Merge pull request #347 from Channingss/Conv

add cast(int64) for embedding, fix bug of python2, delete repeated Geater

README.md

@@ -15,7 +15,7 @@ paddlepaddle >= 1.8.0
 **按需安装以下依赖**  
 tensorflow ： tensorflow == 1.14.0  
 caffe ： 无  
-onnx ： onnx == 1.6.0
+onnx ： onnx >= 1.6.0
 
 ## 安装
 ### 安装方式一（推荐）

x2paddle/convert.py

@@ -170,8 +170,8 @@ def onnx2paddle(model_path, save_dir, params_merge=False):
     try:
         import onnx
         version = onnx.version.version
-        if version != '1.6.0':
-            print("[ERROR] onnx==1.6.0 is required")
+        if version < '1.6.0':
+            print("[ERROR] onnx>=1.6.0 is required")
             return
     except:
         print("[ERROR] onnx is not installed, use \"pip install onnx==1.6.0\".")

x2paddle/decoder/onnx_shape_inference.py

@@ -267,8 +267,9 @@ class SymbolicShapeInference:
 
         if pending_nodes and self.verbose_ > 0:
             print('SymbolicShapeInference: orphaned nodes discarded: ')
-            print('\n'.join(
-                [n.op_type + ': ' + n.output[0] for n in pending_nodes]))
+            for n in pending_nodes:
+                print(n.op_type + ': ' + n.output[0])
+
         if input_shapes is not None:
             for input_name, shape in input_shapes.items():
                 for idx in range(len(self.out_mp_.graph.input)):

x2paddle/op_mapper/onnx2paddle/opset9/opset.py

@@ -487,16 +487,6 @@ class OpSet9():
         node.fluid_code.add_layer(
             'hard_shrink', inputs=val_x, output=node, param_attr=attr)
 
-    def Greater(self, node):
-        val_x = self.graph.get_input_node(node, idx=0, copy=True)
-        val_y = self.graph.get_input_node(node, idx=1, copy=True)
-        node.fluid_code.add_layer(
-            'greater_than',
-            inputs={'x': val_x,
-                    'y': val_y},
-            output=node,
-            param_attr=None)
-
     @print_mapping_info
     def Constant(self, node):
         val_output = self.graph.get_node(node.layer.output[0], copy=True)
@@ -566,25 +556,26 @@ class OpSet9():
     def Expand(self, node):
         val_x = self.graph.get_input_node(node, idx=0, copy=True)
         val_shape = self.graph.get_input_node(node, idx=1, copy=True)
-
         if len(val_shape.outputs) == 1:
             self.omit_nodes.append(val_shape.layer_name)
-
-        val_y = self.graph.get_node(node.layer.output[0], copy=True)
-        out_shape = node.out_shapes[0]
         val_x_dtype = val_x.dtype
-
         name_ones = node.layer_name + '_ones'
-        attr_ones = {'shape': out_shape, 'dtype': string(val_x_dtype)}
+        attr_ones = {
+            'shape': val_shape.layer_name,
+            'dtype': string(val_x_dtype),
+            'value': 1
+        }
         node.fluid_code.add_layer(
-            'ones', inputs=None, output=name_ones, param_attr=attr_ones)
+            'fill_constant',
+            inputs=None,
+            output=name_ones,
+            param_attr=attr_ones)
         inputs = {'x': name_ones, 'y': val_x}
-        attr = {'name': string(node.layer_name)}
         node.fluid_code.add_layer(
             'elementwise_mul',
             inputs=inputs,
             output=node.layer_name,
-            param_attr=attr)
+            param_attr=None)
 
