peyer

Train-logloss/bbox-MSE = nan

2

 Anyone could help me, I cannot solve this problem which shown above

How to save the trained model

5

If I want to save the pnet/rnet/onet model , should I add mod.save(...) after mod.fit(...) in train.py?

Hi , Seanlinx: when I train the P net , I find that since the second DataBatch as Input, the 4_2th conv op's output (pred_delta) are almost 10^25 , so...

@rexbu I have successfully compiled files under libGPU/bs to generate library of libbs. However, when I tried to compile files under libGPU/filter, it shown that some files rely on libGPU/base,...

@rexbu. Amazed for your great work about building framework with openGL ES. However, I used caffe to train model on GPU mostly, do you have plan to add support for...

How about Google Pixel Mobile?

1

@madeye, Hi, thanks for your great work. However, I used Google Pixel Mobile with Android 8.1.0 and Adreno 530 GPU, could you update your project?

I have solve #10 after reading tips from https://github.com/rdolbeau/enable_arm_pmu/blob/master/README.md.  What I want to know is does PMCCNTR_EL0 register being shared with each cpu core and will increment it self...

Got "illegal instruction" after insmod ko and read pmccntr_el0 in user-space application

3

@jerinjacobk I have cross-compiled your provided kernel-space code, and insmod it on my develop board, but I still got "illegal instruction", How could I fix it ? I have typed...

I have comment a problem while training a model for face recognition that when increasing num_output to 800k, the prob of each category became 1.0/800k and never channged! I have...

operator of CudnnConvolutionBackward(ThnnConv2DBackward), ThresholdBackward1 Not found

5

Hi, @longcw, thanks for your work. Recently, I need to transform a pytorch model(RFBNet) which has two sibling output to caffe's format. When I used your pytorch2caffe function, it shown...