Update prune interface (#3336)

tools/infer.py

@@ -106,12 +106,12 @@ def prune_feed_vars(feeded_var_names, target_vars, prog):
     """
     exist_var_names = []
     prog = prog.clone()
-    prog = prog._prune(targets=target_vars)
+    prog = prog._prune(feeded_var_names, targets=target_vars)
     global_block = prog.global_block()
     for name in feeded_var_names:
         try:
             v = global_block.var(name)
-            exist_var_names.append(v.name)
+            exist_var_names.append(v.name.encode('utf-8'))
         except Exception:
             logger.info('save_inference_model pruned unused feed '
                         'variables {}'.format(name))
@@ -127,8 +127,9 @@ def save_infer_model(FLAGS, exe, feed_vars, test_fetches, infer_prog):
     feeded_var_names = prune_feed_vars(feeded_var_names, target_vars,
                                        infer_prog)
     logger.info("Save inference model to {}, input: {}, output: "
-                "{}...".format(save_dir, feeded_var_names,
-                               [var.name for var in target_vars]))
+                "{}...".format(save_dir, feeded_var_names, [
+                    var.name.encode('utf-8') for var in target_vars
+                ]))
     fluid.io.save_inference_model(
         save_dir,
         feeded_var_names=feeded_var_names,
@@ -216,7 +217,7 @@ def main():
         from tb_paddle import SummaryWriter
         tb_writer = SummaryWriter(FLAGS.tb_log_dir)
         tb_image_step = 0
-        tb_image_frame = 0 # each frame can display ten pictures at most. 
+        tb_image_frame = 0  # each frame can display ten pictures at most. 
 
     imid2path = reader.imid2path
     for iter_id, data in enumerate(reader()):
@@ -257,7 +258,7 @@ def main():
                                       int(im_id), catid2name,
                                       FLAGS.draw_threshold, bbox_results,
                                       mask_results)
-            
+
             # use tb-paddle to log image with bbox
             if FLAGS.use_tb:
                 infer_image_np = np.array(image)