Zealoe

Results 11 comments of Zealoe

How to test on Make3D Dataset?

Thanks for your reply~

unmatched shapes in tf.concat()

So how do you fix the problem? i also meet the same problem. Hope for your reply~ @sampepose @Beanocean @Co1dAt0m

unmatched shapes in tf.concat()

@Beanocean thanks very much！

Segmentation fault when run demo.py or training

make again. it may help

do you use different bounding box regression?

when changing the pooling layers or the conv layers, the perception field may be affected, so there may be some effects in the bounding box regression operation. are there any...

Did you find that it failed when you use padding="VALID"?

the input feature map is 24*24, and the offset conv setting is ![image](https://user-images.githubusercontent.com/30797481/35774321-e1ee4a66-0963-11e8-8206-ecd0e27e47d0.png) the deformable conv setting is ![image](https://user-images.githubusercontent.com/30797481/35774322-ee7746de-0963-11e8-8e4e-994f84df2cb6.png)

Did you find that it failed when you use padding="VALID"?

sorry to disturb, but when i change GetWindowedOutputSize in deform_conv.cc and rebuild, it failed when i run g++_complie.sh. So what should i do to make it work? Thanks! ![image](https://user-images.githubusercontent.com/30797481/35799745-afb6eac8-0aa1-11e8-8169-b2321b7c5712.png)

Did you find that it failed when you use padding="VALID"?

What version of tensorflow do you use? I am using 1.1.0 and GetWindowedOutputSizeV2 is not defined in common_shape_fns.h .. if i copy the file common_shape_fns.h including GetWindowedOutputSizeV2 to my tensorflow,...

How many epochs are required to converge?

also concerned.

AP = -1 for every class on custom dataset

@thanasissdr How do you fix the problem?