[Dy2Stat] Fix bug in convert_call (#28368)

* Fix bug in convert_call

* refine unittest

* refine code

* refine code

* fix unittest failed

* add assert

python/paddle/fluid/dygraph/dygraph_to_static/convert_call_func.py

@@ -142,7 +142,7 @@ def convert_call(func):
             # Note(Aurelius84): Because `@declarative` returns a class instance instead of
             # a function. This will modify the value referring to itself in `__globals__`.
 
-            # For example: 
+            # For example:
             #
             #      @declarative
             #      def foo(x):
@@ -150,7 +150,7 @@ def convert_call(func):
             #
             # `foo` will be converted into a wrapper class, suppose as `StaticFunction`.
             # And `foo.__globals__['foo']` will still return this `StaticFunction` instead of
-            # `foo` function. So `isinstance(fn, StaticFunction)` is added here. 
+            # `foo` function. So `isinstance(fn, StaticFunction)` is added here.
             global_functions = set()
             for fn in func.__globals__.values():
                 if inspect.isfunction(fn):
@@ -193,8 +193,10 @@ def convert_call(func):
             try:
                 _, forward_func = unwrap_decorators(func.forward)
                 forward_func = convert_to_static(forward_func)
-                setattr(func, 'forward', forward_func)
-                func_self = func
+                # Bound mothod will be convert into plain function after `convert_to_static`.
+                # So descriptor mechanism is used to bound `self` instance on function to
+                # keep it as bound method.
+                setattr(func, 'forward', forward_func.__get__(func))
             except Exception:
                 # NOTE: func.forward may have been decorated.
                 func_self = None if func_self else func_self

python/paddle/fluid/tests/unittests/dygraph_to_static/test_lstm.py

@@ -18,14 +18,24 @@ import unittest
 from paddle import nn
 
 
+class LSTMLayer(nn.Layer):
+    def __init__(self, in_channels, hidden_size):
+        super(LSTMLayer, self).__init__()
+        self.cell = nn.LSTM(
+            in_channels, hidden_size, direction='bidirectional', num_layers=2)
+
+    def forward(self, x):
+        x, _ = self.cell(x)
+        return x
+
+
 class Net(nn.Layer):
     def __init__(self, in_channels, hidden_size):
         super(Net, self).__init__()
-        self.lstm = nn.LSTM(
-            in_channels, hidden_size, direction='bidirectional', num_layers=2)
+        self.lstm = LSTMLayer(in_channels, hidden_size)
 
     def forward(self, x):
-        x, _ = self.lstm(x)
+        x = self.lstm(x)
         return x
 
 
@@ -115,5 +125,21 @@ class TestSaveInEvalMode(unittest.TestCase):
                                                             infer_out))
 
 
+class TestEvalAfterSave(unittest.TestCase):
+    def test_eval_after_save(self):
+        x = paddle.randn((2, 10, 12)).astype('float32')
+        net = Net(12, 2)
+        dy_out = net(x)
+        # save model
+        paddle.jit.save(net, 'jit.save/lstm', input_spec=[x])
+        load_net = paddle.jit.load('jit.save/lstm')
+        load_out = load_net(x)
+        self.assertTrue(np.allclose(dy_out.numpy(), load_out.numpy()))
+        # eval
+        net.eval()
+        out = net(x)
+        self.assertTrue(np.allclose(dy_out.numpy(), out.numpy()))
+
+
 if __name__ == "__main__":
     unittest.main()