Pinnh

icon location CV，ML

Results 7 comments of Pinnh

MACE运行mtcnn的Pnet不认PRELU这种activation

output_shapes: - 1,1,1,1 - 1,1,1,4 改为 output_shapes: - 1,1,1,2 - 1,1,1,4 不知道是不是这个错误

Value error

I find set smoothing_window_size=0 or less than 800 is OK， because barrier - offset is

How to test v100_test.py without GPU ?

It seems that you can't run the v100_test.py without GPU, because the neural render is written by CUDA c++ code which not support cpu

Handle Negative Latitude and Longitude Correctly

I solved this using math.floor() method ,codes available at [here](https://github.com/Pinnh/srtm-python/blob/master/srtm.py)

training error: Data layer prefetch queue empty

I meet the same problem using 1080Ti & CUDA 10.0, but there is no problem when using gtx1060 & CUDA 8.0 I don't think its GPU power problem too, The...

training error: Data layer prefetch queue empty

I get Resolved, in **src/caffe/util/sampler.cpp** **caffe_rng_uniform(1, 0.f, 1 - bbox_width, &w_off); caffe_rng_uniform(1, 0.f, 1 - bbox_height, &h_off);** **caffe_rng_uniform will get block, when bbox_width or bbox_height near 1.0 , (1 -...

Maybe Noise Optimization improve the encoder's restoration effect

I met the same problem, it happens when you compute mse loss with images that being resized . 256x256?