     @print_mapping_info
     def Gather(self, node):
@@ -652,9 +643,15 @@ class OpSet9():
         elif axis == 0 and len(indices_shape) > 1:
             if val_x.out_shapes[0] is not None and isinstance(
                     val_x, ONNXGraphDataNode):
+                indices_cast = indices.layer_name + '_cast'
                 node.fluid_code.add_layer(
-                    'embedding',
+                    'cast',
                     inputs=indices,
+                    output=indices_cast,
+                    param_attr={'dtype': string('int64')})
+                node.fluid_code.add_layer(
+                    'embedding',
+                    inputs=indices_cast,
                     output=node,
                     use_fluid=True,
                     param_attr={
@@ -663,7 +660,6 @@ class OpSet9():
                     })
             else:
                 from functools import reduce
-                #indices_shape = [1,7]
                 reshape_shape = reduce(lambda x, y: x * y, indices_shape)
                 indices_reshape = indices.layer_name + '_shape'
                 node.fluid_code.add_layer(
@@ -703,7 +699,7 @@ class OpSet9():
             perm = list(range(len(val_x.out_shapes[0])))
             perm = [axis] + perm[:axis] + perm[axis + 1:]
             attr_trans = {'perm': perm}
-            name_trans = val_x.layer_name + '_trans'
+            name_trans = val_x.layer_name + '_transpose'
             node.fluid_code.add_layer(
                 'transpose',
                 inputs=val_x,
@@ -715,8 +711,12 @@ class OpSet9():
                         'index': indices_reshape},
                 output=node,
                 param_attr=None)
+            input_transpose = node.layer_name + '_transpose'
             node.fluid_code.add_layer(
-                'transpose', inputs=node, output=node, param_attr=attr_trans)
+                'transpose',
+                inputs=node,
+                output=input_transpose,
+                param_attr=attr_trans)
             val_x_shape = val_x.out_shapes[0]
             reshaped_shape = []
             for i in perm:
@@ -725,7 +725,7 @@ class OpSet9():
                 reshaped_shape.append(i)
             node.fluid_code.add_layer(
                 'reshape',
-                inputs=node,
+                inputs=input_transpose,
                 output=node,
                 param_attr={'shape': reshaped_shape})
 
@@ -859,17 +859,21 @@ class OpSet9():
                 }
             else:
                 if starts.dtype != 'int32':
+                    starts_cast = starts.layer_name + '_cast'
                     node.fluid_code.add_layer(
                         'cast',
                         inputs=starts,
-                        output=starts,
+                        output=starts_cast,
                         param_attr={'dtype': string('int32')})
+                    attr['starts'] = starts_cast
                 if ends.dtype != 'int32':
+                    ends_cast = ends.layer_name + '_cast'
                     node.fluid_code.add_layer(
                         'cast',
                         inputs=ends,
-                        output=ends,
+                        output=ends_cast,
                         param_attr={'dtype': string('int32')})
+                    attr['ends'] = ends_cast
         else:
             starts = node.get_attr('starts')
             ends = node.get_attr('ends')
@@ -1138,7 +1142,7 @@ class OpSet9():
         x_shape = val_x.out_shapes[0]
         y_shape = val_y.out_shapes[0]
         inputs = {"x": val_x, "y": val_y}
-        if y_shape[0] == 1 and x_shape[-1] != 1:
+        if y_shape[0] == 1 and x_shape[-1] != 1 and x_shape[0] != 1:
             y_squeeze = val_y.layer_name + '_squeeze'
             node.fluid_code.add_layer(
                 "squeeze",
@@ -1286,7 +1290,6 @@ class OpSet9():
                     'y': cast_condition},
             output=mul_val_x,
             param_attr=None)
-
         mul_val_y = val_y.layer_name + '_mul'
         node.fluid_code.add_layer(
             "elementwise_mul",
@@ -1339,7 +1342,8 @@ class OpSet9():
             if val_repeats.dtype != 'int32':
                 attr = {"dtype": string("int32")}
                 node.fluid_code.add_layer(
-                    "cast", inputs=repeats, 
+                    "cast",
+                    inputs=repeats,
                     output="{}.tmp".format(repeats),
                     param_attr=attr)
                 repeats = "{}.tmp".format(repeats)