Timeline for Training an LSTM to track sine waves
Current License: CC BY-SA 4.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 22, 2018 at 21:10 | history | bumped | CommunityBot | This question has answers that may be good or bad; the system has marked it active so that they can be reviewed. | |
| Jun 22, 2018 at 20:47 | history | bumped | CommunityBot | This question has answers that may be good or bad; the system has marked it active so that they can be reviewed. | |
| May 23, 2018 at 7:57 | comment | added | Lafayette | Yes, I have replaced that with a linear activation, as well as replacing the last layer with a time-distributed dense layer - as can be seen below. | |
| May 22, 2018 at 17:08 | comment | added | kbrose | Your problem may be having tanh in your output layer. You’re restricting the output range that way. I suggest linear (no) activation in final layer. | |
| May 22, 2018 at 14:02 | answer | added | Lafayette | timeline score: 3 | |
| May 22, 2018 at 13:53 | comment | added | Lafayette | I trained it for 100 epochs; Both validation and train loss got "stuck" so to speak after this period (given the task I do not have an accuracy statistic). Default LSTM activations were used, that is, tanh for the "data pathway" with sigmoids controlling the gates. | |
| May 22, 2018 at 13:38 | comment | added | kbrose | How long did you train it for? Did you wait until your validation accuracy started to decrease? What are the default activations for LSTM? | |
| May 22, 2018 at 7:30 | history | edited | Lafayette | CC BY-SA 4.0 |
edited body
|
| May 21, 2018 at 14:50 | history | edited | Lafayette | CC BY-SA 4.0 |
typo correction
|
| May 21, 2018 at 12:28 | history | asked | Lafayette | CC BY-SA 4.0 |