Dong Wang comments

Results 41 comments of


                                            Dong Wang

Running PipeCNN on U200

Could you run sw emulation ?

difference between opencl for gpu and opencl for fpga

OpenCL for FPGA is a new way for modeling macro architecture/digital circuit in high-level programming langurage OpenCL for GPU is still parallel programing on SIMD processor

difference between opencl for gpu and opencl for fpga

Hi, you will find a proper way to implement YOLO based-on PipeCNN for sure, because we have verified that. The first thing to do is to fully understand the computation...

difference between opencl for gpu and opencl for fpga

Basicly for YOLO, you only need to modify the convolution kernel to accelerator the algorithm, since the conv operations consumes most of the execution time. You don't really need new...

Input format for resnet-50.

Please refer to the function "loadImageToBuffer" in the host program, which loads the JPEGs into the FPGA.

An error when compiling the project

It seems you are building for ARM processor, make sure you have correct cross-compiling env set up .

Update is messed up when mmyolo adds new learnable parameters for training.

having the same issue on my side ...

Xrt can not correctly recognize the clock freq. in v2021.2

@jvillarre Hello, did v2023.1 fix this issue ? Is there any way that we could mannually fix it in previous versions like v2022.1 ?

`ap_hs` doesn't work with `hls::stream`

I am having the same issue with ap_fifo. Moreover, non-blocking write and read of hls:stream interfaces can not be compiled. However, according the document (), it should be OK to...

Failed to create large buffer : vmap allocation for size XXX failed

@ikwzm Thanks for you quick reply. After enlarging the vmalloc space, this issue was solved. One more question, there is any performance penalties when using very large buffers ? for...