Recurrent Neural Network Implementation from Scratch

English Version mxnet
1

https://d2l.ai/chapter_recurrent-neural-networks/rnn-scratch.html

2

There seems to be a broken reference in section 8.5.3 in the second paragraph " :numref: sec_mlp"

3

Hi @HarryDC, eagle eyes! Can you post an issue on our github? Thanks for contributing!

1 Like
4

Thanks. Now it’s fixed. See comments in https://github.com/d2l-ai/d2l-en/issues/1448

1 Like
5

the second traning (use_random_iter=True) uses the model trained by the first time:
(1)the ppl is from about 2.0
(2)the curve is bumpy

So misleading :sweat_smile:

1 Like
6

I argee with “uses the model trained by the first time”. Thank you for reminding us.
@zhangjiekui
I think use_random_iter=True maybe is a way to converge better when it is hard to converge our model.

7

There is a statement H, = state in function rnn.
I’ve never seen such a usage. What does it mean?
I asked some colleagues around me, but nobody knows.
Can somebody kindly explain it to me?
Many thanks!

8

@ljppro
This statement unpacks an iterable of length one

x, = [2]
>>x
2
9

The second training still uses the same data iterator which is sequential not random. Am I missing something?