Working with Sequences

English Version tensorflow
1

https://d2l.ai/chapter_recurrent-neural-networks/sequence.html

2

In section 8.1.1.1

This leads to models that estimate $x_t$ with $\hat{x}t = P(x_t \mid h{t})$

But going by the figure 8.1.2, isn’t x^_t calculated through a combination of h_t and x_t-1?

3

\hat{x}_t is calculated based on h_t, where β„Ž_𝑑=𝑔(β„Ž_{π‘‘βˆ’1},π‘₯_{π‘‘βˆ’1})

5

Having trouble with
d2l.plot(time, [x], β€˜time’, β€˜x’, xlim=[1, 1000], figsize=(6, 3))
in tensorflow, no error but no ploted image either, what am I doing wrong?