Censor

FIFO file questions?

3

@wangyu- 1.When write mode/timeout/queue-len to fifo file, after checking log file found that queue-len parameter still keep default value 200. In below logs, print_parameter function printed 'fec_queue_len=200'. But expected queue-len=2....

May i know how to turn off logging when i run demo.py. I try to add os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3' and logger.disabled = True. But all not working. I am newcomer...

Handling multiple IPs exposed on the same network interfaces

10

@davidzhao @boks1971 I have kept all the logs this time, please check the attachment. Here are some of the errors I've highlighted. The binary I'm running is compiled from the...

**你的新功能建议是否牵扯到某个常见的问题？** 播放的时候获取每一帧的回调 **你希望达到的效果** 获取到每一帧回调后，我们会对获取到的视频帧进行二次渲染，然后将渲染后的帧回显给用户 **你能考虑到的可选实现方案** ZFPlayer可以支持，但是我们想换成作者的SJVideoPlayer

devzwy的新tflite文件是否兼容

1

我看到android版的作者devzwy更新了tflite文件的模型。模型大小改动较大，原量化模型虽小(6M)，但对GPU加速支持不友好，新模型大约23M，完美支持GPU加速并优化识别精度，加速效果明显。 不知道他的nsfw.tflite能否用于本项目。如果不行，有没有升级的计划。非常感谢你的项目。

Currently only one GPU can be used. Can multiple Gpus be trained at the same time? One GPU is too slow.