-
An exciting work!
It seems there is no "path" folder to support futher test or training. Will you upload is later?
-
请问您有打算发布cuda框架下使用adder方法的代码吗?
有的话请问大概什么时候能发布呢?加速大吗?
-
imagenet训练代码可以公布吗?
-
EfficientDet: Scalable and Efficient Object Detection
* paper: https://arxiv.org/abs/1911.09070v1
> First, we propose a weighted bi-directional feature pyramid network (BiFPN), which allows ea…
-
Hello, I want to realize the Transposed Adder layer according to the relationship between convolution and transposed convolution, but from the result I found that the image it generates has very obvio…
-
Hi,
I see recent you have applying addernet for EDSR quantization, which is a great work.
But for some other SR, like EDVR, using deformable conv for the frame alignment.
I wonder whether adder…
-
AdderNet模型量化
请问一下AdderNet模型的int8量化代码是否有打算开源?如果不方便的话,能否说一下
AdderNet and Its Minimalist Hardware Design for Energy-Efficient Artificial Intelligence
量化部分参考的论文/代码?
我们发现模型训练出来的权重范围比ConvNet大比较多,大约-7~6,…
-
Hi. I have a CNN architecture that I trained on CIFAR-10 with and without AdderNet. I could reach an accuracy of over %80 without AdderNet but when I used AdderNet it got stuck at %10 accuracy. Is the…
-
您好,我想问一下adder.py 第46行,
为何梯度需要乘上math.sqrt(W_col.size(1)*W_col.size(0))?
此外,后面的“/5”是怎么得到的?
目前我在改FC层为加法网络,想问一下在梯度上是需要做哪些修改?谢谢!
-
Hi! I have a pre-trained Torch model that was trained on GPU (using CUDA acceleration). I scripted this model with `torch.jit.trace`, and I used the same code provided in one of this repo's [examples]…