slmdfin

Hi, I install caffe using the anaconda, and I also have this error. But it looks like the the solution of #1 doesn't fit me, because I can't find the...

Hi, I run this script and I got 83.5% accuracy on test dataset, too. Here is the visualization using tensorboard:  I don't know why, can you give me...

@ybpaopao Hi, I met this problem, too. I test the code, there is no DS_store file in the code, how should I delete them ? I can't find the DS_store.......

您好，可以请问一下您是在什么环境下跑的yolo呢?我在titan x上剪枝完测试，发现时间并没有提高。。。或许是因为剪枝在GPU提升的不明显？

@talebolano 有减少，不是一边播放一边测试的，我就是在测试集上测试的图片。我决定再剪枝试试效果！ 还有一个问题，就是我看到您分享了RETHINKING THE SMALLER-NORM-LESSINFORMATIVE代码，您有写这个论文的阅读博客嘛？因为我看这个论文的时候看的比较费力，而且没有好的参考资料。。。不知道您有写博文吗？

@talebolano 您好，谢谢您的回复。我后来又训练了一次，剪枝前的参数量是48.4M，flops是22.59G；剪枝后的参数量是34M，flops是16.10G。模型体积有减小，但是inference时间还是没有变化，是因为flops减少的小吗？还想请教一个问题，就是在额外惩罚BN的时候，不会对原来的任务有影响吗？

There are two ways to restore models, you need to use the right code and use the right model.

1. Use to load from **ckpt** file saver = tf.train.Saver() saver.restore(sess, model_data_path) 2. Use to load from **npy** file net.load(model_data_path, sess)

Hi, thanks for sharing the great work again. I spent some time reading your source code. I have something confused. I prune the model under the threshold, and then what...

@xiaoyaozzr Hi, I have finished converting caffe to tensorflow for mobilenetv1. But I am a little confused about conv function: `conv(3,3,32,1,1,biased=False,group=32,relu=False,name='conv2_1_dw')` Can you tell me how to write the conv...