suxb201

https://github.com/kenshohara/video-classification-3d-cnn-pytorch/issues/47#issuecomment-512637417

``` /home/northmachine/Desktop/video-classification-3d-cnn-pytorch-master/classify.py:25: UserWarning: volatile was removed and now has no effect. Use with torch.no_grad(): instead. inputs = Variable(inputs, volatile=True) ``` You need to add `torch.no_grad():` 

This is my command, may be helpful for you. `python main.py --input ./input --video_root ~/hdd/dataset/tacos/videos/ --output ./output.json --model ./resnext-101-64f-kinetics.pth --mode feature --model_depth 101 --model_name resnext --resnet_shortcut B` NOTE: `--resnet_shortcut B`...

试一下：https://github.com/alibaba/RedisShake/releases/tag/v3.0.0-beta.9

@wwwbjqcom 目前是的，但是资源占用并不会多

理论上没问题，source 写主节点就可以

看这里：https://github.com/alibaba/RedisShake/blob/v3/redis-shake.toml

哨兵和单实例一样，source 只写一个，写主写从都可以。targe 写随便一个节点就行。 参照：https://github.com/alibaba/RedisShake/blob/v3/docs/quick_start.md

@tearstainss 1. 3.0 报错是什么，麻烦贴一下 2. 不正常，再慢也不会一天搞不完的

目的端和源端 redis 是多少？