igo312

@jacobgil I find there is no limits when using `optimizer.step()`. So the pruned weight will get a gradient and after stepping, it will be no longer **0** which means it...

here's the equation  Let's let `Lreg` as `u` and `|-log(IoU)|` as `I`  @yangxue0827 I know there must be something wrong. So could you pleast tell the right wat...

Hi there, I find there should be more instances than `COCO`'s in one image, and I am tring to detect them with Faster RCNN. I have found the sample num...

This error caused because this code `output = tf.reshape(tf.concat(1, outputs), [-1, rnn_size])`, I want to know why a list could be connected with int and why do that

In my label file some file named `P1670_0`, `1650_0`, somelike that. But I cannot find corresponding image in my imagsets. the length of `labelTxt` and `Imagests` are same. Any idea...

Why do pad_to_bounding_box in ram.py?

3

As the title said, I found there is a `tf.image.pad_to_bounding_box ` to get a zero padding.However I did not found the process in the paper. What's that mean?Actually, I also...

thanks for your work, it's really great! But I am not sure what ghost box is, could you please show me a example to help me understand it better? thank...

Hi, Thank you for open-sourcing this project! I have been using it recently and came across an observation. It seems that the `traffic_light` is bound to a specific point on...