[Prim] support amp O1 in prim (#52598)

python/paddle/jit/dy2static/partial_program.py

@@ -287,6 +287,8 @@ class PartialProgramLayer:
                 amp_program, self._amp_list
             )
         if is_infer_mode:
+            if self._hooker:
+                amp_program = self._hooker.after_infer(amp_program)
             return amp_program
         else:
             train_amp_program = self._append_backward_desc(amp_program)

python/paddle/jit/dy2static/program_translator.py

@@ -19,7 +19,6 @@ import threading
 import warnings
 import weakref
 
-from paddle.amp.auto_cast import _in_amp_guard
 from paddle.fluid import _non_static_mode, core, framework
 from paddle.fluid.data_feeder import check_type
 from paddle.fluid.dygraph.base import (
@@ -1228,7 +1227,7 @@ class ProgramCache:
         partial_program = partial_program_from(
             concrete_program, cache_key.class_instance is not None
         )
-        if core._is_fwd_prim_enabled() and not _in_amp_guard():
+        if core._is_fwd_prim_enabled():
             partial_program.set_hooker(
                 PrimHooker(concrete_program.main_program)
             )

test/prim/process/test_prim_amp.py

+# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import unittest
+
+import numpy as np
+
+import paddle
+import paddle.nn.functional as F
+from paddle import nn
+from paddle.fluid import core, framework
+from paddle.nn import BatchNorm
+
+np.random.seed(2023)
+
+
+class PrimeNet(paddle.nn.Layer):
+    def __init__(self):
+        super().__init__()
+        self.conv = nn.Conv2D(2, 4, (3, 3), bias_attr=False)
+        self.bn = BatchNorm(4, act="relu")
+
+    def forward(self, x):
+        y = self.conv(x)
+        out = self.bn(y)
+        res = F.max_pool2d(out, kernel_size=2, stride=2, padding=0)
+        return res
+
+
+class TestPrimAMPO1(unittest.TestCase):
+    """
+    Test PrimeNet with @to_static + prim v.s Dygraph in AMPO1.
+    """
+
+    def setUp(self):
+        paddle.seed(2022)
+        self.x = paddle.randn([4, 2, 6, 6], dtype="float32")
+        self.x.stop_gradient = False
+
+    def train(self, use_prim):
+        core._set_prim_all_enabled(use_prim)
+        paddle.seed(2022)
+        net = PrimeNet()
+        sgd = paddle.optimizer.SGD(
+            learning_rate=0.1, parameters=net.parameters()
+        )
+
+        if use_prim:
+            net = paddle.jit.to_static(net, build_strategy=False)
+        with paddle.amp.auto_cast(level='O1'):
+            out = net(self.x)
+            loss = paddle.mean(out)
+            loss.backward()
+            sgd.step()
+            sgd.clear_grad()
+            return loss
+
+    def test_amp_01(self):
+        if not isinstance(framework._current_expected_place(), core.CPUPlace):
+            expected = self.train(False)
+            actual = self.train(True)
+            np.testing.assert_allclose(
+                expected,
+                actual,
+                rtol=1e-3,
+                atol=1e-3,
+            )
+
+    def test_amp_O1_infer(self):
+        if not isinstance(framework._current_expected_place(), core.CPUPlace):
+            net = PrimeNet()
+            core._set_prim_all_enabled(False)
+            net.eval()
+            static_net = paddle.jit.to_static(net, build_strategy=False)
+            res = static_net(self.x)
+
+            # set prim all enabled
+            core._set_prim_all_enabled(True)
+            net.eval()
+            static_net = paddle.jit.to_static(net, build_strategy=False)
+            with paddle.amp.auto_cast(level='O1'):
+                res_amp = static_net(self.x)
+
+            np.testing.assert_allclose(
+                res,
+                res_amp,
+                rtol=1e-3,
+                atol=1e-3,
+            )
+
+
+if __name__ == '__main__':
+    unittest.main()