Phung Cheng Fei

DeepSpeed initialization with GNN-like model

20

My code is quite similar to some GNN structure : [NN_output = graph.forward(NN_input, types="f")](https://gist.github.com/buttercutter/b6f526c56e20f029d68e6f9041c3f5c0/3d1a3e6844680545fb8b75225267ea625ba9df5b#file-gdas-py-L665) So, [outputs = model_engine(inputs)](https://github.com/microsoft/DeepSpeedExamples/blob/36212dd59cb3eb342c39bc8965aaba04d5491933/cifar/cifar10_deepspeed.py#L281) seems does not really fit in my case ? `args` also does...

When I use [my own modified pytorch-pruning for python 3](https://github.com/promach/pytorch-pruning/) with [this fruits dataset](https://github.com/Horea94/Fruit-Images-Dataset) , I have quite low accuracy value. May I know why ? Besides, how do we...

Questions about DARTS

52

1. For DARTS complexity analysis, anyone have any idea how to derive the *(k+1)\*k/2* expression ? Why 2 input nodes ? How will the calculated value change if graph isomorphism...

1. For DARTS complexity analysis, anyone have any idea how to derive the *(k+1)\*k/2* expression ? Why 2 input nodes ? How will the calculated value change if graph isomorphism...

I have few questions on the [section : Approximate Architecture Gradient in the paper](https://arxiv.org/pdf/1806.09055.pdf#page=4) 1. Why Evaluating the finite difference requires only _two forward passes_ for the weights and _two...

build_graph() does not return

2

I tried to use `hiddenlayer` on this [network architecture search coding](https://gist.github.com/promach/b6f526c56e20f029d68e6f9041c3f5c0#file-gdas-py-L423-L434) for the following [search architecture](https://github.com/D-X-Y/AutoDL-Projects/issues/99#issuecomment-835802887), however the code could not exit and is inside `build_graph()` forever. 

Could anyone explain the [working mechanism of sgx spectre attack](https://github.com/lsds/spectre-attack-sgx/blob/master/SGXSpectre/main/main.c#L118-L150) ? How is `ecall_get_offset()` [caching function](https://github.com/lsds/spectre-attack-sgx/blob/master/SGXSpectre/enclave/enclave_attack.c#L27-L30) actually used in this [sgx spectre exploit](https://en.wikipedia.org/wiki/Spectre_(security_vulnerability)) ? ``` int spectre_main(int argc, char **argv)...

Modify YOLOv3 backbone from DarkNet to AdderNet

17

How to correctly modify https://github.com/eriklindernoren/PyTorch-YOLOv3 to use https://github.com/huawei-noah/AdderNet ? The following [colab ipynb notebook](https://gist.github.com/promach/8ddd9794f242c24ffdaa612bcb0bfa33#file-yolov3_-_addernet-ipynb) is what I have so far with the helps of others: https://colab.research.google.com/drive/1VCafwykgNKAO6144LssBFFy0TmruDNSE#scrollTo=W3e-WcVxnKfs How to solve the...

Cluster-NMS

7

I am trying to understand [Cluster-NMS operations](https://github.com/Zzh-tju/CIoU#description-of-cluster-nms-and-its-usage). The [mathematical proof](https://arxiv.org/pdf/2005.03572.pdf#page=6) seems a bit complicated to follow and comprehend. 1. Why C1 does not change values ? In other words, why...

batch_size issue

2

I ran `python search.py --name cifar10 --dataset cifar10 --batch_size 32` , but I have the following error at [utils.py](https://github.com/khanrc/pt.darts/blob/master/utils.py#L103). Why ? ``` Traceback (most recent call last): File "/home/phung/Downloads/darts/pt.darts/search.py", line...