Merge pull request #486 from cszdrg/main

修改rsqrt样例

Author: cangtianhuang

tester/api_config/torch_error_skip.txt

@@ -2988,3 +2988,16 @@ paddle.nn.functional.interpolate(Tensor([5803, 8, 256, 192],"float32"), size=tup
 paddle.nn.functional.interpolate(Tensor([5803, 8, 256, 192],"float32"), size=tuple(349,262,), mode="bilinear", align_corners=False, )
 paddle.nn.functional.interpolate(Tensor([5803, 8, 256, 192],"float32"), size=tuple(350,263,), mode="bilinear", align_corners=False, )
 paddle.nn.functional.interpolate(Tensor([7761, 3, 280, 350],"float32"), size=tuple(300,375,), mode="bilinear", data_format="NCHW", )
+paddle.rsqrt(Tensor([10000, 143166, 3],"float16"), )
+paddle.rsqrt(Tensor([10000, 2, 214749],"float16"), )
+paddle.rsqrt(Tensor([1431655766, 3],"float16"), )
+paddle.rsqrt(Tensor([16, 268435457],"float16"), )
+paddle.rsqrt(Tensor([2, 2147483649],"float16"), )
+paddle.rsqrt(Tensor([32, 134217729],"float16"), )
+paddle.rsqrt(Tensor([4294967297, 1],"float16"), )
+paddle.rsqrt(Tensor([715827883, 2, 3],"float16"), )
+paddle.rsqrt(x=Tensor([2, 1073741825, 2],"float16"), )
+paddle.rsqrt(x=Tensor([2, 3, 715827883],"float16"), )
+paddle.rsqrt(x=Tensor([2, 715827883, 3],"float16"), )
+paddle.rsqrt(x=Tensor([477218589, 3, 3],"float16"), )
+paddle.rsqrt(x=Tensor([715827883, 3, 2],"float16"), )