diff --git a/onnxsim/onnx_simplifier.py b/onnxsim/onnx_simplifier.py
index 6ac8d1f..cadb9cd 100644
--- a/onnxsim/onnx_simplifier.py
+++ b/onnxsim/onnx_simplifier.py
@@ -106,7 +106,9 @@ def get_np_type_from_elem_type(elem_type: int):
 
 
 def get_inputs(model: onnx.ModelProto) -> List[onnx.ValueInfoProto]:
+    # weights
     initializer_names = [x.name for x in model.graph.initializer]
+    # input not in weights
     return [ipt for ipt in model.graph.input if ipt.name not in initializer_names]
 
 
@@ -127,9 +129,11 @@ def remove_unused_output(model: onnx.ModelProto, unused_output: Sequence[str]) -
         if unused_output_name not in output_names:
             raise RuntimeError(
                 f'The model doesn\'t have output named "{unused_output_name}"')
+
     for graph_output in copy.deepcopy(model.graph.output):
         if graph_output.name in unused_output_names:
             model.graph.output.remove(graph_output)
+
     model = onnxoptimizer.optimize(model, ['eliminate_deadend'],
                                    fixed_point=True)
     onnx.checker.check_model(model)
@@ -144,9 +148,20 @@ def generate_specific_rand_input(model, input_shapes: TensorShapes):
     for key, shape in input_shapes.items():
         shape_np = np.array(shape)
         if not np.all(shape_np > 0):
-            # treat batch size as 1 automatically if dynamic_input_shape is True
+            
             if config.dynamic_input_shape and len(shape_np) >= 3 and np.all(shape_np[1:] > 0):
+                # treat batch size as 1 automatically if dynamic_input_shape is True
                 input_shapes[key] = [1] + shape[1:]
+                print(f"dynamic input shape: {key} {shape} -> {input_shapes[key]}")
+                continue
+
+            if config.dynamic_input_shape:
+                dyn_dim = np.where(shape_np==-1)[0]
+                shape_tmp = copy.deepcopy(shape)
+                for i, idx in enumerate(dyn_dim):
+                    shape_tmp[idx.item()] = 10 + i
+                input_shapes[key] = shape_tmp
+                print(f"dynamic input shape: {key} {shape} -> {shape_tmp}")
                 continue
 
             raise RuntimeError(
@@ -362,6 +377,7 @@ def optimize(model: onnx.ModelProto, skip_fuse_bn: bool, skipped_optimizers: Opt
     onnx.checker.check_model(model)
     onnx.helper.strip_doc_string(model)
     optimizers_list = onnxoptimizer.get_fuse_and_elimination_passes()
+    print(f'opt passes: {optimizers_list}')
     if skip_fuse_bn:
         optimizers_list.remove('fuse_bn_into_conv')
     if skipped_optimizers is not None: