[Frontend][Paddle] [PaddlePaddle Hackathon 4]add attribute support fo…

…r dropout/hard_sigmoid/pixel_shuffle (apache#14575)

* add attr support for paddle ops

* Update paddlepaddle.py

* Update test_forward.py

* codestyle

* remove approximate strategy in gelu

* remove gelu test

* codestyle

* add dropout_implementation attr for dropout

* add tests for dropout

* codestyle

python/tvm/relay/frontend/paddlepaddle.py

@@ -502,7 +502,13 @@ def convert_dropout(g, op, block):
     """Operator converter for dropout."""
 
     x = g.get_node(op.input("X")[0])
-    g.add_node(op.output("Out")[0], x)
+    dropout_prob = op.attr("dropout_prob")
+    dropout_implementation = op.attr("dropout_implementation")
+    if dropout_implementation == "downgrade_in_infer":
+        out = _op.nn.dropout(x, dropout_prob) * _expr.const(1 - dropout_prob, dtype="float32")
+    else:
+        out = _op.nn.dropout(x, dropout_prob)
+    g.add_node(op.output("Out")[0], out)
 
 
 def convert_dot(g, op, block):
@@ -897,8 +903,9 @@ def convert_hard_sigmoid(g, op, block):
     """Operator converter for hard_sigmoid."""
 
     slope = op.attr("slope")
+    offset = op.attr("offset")
     x = g.get_node(op.input("X")[0])
-    out = x * _expr.const(slope) + _expr.const(0.5)
+    out = x * _expr.const(slope) + _expr.const(offset)
     out = _op.clip(out, 0, 1)
     g.add_node(op.output("Out")[0], out)
 
@@ -1425,7 +1432,8 @@ def convert_pixel_shuffle(g, op, block):
 
     x = g.get_node(op.input("X")[0])
     upscale_factor = op.attr("upscale_factor")
-    out = _op.nn.depth_to_space(x, upscale_factor, mode="CRD")
+    data_format = op.attr("data_format")
+    out = _op.nn.depth_to_space(x, block_size=upscale_factor, layout=data_format, mode="CRD")
     g.add_node(op.output("Out")[0], out)
 
 

tests/python/frontend/paddlepaddle/test_forward.py

@@ -617,10 +617,22 @@ def test_forward_dropout():
     def dropout(inputs):
         return nn.functional.dropout(inputs)
 
+    @paddle.jit.to_static
+    def dropout1(inputs):
+        return nn.functional.dropout(inputs, 0.1)
+
+    @paddle.jit.to_static
+    def dropout2(inputs):
+        return nn.functional.dropout(inputs, 0.1, mode="downscale_in_infer")
+
     input_shape = [1, 3, 10, 10]
     input_data = paddle.rand(input_shape, dtype="float32")
     verify_model(dropout, input_data=input_data[0, 0])
     verify_model(dropout, input_data=input_data)
+    verify_model(dropout1, input_data=input_data[0, 0])
+    verify_model(dropout1, input_data=input_data)
+    verify_model(dropout2, input_data=input_data[0, 0])
+    verify_model(dropout2, input_data=input_data)
 
 
 def test_forward_elemwise():
@@ -1008,9 +1020,13 @@ def test_forward_hard_sigmoid():
     def hard_sigmoid(inputs):
         return nn.functional.hardsigmoid(inputs)
 
+    def hard_sigmoid1(inputs):
+        return nn.functional.hardsigmoid(inputs, offset=0.6)
+
     input_shape = [1, 3, 10, 10]
     input_data = paddle.rand(input_shape, dtype="float32")
     verify_model(hard_sigmoid, input_data=input_data)
+    verify_model(hard_sigmoid1, input_data=input_data)
 
 
 @tvm.testing.uses_gpu
@@ -1781,9 +1797,9 @@ def forward(self, input1, input2):
 @tvm.testing.uses_gpu
 def test_forward_pixel_shuffle():
     class PixelShuffle(nn.Layer):
-        def __init__(self, upscale_factor):
+        def __init__(self, upscale_factor, data_format="NCHW"):
             super(PixelShuffle, self).__init__()
-            self.pixel_shuffle = paddle.nn.PixelShuffle(upscale_factor)
+            self.pixel_shuffle = paddle.nn.PixelShuffle(upscale_factor, data_format)
 
         @paddle.jit.to_static
         def forward(self, x):
@@ -1794,6 +1810,11 @@ def forward(self, x):
         x = paddle.rand(input_shape, dtype="float32")
         verify_model(PixelShuffle(2), x)
 
+    input_shapes = [[1, 3, 3, 4], [2, 2, 5, 8]]
+    for input_shape in input_shapes:
+        x = paddle.rand(input_shape, dtype="float32")
+        verify_model(PixelShuffle(2, data_format="NHWC"), x)
+
 
 @tvm.testing.uses_gpu
 def test_forward_prelu():