update npu inference api (#779)

Author: YanhuiDua

tools/inference.py

@@ -118,7 +118,7 @@ def create_predictor(model_path,
     elif device == "cpu":
         config.disable_gpu()
     elif device == "npu":
-        config.enable_npu()
+        config.enable_custom_device('npu')
     elif device == "xpu":
         config.enable_xpu()
     else:
@@ -174,11 +174,10 @@ def main():
         random.seed(args.seed)
         np.random.seed(args.seed)
     cfg = get_config(args.config_file, args.opt, show=True)
-    predictor, config = create_predictor(args.model_path, args.device, args.run_mode,
-                                 args.batch_size, args.min_subgraph_size,
-                                 args.use_dynamic_shape, args.trt_min_shape,
-                                 args.trt_max_shape, args.trt_opt_shape,
-                                 args.trt_calib_mode)
+    predictor, config = create_predictor(
+        args.model_path, args.device, args.run_mode, args.batch_size,
+        args.min_subgraph_size, args.use_dynamic_shape, args.trt_min_shape,
+        args.trt_max_shape, args.trt_opt_shape, args.trt_calib_mode)
     input_handles = [
         predictor.get_input_handle(name)
         for name in predictor.get_input_names()
@@ -225,15 +224,15 @@ def main():
         elif model_type == "cyclegan":
             import auto_log
             logger = get_logger(name='ppgan')
-            
+
             size = data['A'].shape
             pid = os.getpid()
             auto_logger = auto_log.AutoLogger(
                 model_name=args.model_type,
                 model_precision=args.run_mode,
                 batch_size=args.batch_size,
                 data_shape=size,
-                save_path=args.output_path+'auto_log.lpg',
+                save_path=args.output_path + 'auto_log.lpg',
                 inference_config=config,
                 pids=pid,
                 process_name=None,
@@ -254,7 +253,11 @@ def main():
             save_image(
                 image_numpy,
                 os.path.join(args.output_path, "cyclegan/{}.png".format(i)))
-            logger.info("Inference succeeded! The inference result has been saved in {}".format(os.path.join(args.output_path, "cyclegan/{}.png".format(i))))
+            logger.info(
+                "Inference succeeded! The inference result has been saved in {}"
+                .format(
+                    os.path.join(args.output_path,
+                                 "cyclegan/{}.png".format(i))))
             auto_logger.times.end(stamp=True)
             auto_logger.report()
             metric_file = os.path.join(args.output_path, "cyclegan/metric.txt")