peterjc123
peterjc123
With https://github.com/alibaba/TinyNeuralNetwork/commit/e35ef92faba445830bb6156c916b1e838801c07e, the weights generated are within the range [-127, 127]. Would you please try again? BTW, I'm just curious how does the model perform during QAT training?
If things still don't work out with that patch, you may try bisecting the model, which should be fairly easy since you have the model descriptive script there.
Suppose you have the following model description file, you may return the intermediate tensors (e.g. `a` or `b`) instead of the original ones, so that you could figure out which...
> My experiment shows no -128 weights anymore for asymmetric per-channel case. Not sure if this is what you said "work out with that patch". Similar to former experiments, asymmetric...
You can just do it on a trained quantized model. Just load the state dict with `strict=False` and it will be fine.
@liamsun2019 Do you have a DingTalk account so that you can join our discussion group? This thread will grow too lengthy if you answer your questions here.
因为目前不支持count_include_pad=False的avgpool2d，如果是same padding的情况下实现会比较简单，但是如果是其他情况下，需要额外添加一个mask乘的操作，会有点麻烦
已经添加了使用乘mask的等效实现的count_include_pad=False的avgpool2d，可以重新拉一下试试
> 对于这些warning如果不修改的话会导致转换后的模型输出不正确是吗？ 不会，图1是pytorch使用floordiv的warning，图2只是int64自动下转成int32的例行打印。两者都跟精度没关系。 > ok 重新拉去后这个问题就消失了，作者太给力了然后我往下运行，出现了：  基本上是光流和transformer相关的，这些算子作者有继续支持的计划吗？ 不知道哪些是常量操作，如果是常量操作可以直接跳过。方便提供下模型定义不？
按难度排个序吧 ## 简单 new_zeros， new_ones 直接跳过，没啥可说的 fill_ 有直接对应 ## 一般 roll 可以用strided_slice配合cat实现 img2col， col2im 可以用arange后的数据跑一遍，然后用gather_nd实现 ## 困难 grid_sample 貌似没有直接对应