upgrading matlab interface by adding new methods to matlab class Net #4996
Conversation
This file contains 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
Currently parse_log.py skips all non timestamped lines only once. When resuming a solver and appending to the same log file, it creates more non timestamped log lines. This change allows the script to silently skip those lines.
sig-ce learns to ignore by zeroing out the loss/diff at targets equal to the configured `ignore_label`. n.b. as of now the loss/diff are not properly normalized when there are ignored targets. sig-ce loss should adopt the same normalization options as softmax loss.
sig-ce loss handles all the same normalizations as the softmax loss; refer to #3296 for more detail. this preserves the default normalization for sig-ce loss: batch size.
Sigmoid Cross-Entropy Loss: ignore selected targets by `ignore_label`
Support solver resumes in parse_log.py
Avoids missing return values during build
the solver checks its proto type (SolverParameter.type) on instantiation: - if the proto type is unspecified it's set according to the class type `Solver::type()` - if the proto type and class type conflict, the solver dies loudly this helps avoid accidental instantiation of a different solver type than intended when the solver def and class differ. guaranteed type information in the SolverParameter will simplify multi-solver coordination too.
OS X: build with latest SDK by default
Add missing spaces besides equal signs in batch_norm_layer.cpp
Solver: check and set type to reconcile class and proto type
…into matlab_interface
|
sorry for the mess, i have opened a new pull request with clean commit : |
|
@wadefelix Anyway, this PR is depcrecated as i replaced it with this one : |
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.
I took the updated matlab interface from ShaoqingRen/caffe, which adds many new methods compared to the current interface master branch.
These changes were originally done to support faster_rcnn project but can be added to caffe master branch as independent update.
After accepting this PR, and merging #4163 , the projects faster_rcnn and py-faster-rcnn can be used alongside caffe master branch without having to use different caffe versions.