Created by: ZERO6666

　　自己尝试通过读入视频帧进行推理，发现框架的reader是要求路径的，想知道如何对视频流的帧输入进行推理。 /home/ref/下载/PaddleDetection/ppdet/data/source/simple_source.py:42: FutureWarning: elementwise comparison failed; returning scalar instead, b in the future will perform elementwise comparison assert image != '' and os.path.isfile(image), \ Traceback (most recent call last): File "tools/infer.py", line 326, in <module> main() File "tools/infer.py", line 182, in main reader = create_reader(test_feed) File "/home/ref/下载/PaddleDetection/ppdet/data/data_feed.py", line 149, in create_reader my_source) File "/home/ref/下载/PaddleDetection/ppdet/data/reader.py", line 140, in create return reader._make_reader(mode, my_source) File "/home/ref/下载/PaddleDetection/ppdet/data/reader.py", line 51, in _make_reader sc = build_source(sc_conf) File "/home/ref/下载/PaddleDetection/ppdet/data/source/__init__.py", line 69, in build_source return SimpleSource(**args) File "/home/ref/下载/PaddleDetection/ppdet/data/source/simple_source.py", line 42, in __init__ assert image != '' and os.path.isfile(image), \ File "/home/ref/anaconda2/envs/py37/lib/python3.7/genericpath.py", line 30, in isfile st = os.stat(path) ValueError: stat: embedded null character in path