Add augmentation option to train_raw_dnn.py#1585
Merged
danpovey merged 6 commits intokaldi-asr:kaldi_52from Apr 27, 2017
Merged
Add augmentation option to train_raw_dnn.py#1585danpovey merged 6 commits intokaldi-asr:kaldi_52from
danpovey merged 6 commits intokaldi-asr:kaldi_52from
Conversation
Contributor
|
I think it's because I previously committed using 'squash and merge', so those exact commits are not showing up. If you were to rebase against kaldi_52, the issue would probably go away. But I can merge anyway. |
danpovey
reviewed
Apr 26, 2017
| raw_model_string, egs_dir, | ||
| momentum, max_param_change, | ||
| shuffle_buffer_size, minibatch_size_str, | ||
| image_augmentation_opts, |
Contributor
There was a problem hiding this comment.
You need to have this default to None; otherwise this function would crash when called from elsewhere.
Contributor
Author
There was a problem hiding this comment.
OK. I guess the same goes with train_new_models, right?
Contributor
|
I think I commented on the wrong function. It's train_one_iteration where
it's important-- that's called from elsewhere. But I think
train_new_models is only called from there; grep around, though, to check.
…On Wed, Apr 26, 2017 at 6:00 PM, Hossein Hadian ***@***.***> wrote:
***@***.**** commented on this pull request.
------------------------------
In egs/wsj/s5/steps/libs/nnet3/train/frame_level_objf/common.py
<#1585 (comment)>:
> @@ -27,6 +27,7 @@ def train_new_models(dir, iter, srand, num_jobs,
raw_model_string, egs_dir,
momentum, max_param_change,
shuffle_buffer_size, minibatch_size_str,
+ image_augmentation_opts,
OK. I guess the same goes with train_new_models, right?
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#1585 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/ADJVu6_XAW-0t0YcSgVEaNTpqum-zCC9ks5rz76KgaJpZM4NJeYp>
.
|
Contributor
Author
|
|
Skaiste
pushed a commit
to Skaiste/idlak
that referenced
this pull request
Sep 26, 2018
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
It seems github is acting odd. It includes my previous commits everytime but merges them w/o a problem, so we are fine I guess.
I tried augmentation (only flipping and vertical shift, I'll fix horizontal shift after I'm done with trying dropout) with the same config as 1a, and I got 79.5% final valid accuracy.