gy910210 / rnn-from-scratch Goto Github PK

@pangolulu

Hi, thank you for a wonderful work!

If you do not mind it, would you please specify the license of this repository?
I hope that people learning more DeepLearning will be able to use this wonderful theme.

Best Regards

Hi,
Why ds = dsv + diff_s
what is the difference between diff_s and ds?
I'am confused.
thank you!

Thanks for the wonderful write up! We can cache the layer output and intermediate to make this implementation fast.

不太明白这个RNN在做什么，X是句子，Y是句子，

X_train shape: (78483,)
y_train shape: (78483,)
x:
SENTENCE_START what are n't you understanding about this ? !
[0, 51, 27, 16, 10, 857, 53, 25, 34, 69]
y:
what are n't you understanding about this ? ! SENTENCE_END
[51, 27, 16, 10, 857, 53, 25, 34, 69, 1]

你好！
首先表示感谢，代码非常清晰易懂，唯一有点疑惑的是bptt的实现方式。
代码中是从前向后遍历T，用的是用t时刻的diff去算t-1 ~ 0 的梯度，是否使用t时刻diff由t+1时刻的梯度决定的方式会比较好一些，这样从后向前遍历T，一个循环就足够了。不过这样的话无法控制梯度传递的时间长度。

Thank you .