Commit 7f0e442

authored

循环的最后一个元素和下一次循环的第一个元素时间不连续

steps = np.linspace(start, end, TIME_STEP, dtype=np.float32) 其中默认参数endpoint=True,也就是说这一次循环的最后一个元素和下一次循环的第一个元素是一致的,按照RNN传递的state的特点,应该保证输入是关于时间连续的,所以不应该有相邻的时间是一致的情况,这可能不是想要的,所以我认为这里是一个小bug。

1 parent 51c6c66 commit 7f0e442Copy full SHA for 7f0e442

1 file changed

Lines changed: 1 addition & 1 deletion

File tree

tutorial-contents
- 403_RNN_regressor.py

`‎tutorial-contents/403_RNN_regressor.py‎`

Lines changed: 1 addition & 1 deletion

Original file line number	Diff line number	Diff line change
`@@ -77,7 +77,7 @@ def forward(self, x, h_state):`
`77`	`77`	`for step in range(100):`
`78`	`78`	`start, end = step * np.pi, (step+1)*np.pi # time range`
`79`	`79`	`# use sin predicts cos`
`80`		`- steps = np.linspace(start, end, TIME_STEP, dtype=np.float32) # float32 for converting torch FloatTensor`
	`80`	`+ steps = np.linspace(start, end, TIME_STEP, dtype=np.float32, endpoint=False) # float32 for converting torch FloatTensor`
`81`	`81`	`x_np = np.sin(steps)`
`82`	`82`	`y_np = np.cos(steps)`
`83`	`83`

0 commit comments

Comments

(0)

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Commit 7f0e442

File tree

`‎tutorial-contents/403_RNN_regressor.py‎`

0 commit comments