perfect unittest for pir (PaddlePaddle#61076)

paddle/fluid/pir/dialect/op_generator/op_build_gen.py

@@ -26,6 +26,7 @@
     'FusedConv2dAddActInferMeta',
     'InterpolateInferMeta',
     'DeformableConvInferMeta',
+    'MatrixNMSInferMeta',
 }
 
 _PREPARE_DATA_WITH_VECTOR_INT64_MTTABLE_ATTRIBUTE = {'FrobeniusNormOp'}

python/paddle/tensor/math.py

@@ -2154,49 +2154,50 @@ def mm(input, mat2, name=None):
 
 
     """
-    if in_dynamic_or_pir_mode():
+    if in_dynamic_mode():
         return _C_ops.matmul(input, mat2, False, False)
-    else:
 
-        def __check_input(x, y):
-            var_names = {'x': x, 'y': y}
-            for name, val in var_names.items():
-                check_variable_and_dtype(
-                    val, name, ['float16', 'float32', 'float64'], 'mm'
+    def __check_input(x, y):
+        var_names = {'x': x, 'y': y}
+        for name, val in var_names.items():
+            check_variable_and_dtype(
+                val, name, ['float16', 'float32', 'float64'], 'mm'
+            )
+        x_shape = list(x.shape)
+        y_shape = list(y.shape)
+        if len(x_shape) == 1:
+            x_shape = [1] + x_shape
+        if len(y_shape) == 1:
+            y_shape = y_shape + [1]
+
+        # check the inner 2 dimensions
+        if x_shape[-1] != y_shape[-2]:
+            if not ((x_shape[-1] == -1) or (y_shape[-2] == -1)):
+                raise ValueError(
+                    "After performing an optional transpose, Input X's width should be "
+                    "equal to Y's width for multiplication "
+                    "prerequisites. But received X's shape: {}, Y's shape: {}\n".format(
+                        x_shape, y_shape
+                    )
                 )
-            x_shape = list(x.shape)
-            y_shape = list(y.shape)
-            if len(x_shape) == 1:
-                x_shape = [1] + x_shape
-            if len(y_shape) == 1:
-                y_shape = y_shape + [1]
 
-            # check the inner 2 dimensions
-            if x_shape[-1] != y_shape[-2]:
-                if not ((x_shape[-1] == -1) or (y_shape[-2] == -1)):
+        if len(y_shape) > 2 and len(x_shape) > 2:
+            for i, dim_x in enumerate(x_shape[:-2]):
+                # don't check neg shape
+                if dim_x < 0 or y_shape[i] < 0:
+                    continue
+                if dim_x != y_shape[i]:
                     raise ValueError(
-                        "After performing an optional transpose, Input X's width should be "
-                        "equal to Y's width for multiplication "
-                        "prerequisites. But received X's shape: {}, Y's shape: {}\n".format(
-                            x_shape, y_shape
-                        )
+                        "When the matrix is larger than 2 dimensions, the higher "
+                        "dimensional values of the two matrices need to be equal. "
+                        "But received x_shape[%d] != y_shape[%d]. X's shape: %s, "
+                        "Y's shape: %s.\n" % (i, i, x_shape, y_shape)
                     )
 
-            if len(y_shape) > 2 and len(x_shape) > 2:
-                for i, dim_x in enumerate(x_shape[:-2]):
-                    # don't check neg shape
-                    if dim_x < 0 or y_shape[i] < 0:
-                        continue
-                    if dim_x != y_shape[i]:
-                        raise ValueError(
-                            "When the matrix is larger than 2 dimensions, the higher "
-                            "dimensional values of the two matrices need to be equal. "
-                            "But received x_shape[%d] != y_shape[%d]. X's shape: %s, "
-                            "Y's shape: %s.\n" % (i, i, x_shape, y_shape)
-                        )
-
-        __check_input(input, mat2)
-
+    __check_input(input, mat2)
+    if in_pir_mode():
+        return _C_ops.matmul(input, mat2, False, False)
+    else:
         helper = LayerHelper('mm', **locals())
         out = helper.create_variable_for_type_inference(dtype=input.dtype)
         helper.append_op(
@@ -2514,33 +2515,33 @@ def inner(x, y, name=None):
         nx = x.reshape((-1, xshape[-1]))
         ny = y.reshape((-1, yshape[-1]))
 
