Skip to content
Permalink

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.
base repository: VowpalWabbit/vowpal_wabbit
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: master
Choose a base ref
...
head repository: duongnt/vowpal_wabbit
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: master
Choose a head ref
Checking mergeability… Don’t worry, you can still create the pull request.
  • 19 commits
  • 8 files changed
  • 4 contributors

Commits on May 9, 2013

  1. Matrix factorization added.

    Thi-Duong Nguyen committed May 9, 2013

    Unverified

    This commit is not signed, but one or more authors requires that any commit attributed to them is signed.
    Learn about vigilant mode
    Copy the full SHA
    a882314 View commit details

Commits on May 11, 2013

  1. Matrix factorization v.2. Issues: Segmentation fault. Wrong number of… 

    … features. Loss blows up with L2 reg. Loss doesn't look right (negative loss).
    Thi-Duong Nguyen committed May 11, 2013
    Copy the full SHA
    cd19a82 View commit details

Commits on May 12, 2013

  1. Fixed Segmentation fault due to changing all.pairs. Hardcoding parse_… 

    …regressor to initialize all weights and not only the linear weights to random / initial value.
    thiduong committed May 12, 2013
    Copy the full SHA
    e81c82c View commit details

  2. Fixing Indexes and Free Slots for atomics

    Amortized authored and Amortized committed May 12, 2013
    Copy the full SHA
    610b2b5 View commit details

  3. Merge pull request #1 from Amortized/master 

    Fixed hardcoding in mf.cc
    @duongnt
    duongnt committed May 12, 2013
    Copy the full SHA
    3e0bca6 View commit details

  4. Fixed a Bug in finding free slots for atomics indexes

    Amortized authored and Amortized committed May 12, 2013
    Copy the full SHA
    1a94803 View commit details

  5. Merge pull request #2 from Amortized/master 

    Fixed the Bug in Atomics
    @duongnt
    duongnt committed May 12, 2013
    Copy the full SHA
    86e2bd4 View commit details

  6. hardcoding random weights init in parse_regressor.cc

    thiduong committed May 12, 2013
    Copy the full SHA
    09ca799 View commit details

  7. Merge branch 'master' of https://github.com/duongnt/vowpal_wabbit

    thiduong committed May 12, 2013
    Copy the full SHA
    5a545a9 View commit details

Commits on May 14, 2013

  1. Factorization Rewritten with dot products calc by calling base.learn

    Thi-Duong Nguyen committed May 14, 2013
    Copy the full SHA
    32c52af View commit details

Commits on May 18, 2013

  1. MF v2

    Thi-Duong Nguyen committed May 18, 2013
    Copy the full SHA
    be906c9 View commit details

  2. Model Reverted

    Thi-Duong Nguyen committed May 18, 2013
    Copy the full SHA
    cfc6f30 View commit details

  3. Model Reverted

    Thi-Duong Nguyen committed May 18, 2013
    Copy the full SHA
    4c456c3 View commit details

Commits on May 19, 2013

  1. Removing Dead Code

    Amortized authored and Amortized committed May 19, 2013
    Copy the full SHA
    359fb41 View commit details

  2. Merge pull request #3 from Amortized/master 

    Dead Code Removed
    @duongnt
    duongnt committed May 19, 2013
    Copy the full SHA
    2bf881c View commit details

  3. merge head

    thiduong committed May 19, 2013
    Copy the full SHA
    ad959e6 View commit details

  4. Merge branch 'master' of https://github.com/duongnt/vowpal_wabbit

    thiduong committed May 19, 2013
    Copy the full SHA
    9e0961f View commit details

  5. restored parse_regressor.cc

    thiduong committed May 19, 2013
    Copy the full SHA
    170fff6 View commit details

Commits on May 22, 2013

  1. Removed debugging comments

    thiduong committed May 22, 2013
    Copy the full SHA
    95dd97a View commit details
Showing with 1,204 additions and 897 deletions.
  1. +1 −1 Makefile
  2. +1 −1 vowpalwabbit/Makefile.am
  3. +634 −643 vowpalwabbit/gd.cc
  4. +329 −0 vowpalwabbit/mf.cc
  5. +18 −0 vowpalwabbit/mf.h
  6. +8 −4 vowpalwabbit/parse_args.cc
  7. +212 −248 vowpalwabbit/parse_regressor.cc
  8. +1 −0 vowpalwabbit/sparse_dense.h
2 changes: 1 addition & 1 deletion Makefile
View file
Original file line number Diff line number Diff line change
@@ -31,7 +31,7 @@ WARN_FLAGS = -Wall -pedantic
endif

# for normal fast execution.
FLAGS = $(ARCH) $(WARN_FLAGS) $(OPTIM_FLAGS) -D_FILE_OFFSET_BITS=64 -I $(BOOST_INCLUDE) #-DVW_LDA_NO_SSE
FLAGS = $(ARCH) $(WARN_FLAGS) $(OPTIM_FLAGS) -D_FILE_OFFSET_BITS=64 -I $(BOOST_INCLUDE) #-DVW_LDA_NO_SSE -g

# for profiling
#FLAGS = $(ARCH) $(WARN_FLAGS) -O3 -fno-strict-aliasing -ffast-math -D_FILE_OFFSET_BITS=64 -I $(BOOST_INCLUDE) -pg #-DVW_LDA_NO_SSE
2 changes: 1 addition & 1 deletion vowpalwabbit/Makefile.am
View file
Original file line number Diff line number Diff line change
@@ -4,7 +4,7 @@ liballreduce_la_SOURCES = allreduce.cc

bin_PROGRAMS = vw active_interactor

libvw_la_SOURCES = hash.cc global_data.cc io_buf.cc parse_regressor.cc parse_primitives.cc unique_sort.cc cache.cc rand48.cc simple_label.cc oaa.cc ect.cc autolink.cc binary.cc csoaa.cc cb.cc wap.cc beam.cc searn.cc searn_sequencetask.cc parse_example.cc sparse_dense.cc network.cc parse_args.cc accumulate.cc gd.cc lda_core.cc gd_mf.cc bfgs.cc noop.cc example.cc parser.cc loss_functions.cc sender.cc nn.cc
libvw_la_SOURCES = hash.cc global_data.cc io_buf.cc parse_regressor.cc parse_primitives.cc unique_sort.cc cache.cc rand48.cc simple_label.cc oaa.cc ect.cc autolink.cc binary.cc csoaa.cc cb.cc wap.cc beam.cc searn.cc searn_sequencetask.cc parse_example.cc sparse_dense.cc network.cc parse_args.cc accumulate.cc gd.cc lda_core.cc mf.cc gd_mf.cc bfgs.cc noop.cc example.cc parser.cc loss_functions.cc sender.cc nn.cc

# accumulate.cc uses all_reduce
libvw_la_LIBADD = liballreduce.la
Loading