3FNUZ, [10])
+    y = helper.make_tensor_value_info('y', TensorProto.FLOAT8E4M3FNUZ, [10])
+
+    node = onnx.helper.make_node(
+        'Sin',
+        inputs=['x'],
+        outputs=['y'],
+    )
+
+    return ([node], [x], [y])
+
+
 @onnx_test()
 def sinh_test():
     x = helper.make_tensor_value_info('x', TensorProto.FLOAT, [10])
@@ -10070,6 +10262,19 @@ def size_int_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def size_fp8_test():
+    x = helper.make_tensor_value_info('x', TensorProto.FLOAT8E4M3FNUZ,
+                                      [2, 5, 3])
+    y = helper.make_tensor_value_info('y', TensorProto.INT64, [1])
+    node = onnx.helper.make_node(
+        'Size',
+        inputs=['x'],
+        outputs=['y'],
+    )
+    return ([node], [x], [y])
+
+
 @onnx_test()
 def size_verify_test():
     x = helper.make_tensor_value_info('x', TensorProto.FLOAT, [2, 5, 3])
@@ -10913,6 +11118,22 @@ def sqrt_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def sqrt_fp8_test():
+    x = helper.make_tensor_value_info('x', TensorProto.FLOAT8E4M3FNUZ,
+                                      [10, 15])
+    y = helper.make_tensor_value_info('y', TensorProto.FLOAT8E4M3FNUZ,
+                                      [10, 15])
+
+    node = onnx.helper.make_node(
+        'Sqrt',
+        inputs=['x'],
+        outputs=['y'],
+    )
+
+    return ([node], [x], [y])
+
+
 @onnx_test()
 def squeeze_axes_input_test():
     x = helper.make_tensor_value_info('x', TensorProto.FLOAT, [3, 1, 5, 1])