Timeline for answer to Why does Keras LSTM batch size used for prediction have to be the same as fitting batch size? by Hai-Anh Trinh
Current License: CC BY-SA 3.0
Post Revisions
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 25, 2020 at 2:33 | comment | added | user11747064 | @Hai-AnhTrinh Do you have link to blog post to implementation of rnn using tensorflow computation graph? | |
| Nov 14, 2019 at 10:50 | comment | added | Voy |
What makes you think author meant num_steps not batch_size? Whilst your answer is related and somewhat useful, I think you're making a wrong assumption. If I'm not misreading something, they clearly speak about the various batch_size, not the num_step (which they refer to as sl). Other answers seem to provide better solutions to the specific problem author describes. Still, thanks for putting in the effort to write such a detailed answer!
|
|
| S Apr 13, 2018 at 14:09 | history | suggested | Robert Pollak | CC BY-SA 3.0 |
Fixed a link.
|
| Apr 13, 2018 at 12:55 | review | Suggested edits | |||
| S Apr 13, 2018 at 14:09 | |||||
| Feb 27, 2018 at 6:29 | vote | accept | DanielSon | ||
| Dec 27, 2017 at 11:44 | comment | added | Tomasz Sętkowski |
Here is a complete short example on how to restore tensorflow model with LSTM cells when using different batch_size and num_steps than when it was trained.
|
|
| May 7, 2017 at 5:49 | history | bounty awarded | DanielSon | ||
| May 3, 2017 at 22:51 | history | edited | Hai-Anh Trinh | CC BY-SA 3.0 |
deleted 1 character in body; added 29 characters in body
|
| May 3, 2017 at 22:50 | comment | added | Hai-Anh Trinh | For RNN models, the inputs are usually 3D tensors (batch_size, num_steps, num_features) meaning you train on multiple sequences in the same batch, each sequence have length num_steps, each num steps have num_features. | |
| May 3, 2017 at 22:10 | comment | added | DanielSon |
Hey thanks for a really good reply. Can you explain the difference between batch_size and num_steps again? I've never actually used or seen num_step and thought batch_size was just how many windows are trained on at the same time. What is the difference between a sequence and a time step?
|
|
| May 3, 2017 at 22:03 | history | answered | Hai-Anh Trinh | CC BY-SA 3.0 |