【PIR API adaptor No.261、273、283、285、286、313、315】 Migrate is_tensor/median/nanmean/nansum/neg/Unflatten/var into pir

python/paddle/tensor/logic.py

@@ -1106,8 +1106,10 @@ def is_tensor(x):
             False
 
     """
-    if in_dynamic_mode():
-        return isinstance(x, (Tensor, paddle.base.core.eager.Tensor))
+    if in_dynamic_or_pir_mode():
+        return isinstance(
+            x, (Tensor, paddle.base.core.eager.Tensor, paddle.pir.OpResult)
+        )
     else:
         return isinstance(x, Variable)
 

python/paddle/tensor/manipulation.py

@@ -5596,7 +5596,7 @@ def unflatten(x, axis, shape, name=None):
         new_shape = (
             list(x.shape[:axis]) + list(shape) + list(x.shape[axis + 1 :])
         )
-    elif isinstance(shape, Variable):
+    elif isinstance(shape, (Variable, paddle.pir.OpResult)):
         # The data type returned by `paddle.shape` is only 'int32'.
         new_shape = paddle.concat(
             [

python/paddle/tensor/stat.py

@@ -16,6 +16,7 @@
 
 import paddle
 from paddle import _C_ops
+from paddle.base.libpaddle import DataType
 from paddle.framework import in_dynamic_mode, in_dynamic_or_pir_mode
 
 from ..base.data_feeder import check_type, check_variable_and_dtype
@@ -313,7 +314,7 @@ def nanmedian(x, axis=None, keepdim=False, name=None):
             >>> print(y4.numpy())
             2.0
     """
-    if not isinstance(x, Variable):
+    if not isinstance(x, (Variable, paddle.pir.OpResult)):
         raise TypeError("In median, the input x should be a Tensor.")
 
     if isinstance(axis, (list, tuple)) and len(axis) == 0:
@@ -403,10 +404,11 @@ def median(x, axis=None, keepdim=False, name=None):
             [[4., 5., 6., 7.]])
 
     """
-    if not isinstance(x, Variable):
+    if not isinstance(x, (Variable, paddle.pir.OpResult)):
         raise TypeError("In median, the input x should be a Tensor.")
 
-    if x.size == 0:
+    if in_dynamic_mode() and x.size == 0:
+        # TODO: Currently, `__eq__` don't support arguments (`pir.Value` & `int`)
         raise ValueError("In median, the size of input x should not be 0.")
 
     is_flatten = False
@@ -435,7 +437,11 @@ def median(x, axis=None, keepdim=False, name=None):
     sz = x.shape[axis]
     kth = sz >> 1
     tensor_topk, idx = paddle.topk(x, kth + 1, axis=axis, largest=False)
-    dtype = 'float64' if x.dtype == core.VarDesc.VarType.FP64 else 'float32'
+    dtype = (
+        'float64'
+        if x.dtype in [core.VarDesc.VarType.FP64, DataType.FLOAT64]
+        else 'float32'
+    )
     if sz & 1 == 0:
         out_tensor = paddle.slice(
             tensor_topk, axes=[axis], starts=[kth - 1], ends=[kth]

test/legacy_test/test_bce_with_logits_loss.py

@@ -167,7 +167,7 @@ def test_BCEWithLogitsLoss(self):
                 np.testing.assert_allclose(dy_functional, expected, rtol=1e-05)
 
                 @test_with_pir_api
-                def test_dynamic_or_pir_mode():
+                def test_static_or_pir_mode():
                     static_result = test_static(
                         place, logit_np, label_np, reduction=reduction
                     )
@@ -192,7 +192,7 @@ def test_dynamic_or_pir_mode():
                         static_functional, dy_functional, rtol=1e-05
                     )
 
-                test_dynamic_or_pir_mode()
+                test_static_or_pir_mode()
 
     def test_BCEWithLogitsLoss_weight(self):
         logit_np = np.random.uniform(0.1, 0.8, size=(2, 3, 4, 10)).astype(
@@ -230,7 +230,7 @@ def test_BCEWithLogitsLoss_weight(self):
             np.testing.assert_allclose(dy_functional, expected, rtol=1e-05)
 
             @test_with_pir_api
-            def test_dynamic_or_pir_mode():
+            def test_static_or_pir_mode():
                 static_result = test_static(
                     place,
                     logit_np,
@@ -257,7 +257,7 @@ def test_dynamic_or_pir_mode():
                     static_functional, dy_functional, rtol=1e-05
                 )
 
-            test_dynamic_or_pir_mode()
+            test_static_or_pir_mode()
 
     def test_BCEWithLogitsLoss_pos_weight(self):
         logit_np = np.random.uniform(0.1, 0.8, size=(2, 3, 4, 10)).astype(
@@ -294,7 +294,7 @@ def test_BCEWithLogitsLoss_pos_weight(self):
         np.testing.assert_allclose(dy_functional, expected, rtol=1e-05)
 
