-
Notifications
You must be signed in to change notification settings - Fork 18.6k
InverseMVNLayer #1895
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Closed
Closed
InverseMVNLayer #1895
Conversation
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
* A sample code was added. * `slice_dim` and `slice_point` attributes were explained.
[docs] brief explanation of SLICE layer's attributes
Correct 'epochs' to 'iterations'
fdc0e30 to
c007a43
Compare
Next: release candidater
fix Imagenet example path
set the right rpath for tools and examples respectively thanks for the report @mees!
[build] fix dynamic linking of tools
Contributor
Author
|
Oops this is broken for across_channels=true. I will fix and update as am able. Or close. |
… was overwritten with symlink created at build time and installed with install(DIRECTORY ...)
… systems). This commit specifies Python2 with which cpp_lint.py works :-)
[cmake] fix install rpath for pycaffe
c007a43 to
3c4295f
Compare
APPLE was misspelled in Line 27
fixes: cpp_lint.py fails silently with Python3
Check caffe tool runs
Making python3 work with cmake and the new python wrapper
Commands, such as $(error ...), are not allowed to be indented with tabs outside of targets, throwing an error instead of outputting the actual error. The solution is to use innocuous spaces instead. Ideally, spaces should be used everywhere outside targets, but since make does not mind it if variable assignments are tab-indented outside targets, a complete overhaul is not necessary. However, if more errors are added, it might make more sense to be consistent. Also, make will already add a period so I removed it.
fix accelerate / veclib path for OS X 10.10
Replaced illegal tab in Makefile with spaces.
Decoding the datum before feeding it into the reshaping data layer
9002c93 to
ce22509
Compare
ce22509 to
0348894
Compare
Closed
Contributor
Author
|
Replaced with 1979. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
This PR extends the MVNLayer to allow the mean and variance blobs to be exported as top blobs. It adds a new layer type InverseMVNLayer which takes the mean and variance as bottom blobs, and performs the inverse operation (adding the mean back, and denormalizing for variance). A use case for this is an autoencoder that feeds input into the MVNLayer, and generates the output from the InverseMVNLayer, with autoencoding layers in-between.
There was also a problem with the MVNLayer that was fixed: if it is given input that has exactly zero variance (e.g. a solid color RGB image, with across_channels=false), it computes the variance as E(X^2) - (EX)^2, but the result isn't usually exactly zero, but has small negative values due to floating point resolution. The subsequent square root operation then produces NaN. This PR also fixes this issue.