deep-learning-from-scratch-3

deep-learning-from-scratch-3 copied to clipboard

```py class Variable: # ... def backward(self, retain_grad=False): if self.grad is None: self.grad = np.ones_like(self.data) funcs = [] seen_set = set() def add_func(f): if f not in seen_set: funcs.append(f) seen_set.add(f)...

LeiLiGithub

Excuse me When I coding with Colab and call DeZero,it show AttributeError  How do I deal with YHX

SonRule

sonald

As described in the title. I think it should be a typo by the author.

wsbao

In the implementation of MatMul, the used dot does not conform to the behavior of matrix multiplication in Torch, especially it is not suitable for cases with dimensions higher than...

owenliang