未验证 提交 a5a1546a 编写于 作者: G Guanghua Yu 提交者: GitHub

Update paddle_inference_eval.py

上级 fb3fac19
......@@ -338,7 +338,7 @@ def predict_image(predictor,
img, scale_factor = image_preprocess(image_file, image_shape)
inputs = {}
inputs["image"] = img
if include_nms:
if FLAGS.include_nms:
inputs['scale_factor'] = scale_factor
input_names = predictor.get_input_names()
for i, _ in enumerate(input_names):
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册