+        def __check_input(x, y):
+            var_names = {'x': x, 'y': y}
+            for name, val in var_names.items():
+                check_variable_and_dtype(
+                    val, name, ['float16', 'float32', 'float64'], 'inner'
+                )
+            x_shape = list(xshape)
+            y_shape = list(yshape)
+
+            # check the inner 2 dimensions
+            if x_shape[-1] != y_shape[-1]:
+                if not ((x_shape[-1] == -1) or (y_shape[-1] == -1)):
+                    raise ValueError(
+                        "After performing an optional transpose, Input X's last dim should be "
+                        "equal to Y's last dim for multiplication "
+                        "prerequisites. But received X's shape: {}, Y's shape: {}\n".format(
+                            x_shape, y_shape
+                        )
+                    )
+
+        __check_input(nx, ny)
+
         if in_dynamic_or_pir_mode():
             return _C_ops.matmul(
                 nx, paddle.transpose(ny, [1, 0]), False, False
             ).reshape(dstshape)
         else:
-
-            def __check_input(x, y):
-                var_names = {'x': x, 'y': y}
-                for name, val in var_names.items():
-                    check_variable_and_dtype(
-                        val, name, ['float16', 'float32', 'float64'], 'inner'
-                    )
-                x_shape = list(xshape)
-                y_shape = list(yshape)
-
-                # check the inner 2 dimensions
-                if x_shape[-1] != y_shape[-1]:
-                    if not ((x_shape[-1] == -1) or (y_shape[-1] == -1)):
-                        raise ValueError(
-                            "After performing an optional transpose, Input X's last dim should be "
-                            "equal to Y's last dim for multiplication "
-                            "prerequisites. But received X's shape: {}, Y's shape: {}\n".format(
-                                x_shape, y_shape
-                            )
-                        )
-
-            __check_input(nx, ny)
             helper = LayerHelper('inner', **locals())
             out = helper.create_variable_for_type_inference(dtype=nx.dtype)
             helper.append_op(
@@ -2584,22 +2585,23 @@ def outer(x, y, name=None):
     nx = x.reshape((-1, 1))
     ny = y.reshape((1, -1))
 
-    if in_dynamic_or_pir_mode():
+    if in_dynamic_mode():
         return _C_ops.matmul(nx, ny, False, False)
-    else:
-
-        def __check_input(x, y):
-            var_names = {'x': x, 'y': y}
-            for name, val in var_names.items():
-                check_variable_and_dtype(
-                    val,
-                    name,
-                    ['float16', 'float32', 'float64', 'int32', 'int64'],
-                    'outer',
-                )
 
-        __check_input(nx, ny)
+    def __check_input(x, y):
+        var_names = {'x': x, 'y': y}
+        for name, val in var_names.items():
+            check_variable_and_dtype(
+                val,
+                name,
+                ['float16', 'float32', 'float64', 'int32', 'int64'],
+                'outer',
+            )
 
+    __check_input(nx, ny)
+    if in_pir_mode():
+        return _C_ops.matmul(nx, ny, False, False)
+    else:
         helper = LayerHelper('outer', **locals())
         out = helper.create_variable_for_type_inference(dtype=nx.dtype)
         helper.append_op(

test/legacy_test/test_activation_op.py

@@ -692,6 +692,7 @@ def test_dygraph_api(self):
             np.testing.assert_allclose(out_ref, r.numpy(), rtol=1e-05)
         paddle.enable_static()
 
+    @test_with_pir_api
     def test_errors(self):
         with static_guard():
             with paddle.static.program_guard(paddle.static.Program()):
@@ -4910,7 +4911,7 @@ def test_check_grad(self):
 create_test_act_fp16_class(
     TestSilu, check_prim=True, enable_cinn=True, check_prim_pir=True
 )
-create_test_act_fp16_class(TestLogSigmoid)
+create_test_act_fp16_class(TestLogSigmoid, check_pir=True)
 create_test_act_fp16_class(
     TestTanh, check_prim=True, check_prim_pir=True, enable_cinn=True
 )
@@ -5100,7 +5101,7 @@ def test_check_grad(self):
     TestSigmoid, check_prim=True, check_pir=True, check_prim_pir=True
 )
 create_test_act_bf16_class(TestSilu, check_prim=True, check_prim_pir=True)
-create_test_act_bf16_class(TestLogSigmoid)
+create_test_act_bf16_class(TestLogSigmoid, check_pir=True)
 create_test_act_bf16_class(TestTanh, check_prim=True, check_prim_pir=True)
 create_test_act_bf16_class(TestTanhshrink, check_pir=True)
 create_test_act_bf16_class(TestHardShrink, check_pir=True)

test/legacy_test/test_inner.py

@@ -124,6 +124,7 @@ def test_multiply_dynamic_case5(self):
 
 
 class TestMultiplyError(unittest.TestCase):
+    @test_with_pir_api
     def test_errors_static_case1(self):
         # test static computation graph: dtype can not be int8
         paddle.enable_static()
@@ -134,6 +135,7 @@ def test_errors_static_case1(self):
             y = paddle.static.data(name='y', shape=[100], dtype=np.int8)
             self.assertRaises(TypeError, paddle.inner, x, y)
 