         @test_with_pir_api
-        def test_dynamic_or_pir_mode():
+        def test_static_or_pir_mode():
             static_result = test_static(
                 place, logit_np, label_np, weight_np, reduction, pos_weight_np
             )
@@ -315,7 +315,7 @@ def test_dynamic_or_pir_mode():
                 static_functional, dy_functional, rtol=1e-05
             )
 
-        test_dynamic_or_pir_mode()
+        test_static_or_pir_mode()
 
     def test_BCEWithLogitsLoss_error(self):
         paddle.disable_static()

test/legacy_test/test_is_tensor.py

@@ -15,6 +15,7 @@
 import unittest
 
 import paddle
+from paddle.pir_utils import test_with_pir_api
 
 DELTA = 0.00001
 
@@ -49,10 +50,12 @@ def setUp(self):
     def tearDown(self):
         paddle.disable_static()
 
+    @test_with_pir_api
     def test_is_tensor(self):
         x = paddle.rand([3, 2, 4], dtype='float32')
         self.assertTrue(paddle.is_tensor(x))
 
+    @test_with_pir_api
     def test_is_tensor_array(self):
         x = paddle.tensor.create_array('float32')
         self.assertTrue(paddle.is_tensor(x))

test/legacy_test/test_median.py

@@ -17,7 +17,7 @@
 import numpy as np
 
 import paddle
-from paddle.static import Program, program_guard
+from paddle.pir_utils import test_with_pir_api
 
 DELTA = 1e-6
 
@@ -32,10 +32,10 @@ def static_single_test_median(self, lis_test):
         paddle.enable_static()
         x, axis, keepdims = lis_test
         res_np = np.median(x, axis=axis, keepdims=keepdims)
-        main_program = Program()
-        startup_program = Program()
+        main_program = paddle.static.Program()
+        startup_program = paddle.static.Program()
         exe = paddle.static.Executor()
-        with program_guard(main_program, startup_program):
+        with paddle.static.program_guard(main_program, startup_program):
             x_in = paddle.static.data(shape=x.shape, dtype=x.dtype, name='x')
             y = paddle.median(x_in, axis, keepdims)
             [res_pd] = exe.run(feed={'x': x}, fetch_list=[y])
@@ -48,6 +48,7 @@ def dygraph_single_test_median(self, lis_test):
         res_pd = paddle.median(paddle.to_tensor(x), axis, keepdims)
         self.check_numpy_res(res_pd.numpy(False), res_np)
 
+    @test_with_pir_api
     def test_median_static(self):
         h = 3
         w = 4

test/legacy_test/test_nanmean_api.py

@@ -18,6 +18,7 @@
 
 import paddle
 from paddle.base import core
+from paddle.pir_utils import test_with_pir_api
 
 np.random.seed(10)
 
@@ -38,6 +39,7 @@ def setUp(self):
             else paddle.CPUPlace()
         )
 
+    @test_with_pir_api
     def test_api_static(self):
         paddle.enable_static()
         with paddle.static.program_guard(paddle.static.Program()):
@@ -87,6 +89,7 @@ def test_case(x, axis=None, keepdim=False):
         test_case(self.x, [0, 1, 2, 3])
         paddle.enable_static()
 
+    @test_with_pir_api
     def test_errors(self):
         paddle.enable_static()
         with paddle.static.program_guard(paddle.static.Program()):

test/legacy_test/test_nansum_api.py

@@ -18,9 +18,11 @@
 
 import paddle
 from paddle import base
+from paddle.pir_utils import test_with_pir_api
 
 
 class API_Test_Nansum(unittest.TestCase):
+    @test_with_pir_api
     def test_static_graph(self):
         paddle.enable_static()
         startup_program = base.Program()
@@ -75,6 +77,7 @@ def test_static_graph(self):
             )
 
     # test nansum api with float16
+    @test_with_pir_api
     def test_static_graph_fp16(self):
         paddle.enable_static()
         startup_program = paddle.static.Program()

test/legacy_test/test_neg_op.py

@@ -17,6 +17,7 @@
 import numpy as np
 
 import paddle
+from paddle.pir_utils import test_with_pir_api
 
 
 class TestNegOp(unittest.TestCase):
@@ -35,6 +36,7 @@ def run_imperative(self):
             dy_result.numpy(), expected_result, rtol=1e-05
         )
 
+    @test_with_pir_api
     def run_static(self, use_gpu=False):
         input = paddle.static.data(
             name='input', shape=[32, 8], dtype=self.dtype