Fixing Softmax handling

llvm/projects/keras/frontend/approxhpvm_translator.py

@@ -351,6 +351,8 @@ class TensorRtTranslator:
     inst_str = "void* " + output_var + " = "
     inst_str += "tensor" + func_name + "(" + input_var + "); \n"
 
+    print ("***** inst_str = ", inst_str, "\n")
+    
     return inst_str
 
   
@@ -392,7 +394,6 @@ class TensorRtTranslator:
       else:
         inst_str += "1); \n"
         
-
       self.program_str += inst_str
 
 
@@ -421,21 +422,22 @@ class TensorRtTranslator:
       # NOTE: Changing output variable
       out_var_name1 = out_var_name2
 
-      
-    if self.hasActivation(cur_node):
-      activation_type = cur_node.activation_type
-      out_var_name3 = self.getVariableName(cur_node)    
-
-      inst_str = self.genActivationCall(out_var_name1, out_var_name3, activation_type)
-      self.program_str += inst_str  
 
-      
     if layer_type == "Activation":
       input_var_name = self.getSingleInputName(cur_node)
 
       inst_str = self.genActivationCall(input_var_name, out_var_name1, cur_node.activation_type)
+      self.program_str += inst_str
+      
+    
+    if self.hasActivation(cur_node) and layer_type != "Activation":
+      activation_type = cur_node.activation_type
+      out_var_name3 = self.getVariableName(cur_node)    
+
+      inst_str = self.genActivationCall(out_var_name1, out_var_name3, activation_type)
       self.program_str += inst_str  
 
+        
 
     if layer_type == "BatchNormalization":
       input_var_name = self.getSingleInputName(cur_node)
@@ -482,7 +484,9 @@ class TensorRtTranslator:
     # Skip visited nodes
     if cur_node.layer_name in visited_nodes:
       return
-      
+
+    print ("-visiting = ", cur_node.layer_name, "\n")
+    
     if dfg.predVisited(cur_node, visited_nodes):
       
       visited_nodes[cur_node.layer_name] = True