Add FP16 & BF16 for nanmedian (#56056)

PaddlePaddle · Aug 9, 2023 · 4ae9945 · 4ae9945
1 parent 08e46d6
commit 4ae9945
Show file tree

Hide file tree

Showing 4 changed files with 52 additions and 4 deletions.
diff --git a/paddle/phi/kernels/gpu/nanmedian_grad_kernel.cu b/paddle/phi/kernels/gpu/nanmedian_grad_kernel.cu
@@ -123,4 +123,5 @@ PD_REGISTER_KERNEL(nanmedian_grad,
                    double,
                    int,
                    int64_t,
-                   phi::dtype::float16) {}
+                   phi::dtype::float16,
+                   phi::dtype::bfloat16) {}
diff --git a/paddle/phi/kernels/gpu/nanmedian_kernel.cu b/paddle/phi/kernels/gpu/nanmedian_kernel.cu
@@ -287,6 +287,7 @@ PD_REGISTER_KERNEL(nanmedian,
                    double,
                    int,
                    int64_t,
-                   phi::dtype::float16) {
+                   phi::dtype::float16,
+                   phi::dtype::bfloat16) {
   kernel->OutputAt(1).SetDataType(phi::DataType::INT64);
 }
diff --git a/python/paddle/tensor/stat.py b/python/paddle/tensor/stat.py
@@ -265,7 +265,7 @@ def nanmedian(x, axis=None, keepdim=False, name=None):
     the average value of both elements in the middle is calculated as the median.
 
     Args:
-        x (Tensor): The input Tensor, it's data type can be int32, int64, float16, float32, float64.
+        x (Tensor): The input Tensor, it's data type can be int32, int64, float16, bfloat16, float32, float64.
         axis (None|int|list|tuple, optional):
             The axis along which to perform median calculations ``axis`` should be int or list of int.
             ``axis`` should be in range [-D, D), where D is the dimensions of ``x`` .
@@ -319,7 +319,7 @@ def nanmedian(x, axis=None, keepdim=False, name=None):
         check_variable_and_dtype(
             x,
             'X',
-            ['int32', 'int64', 'float16', 'float32', 'float64'],
+            ['int32', 'int64', 'float16', 'float32', 'float64', 'uint16'],
             'nanmedian',
         )
 

diff --git a/test/legacy_test/test_nanmedian.py b/test/legacy_test/test_nanmedian.py
@@ -15,6 +15,7 @@
 import unittest
 
 import numpy as np
+from eager_op_test import OpTest, convert_float_to_uint16
 
 import paddle
 from paddle.fluid import core
@@ -243,5 +244,50 @@ def test_check_grad_0d(self):
         np.testing.assert_allclose(x.grad, np.array(0.0))
 
 
+class TestNanmedianFP16Op(OpTest):
+    def setUp(self):
+        self.op_type = "nanmedian"
+        self.python_api = paddle.nanmedian
+        self.public_python_api = paddle.nanmedian
+        self.dtype = np.float16
+        self.python_out_sig = ["Out"]
+        X = np.random.random((100, 100)).astype('float16')
+        Out = np.nanmedian(X)
+        self.inputs = {'X': X}
+        self.outputs = {'Out': Out}
+
+    def test_check_output(self):
+        self.check_output()
+
+    def test_check_grad(self):
+        self.check_grad(['X'], 'Out')
+
+
+@unittest.skipIf(
+    not core.is_compiled_with_cuda()
+    or not core.is_bfloat16_supported(core.CUDAPlace(0)),
+    "core is not complied with CUDA and not support the bfloat16",
+)
+class TestNanmedianBF16Op(OpTest):
+    def setUp(self):
+        self.op_type = "nanmedian"
+        self.python_api = paddle.nanmedian
+        self.public_python_api = paddle.nanmedian
+        self.dtype = np.uint16
+        self.python_out_sig = ["Out"]
+        X = np.random.random((100, 100)).astype('float32')
+        Out = np.nanmedian(X)
+        self.inputs = {'X': convert_float_to_uint16(X)}
+        self.outputs = {'Out': convert_float_to_uint16(Out)}
+
+    def test_check_output(self):
+        place = core.CUDAPlace(0)
+        self.check_output_with_place(place)
+
+    def test_check_grad(self):
+        place = core.CUDAPlace(0)
+        self.check_grad_with_place(place, ['X'], 'Out')
+
+
 if __name__ == "__main__":
     unittest.main()