Wang Jiahao

I have noticed that the quantization mechanism is quantize(w, ak, c) = round(clamp(w, c)/s) × s. It seems like you have use the "Training with simulated quantization" in paper "Quantization...

[Feature] Support InstructBLIP

2

Add InstructBLIP Implementation

Question about visualization

1

Thanks Zechun! Very interesting work! Could you provide the code about Figure. 1, i.e. "the actual optimization landscape from real-valued and binary networks with the same architecture" Thanks a lot!

Hi, author. Thanks for the nice work! 有个问题想请教下作者。看到HardBinaryConv这段代码表示一个二值卷积，但是打印了权重，发现大概是0.005,0.006等等这样的数。看到他代码给二值化卷积torch.sign(real_weights)前面乘了个scaling_factor，这样好像就又变成32位浮点数了。我不知道我这样的理解对不对

k = x.shape[1]

3

Hi! A line of code in PIC.py is: k = x.shape[1]. I wondor what does k means here? Maybe image channels? But why is each channel handled separately?

Hi, author! We notice your another wonderful work published at CVPR2020: ' Memory-Efficient Hierarchical Neural Architecture Search for Image Denoising', which give us a lot inspiration. Thanks for your wonderful...