Liangfu Chen

can you trace the error with pdb to see which line of the code generates such error? on the other hand, I would suggest building independent tvm runtime on your...

i think the root cause of this issue might be your opencl driver, and a correct way to compile tvm with opencl, not quite related to running mobilenetv2 inference. therefore,...

Regarding to your questions, 1. I refer to resnet , which has been reproduced in [this repo](https://github.com/tornadomeet/ResNet), for data augmentation. I use 480xN because resnet training can be reproduced that...

Good question, your success is just around the corner! At epoch around 200, I turned augmentation level to 3, and random scale between 0.533 and 0.6, this step fine-tunes the...

That sounds great, but that might consume a long time for training I guess. How many epochs you got until it converge to 71.7?

Thank you for sharing. I would change my training strategy and try again later. I still think even after your converge to 71.8 without changing aug_level, I suggest try changing...

Training logs have been uploaded, please look into the log folder.

I would agree with > derive the value from the VTA target (i.e. FPGA type) in pkg_config , and avoid requiring a user to specify the `ACC_DEP_DISTANCE` parameter.

Just for a quick note, I'm currently using following code snippet to fetch categorical_feature from the model ```python #### HACK, HACK and HACK def get_categorical_feature(model): import ast model_str_list = str(model.model_to_string()).split("\n")...

The implement of the initial support (Ocra-Max style w/o paged attention) is actually done in local branch (https://github.com/liangfu/vllm/tree/neuron), we just need time to upstream. For paged attention, my rough estimate...