Peng Chen

Hi, could you share the top1 and top5 accuracy in your training?

1

RT

Any one test the accuracy on VOC dataset?

7

Hi, everyone, I test the pre-trained model on the VOC dataset, and found the mAP didn't match the expected ones. Here is my result: yolo_v2_mobilenetv2_voc: 0.721956 yolo_v3_mobilenetv2_voc: 0.766846 ssd_lite_mobilenetv2_voc: 0.708510...

Hi, everyone: I tried to re-implement the inplace-ABN based on my understanding of the concept. Out of my expectation, the code below cost more GPU memory rather than reduce the...

how to get the file descriptor of the memory object?

1

I emloyed: ` private_handle_t * gralloc_handle = reinterpret_cast(const_cast(graphicbuffer->getNativeBuffer()->handle)); int fd = gralloc_handle->fd;` to get the file descriptor But it seemed to be wrong and always returned 0x00

Has anyone successfully gain accuracy after quantification on vgg16?

4

I tried 50% log quantization on the pre-trained vgg16, however failed to re-gain the original accuracy. Have anyone successful with the experiments? Any suggestions on how to re-gain the accuracy...

any example on the project?

1

Hi, @xuanyoya Thanks for the interesting project. I'm reading the related paper 'A Systematic Approach to blocking convolutional neural networks' and found it could be more useful to take this...

Hi, @yeahkun Thanks for your nice work. I compared the code with the origin darknet. All pieces of code were understood expect the IOU computing in the loss layer. In...

Hi, thanks for sharing the PCI virtual device code of the QEMU. I recently is re-implementing a similar module. I also found some other related code, eg vpice https://github.com/texane/vpcie. However,...

Hi, Is there any support for Bi-LSTM

3

Hi, thanks for the helpful work. Coudl I ask if any plan for supporting bidirectional lstm with custom stacks?