fix skip_tensor_list name

python/paddle/fluid/contrib/slim/quantization/post_training_quantization.py

@@ -126,7 +126,7 @@ def __init__(self,
                  onnx_format=False,
                  optimize_model=False,
                  is_use_cache_file=False,
-                 skip_op_list=None,
+                 skip_tensor_list=None,
                  cache_dir=None):
         '''
         Constructor.
@@ -199,7 +199,7 @@ def __init__(self,
                 the model accuracy is usually higher when it is 'channel_wise_abs_max'.
             onnx_format(bool): Whether to export the quantized model with format of ONNX.
                 Default is False.
-            skip_op_list(list): List of skip quant tensor name.
+            skip_tensor_list(list): List of skip quant tensor name.
             optimize_model(bool, optional): If set optimize_model as True, it applies
                 some passes to the model before quantization, and it supports
                 `conv2d/depthwise_conv2d + bn` pass so far. Some targets require the
@@ -303,7 +303,7 @@ def __init__(self,
         self._activation_quantize_type = activation_quantize_type
         self._weight_quantize_type = weight_quantize_type
         self._onnx_format = onnx_format
-        self._skip_op_list = skip_op_list
+        self._skip_tensor_list = skip_tensor_list
         self._is_full_quantize = is_full_quantize
         if is_full_quantize:
             self._quantizable_op_type = self._support_quantize_op_type
@@ -550,10 +550,10 @@ def collect_var_name(var_name_list, persistable_var_names, op_type):
         persistable_var_names = _all_persistable_var_names(self._program)
         for block_id in range(len(self._program.blocks)):
             for op in self._program.blocks[block_id].ops:
-                # skip quant form self._skip_op_list
-                if self._skip_op_list is not None:
+                # skip quant form self._skip_tensor_list
+                if self._skip_tensor_list is not None:
                     for inp_name in utils._get_op_input_var_names(op):
-                        if inp_name in self._skip_op_list:
+                        if inp_name in self._skip_tensor_list:
                             op._set_attr("op_namescope", "skip_quant")
 
                 op_type = op.type

python/paddle/fluid/contrib/slim/tests/test_post_training_quantization_mnist.py

@@ -118,7 +118,7 @@ def generate_quantized_model(self,
                                  batch_size=10,
                                  batch_nums=10,
                                  onnx_format=False,
-                                 skip_op_list=None):
+                                 skip_tensor_list=None):
 
         place = fluid.CPUPlace()
         exe = fluid.Executor(place)
@@ -137,7 +137,7 @@ def generate_quantized_model(self,
             is_full_quantize=is_full_quantize,
             optimize_model=is_optimize_model,
             onnx_format=onnx_format,
-            skip_op_list=skip_op_list,
+            skip_tensor_list=skip_tensor_list,
             is_use_cache_file=is_use_cache_file)
         ptq.quantize()
         ptq.save_quantized_model(self.int8_model_path)
@@ -157,7 +157,7 @@ def run_test(self,
                  infer_iterations=10,
                  quant_iterations=5,
                  onnx_format=False,
-                 skip_op_list=None):
+                 skip_tensor_list=None):
 
         origin_model_path = self.download_model(data_url, data_md5, model_name)
         origin_model_path = os.path.join(origin_model_path, model_name)
@@ -172,7 +172,7 @@ def run_test(self,
         self.generate_quantized_model(
             origin_model_path, algo, round_type, quantizable_op_type,
             is_full_quantize, is_use_cache_file, is_optimize_model, batch_size,
-            quant_iterations, onnx_format, skip_op_list)
+            quant_iterations, onnx_format, skip_tensor_list)
 
         print("Start INT8 inference for {0} on {1} images ...".format(
             model_name, infer_iterations * batch_size))
@@ -423,7 +423,7 @@ def test_post_training_avg_skip_op(self):
         batch_size = 10
         infer_iterations = 50
         quant_iterations = 5
-        skip_op_list = ["fc_0.w_0"]
+        skip_tensor_list = ["fc_0.w_0"]
         self.run_test(
             model_name,
             data_url,
@@ -438,7 +438,7 @@ def test_post_training_avg_skip_op(self):
             batch_size,
             infer_iterations,
             quant_iterations,
-            skip_op_list=skip_op_list)
+            skip_tensor_list=skip_tensor_list)
 
 
 if __name__ == '__main__':