Skip to content
Projects
Groups
Snippets
Help
Loading...
Help
Support
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
S
seminar-breakout
Project overview
Project overview
Details
Activity
Releases
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Issues
0
Issues
0
List
Boards
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Analytics
Analytics
CI / CD
Repository
Value Stream
Wiki
Wiki
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
Shashank Suhas
seminar-breakout
Commits
c223f22318bf31a75bc655307d3765d2a0264443
Switch branch/tag
seminar-breakout
15 Jul, 2017
5 commits
fix build
· c223f223
Yuxin Wu
authored
Jul 14, 2017
c223f223
remove the 'towerp' constant completely. Use custom predict tower names everywhere.
· 4f498c2a
Yuxin Wu
authored
Jul 14, 2017
4f498c2a
fix name scope error
· d6db7efa
Yuxin Wu
authored
Jul 14, 2017
d6db7efa
Use custom name for predictor tower
· 098cbd97
Yuxin Wu
authored
Jul 14, 2017
098cbd97
Reenter the same name_scope when reusing a variable_scope
· ab507874
Yuxin Wu
authored
Jul 14, 2017
ab507874
14 Jul, 2017
11 commits
Fix build
· 79f4760a
Yuxin Wu
authored
Jul 14, 2017
79f4760a
remove PredictorTowerBuilder
· 8a80b036
Yuxin Wu
authored
Jul 14, 2017
8a80b036
let multi-gpu OfflinePredictor use PredictorFactory
· 7699fd9b
Yuxin Wu
authored
Jul 14, 2017
7699fd9b
let DataParallelInference use PredictorFactory
· e1f9cc09
Yuxin Wu
authored
Jul 14, 2017
e1f9cc09
merge Feed/Feedfree InferenceRunner
· 3951aaf7
Yuxin Wu
authored
Jul 14, 2017
3951aaf7
fix setuptools environment marker issue
· 05d1cbe7
Yuxin Wu
authored
Jul 14, 2017
05d1cbe7
rgb option for Saturation and Hue (fix #332)
· ca7bc07f
Yuxin Wu
authored
Jul 14, 2017
ca7bc07f
expose DataFromGenerator (#333)
· f17fab8e
Yuxin Wu
authored
Jul 14, 2017
f17fab8e
InferenceRunnerBase uses PredictorFactory to build graph
· fb501e66
Yuxin Wu
authored
Jul 14, 2017
fb501e66
PredictorFactory build tower by itself.
· 4ee1e735
Yuxin Wu
authored
Jul 13, 2017
4ee1e735
move PredictorFactory to graph_builder
· e839c50d
Yuxin Wu
authored
Jul 13, 2017
e839c50d
13 Jul, 2017
13 commits
remove the many levels of Trainer herarchy
· efe3dfb5
Yuxin Wu
authored
Jul 13, 2017
efe3dfb5
Make SimpleTrainer support InputSource
· 079eb3a9
Yuxin Wu
authored
Jul 13, 2017
079eb3a9
move get_cost_and_grad to ModelDesc
· 20ee19bc
Yuxin Wu
authored
Jul 13, 2017
20ee19bc
use tf.get_name_scope to replace get_name_scope_name
· 696c7830
Yuxin Wu
authored
Jul 13, 2017
696c7830
work around device name problem with colocation (#329)
· 44f603c0
Yuxin Wu
authored
Jul 13, 2017
44f603c0
Add a ModelDescBase without single-cost assumptions
· 61c113b8
Yuxin Wu
authored
Jul 13, 2017
61c113b8
fix import for 'git pull' users
· 7240f877
Yuxin Wu
authored
Jul 13, 2017
7240f877
Move related stuff to graph_builder.
· a8cb7e33
Yuxin Wu
authored
Jul 13, 2017
a8cb7e33
Get tower-owned variables through TowerContext
· 3f5b9d51
Yuxin Wu
authored
Jul 13, 2017
3f5b9d51
add back build_train_tower for compatibility
· a397cebc
Yuxin Wu
authored
Jul 12, 2017
a397cebc
update FAQ about freezing variables
· 3e1f0e53
Yuxin Wu
authored
Jul 12, 2017
3e1f0e53
fix typo in freeze_variables
· cc2f9c12
Yuxin Wu
authored
Jul 12, 2017
cc2f9c12
remove vs_strategy from tower. Use vs_name in a cleaner way.
· a77cc508
Yuxin Wu
authored
Jul 12, 2017
a77cc508
12 Jul, 2017
6 commits
Simplify the connection between ModelDesc and InputSource
· 4d2a7b4c
Yuxin Wu
authored
Jul 12, 2017
4d2a7b4c
make DataParallelFeedInput runnable again
· e46e6bca
Yuxin Wu
authored
Jul 12, 2017
e46e6bca
Use hooks for feed_dict in FeedInput
· d9b96535
Yuxin Wu
authored
Jul 12, 2017
d9b96535
make InputSource subclass methods private
· 1e7fa5f9
Yuxin Wu
authored
Jul 11, 2017
1e7fa5f9
dynamically generate the remapped class
· ef9e27a6
Yuxin Wu
authored
Jul 11, 2017
ef9e27a6
Allow remapping on every InputSource, therefore remove the old 'names' options
· adf51f22
Yuxin Wu
authored
Jul 11, 2017
adf51f22
11 Jul, 2017
5 commits
Improve the use of size() of InputSource
· 20d7fe7f
Yuxin Wu
authored
Jul 11, 2017
20d7fe7f
replace setup_training by get_callbacks in InputSource
· e2f9798a
Yuxin Wu
authored
Jul 11, 2017
e2f9798a
Pass InputDesc instead of ModelDesc to InputSource.setup()
· 14c564cc
Yuxin Wu
authored
Jul 10, 2017
14c564cc
bugfix in InputDesc initialization
· 8b879cb9
Yuxin Wu
authored
Jul 10, 2017
8b879cb9
use InputDesc inside input_source, instead of ModelDesc
· a266459e
Yuxin Wu
authored
Jul 10, 2017
a266459e