+    @test_with_pir_api
     def test_errors_static_case2(self):
         # test static computation graph: inputs must be broadcastable
         paddle.enable_static()

test/legacy_test/test_log_softmax.py

@@ -236,6 +236,7 @@ def test_check_api(self):
             self.check_api(axis)
         self.check_api(-1, 'float64')
 
+    @test_with_pir_api
     def test_errors(self):
         with paddle.static.program_guard(paddle.static.Program()):
             x = paddle.static.data(name='X1', shape=[100], dtype='int32')

test/legacy_test/test_logcumsumexp_op.py

@@ -199,7 +199,6 @@ def test_gpu(self):
 
         self.run_static(use_gpu=True)
 
-    # @test_with_pir_api
     def test_name(self):
         with base.program_guard(base.Program()):
             x = paddle.static.data('x', [3, 4])

test/legacy_test/test_logsumexp.py

@@ -228,6 +228,7 @@ def set_attrs_addition(self):
 
 
 class TestLogsumexpError(unittest.TestCase):
+    @test_with_pir_api
     def test_errors(self):
         with paddle.static.program_guard(paddle.static.Program()):
             self.assertRaises(TypeError, paddle.logsumexp, 1)

test/legacy_test/test_matmul_op.py

@@ -221,6 +221,7 @@ def test_dygraph_without_out(self):
 
 
 class API_TestMmError(unittest.TestCase):
+    @test_with_pir_api
     def test_errors(self):
         with paddle_static_guard():
 

test/legacy_test/test_matrix_nms_op.py

@@ -19,6 +19,7 @@
 from op_test import OpTest
 
 import paddle
+from paddle.pir_utils import test_with_pir_api
 
 
 def python_matrix_nms(
@@ -310,6 +311,7 @@ def set_argument(self):
 
 
 class TestMatrixNMSError(unittest.TestCase):
+    @test_with_pir_api
     def test_errors(self):
         M = 1200
         N = 7

test/legacy_test/test_matrix_power_op.py

@@ -297,6 +297,7 @@ def test_dygraph(self):
 
 
 class TestMatrixPowerAPIError(unittest.TestCase):
+    @test_with_pir_api
     def test_errors(self):
         input_np = np.random.random([4, 4]).astype("float64")
 
@@ -317,13 +318,6 @@ def test_errors(self):
             )
             self.assertRaises(TypeError, paddle.linalg.matrix_power, input, 2)
 
-        # When out is set, the data type must be the same as input.
-        input = paddle.static.data(
-            name="input_1", shape=[4, 4], dtype="float32"
-        )
-        out = paddle.static.data(name="output", shape=[4, 4], dtype="float64")
-        self.assertRaises(TypeError, paddle.linalg.matrix_power, input, 2, out)
-
         # The number of dimensions of input must be >= 2.
         input = paddle.static.data(name="input_2", shape=[4], dtype="float32")
         self.assertRaises(ValueError, paddle.linalg.matrix_power, input, 2)
@@ -348,6 +342,14 @@ def test_errors(self):
             ValueError, paddle.linalg.matrix_power, input, -956301312
         )
 
+    def test_old_ir_errors(self):
+        # When out is set, the data type must be the same as input.
+        input = paddle.static.data(
+            name="input_1", shape=[4, 4], dtype="float32"
+        )
+        out = paddle.static.data(name="output", shape=[4, 4], dtype="float64")
+        self.assertRaises(TypeError, paddle.linalg.matrix_power, input, 2, out)
+
 
 class TestMatrixPowerSingularAPI(unittest.TestCase):
     def setUp(self):

test/legacy_test/test_maxout_op.py

@@ -124,6 +124,7 @@ def test_dygraph_api(self):
         np.testing.assert_allclose(out3_ref, out3.numpy(), rtol=1e-05)
         paddle.enable_static()
 
+    @test_with_pir_api
     def test_errors(self):
         with paddle.static.program_guard(paddle.static.Program()):
             # The input type must be Variable.

test/legacy_test/test_multi_dot_op.py

@@ -260,6 +260,7 @@ def get_inputs_and_outputs(self):
 
 # python API test
 class TestMultiDotOpError(unittest.TestCase):
+    @test_with_pir_api
     def test_errors(self):
         with paddle.static.program_guard(
             paddle.static.Program(), paddle.static.Program()

test/legacy_test/test_outer.py

@@ -142,6 +142,7 @@ def test_multiply_dynamic(self):
 
 
 class TestMultiplyError(unittest.TestCase):
+    @test_with_pir_api
     def test_errors_static(self):
         # test static computation graph: dtype can not be int8
         paddle.enable_static()