| Age | Commit message (Collapse) | Author | Files | Lines | |
|---|---|---|---|---|---|
| 2018-01-31 | v0.5.0 | huifang | 1 | -0/+213 | |
| 2017-08-26 | add support acl batch normal,direct conv, local connect, concat layers | honggui | 7 | -0/+0 | |
| 2017-02-16 | Merge pull request #5074 from garion9013/master | Evan Shelhamer | 1 | -12/+5 | |
| Fix plot_training_log.py.example | |||||
| 2017-01-24 | parse_log.py was not using --verbose argument | Ken Schutte | 1 | -1/+1 | |
| 2017-01-18 | Merge pull request #5142 from lvchaxj/master | Evan Shelhamer | 2 | -0/+15 | |
| Fix parse_log tool for negative time duration if datetime across year boundary | |||||
| 2017-01-06 | Switched multi-GPU to NCCL | Cyprien Noel | 1 | -3/+8 | |
| 2016-12-31 | Fix parse_log.py and parse_log.sh for negative time duration if datetime in ↵ | Xiaojie Deng | 2 | -0/+15 | |
| log across year boundary | |||||
| 2016-12-08 | fix wrongly used marker hash | Young H. Oh | 1 | -12/+5 | |
| 2016-11-27 | fix many typos by using codespell | Zhou Mo | 1 | -1/+1 | |
| 2016-11-14 | support solver resumes in parse_log.py | davidbrai | 1 | -2/+7 | |
| 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. | |||||
| 2016-07-11 | Update parse_log.py | Alessandro Giusti | 1 | -4/+1 | |
| Aligned output description in docstring with actual output returned by parse_log | |||||
| 2016-06-03 | Add phase, level and stages to tools/caffe | Luke Yeager | 1 | -2/+37 | |
| Adds command-line flags for phase, level and stage train -- override level and stages for test_state from solver test -- set level and stages time -- set phase, level and stages | |||||
| 2016-06-02 | Fixing a typo | Chuck Cho | 1 | -1/+1 | |
| 2016-05-18 | Merge pull request #4059 from crazytan/master | Jon Long | 1 | -3/+6 | |
| handle spaces in image file names Thanks @bchu for an earlier implementation. | |||||
| 2016-05-17 | handle image names with spaces | crazytan | 1 | -3/+6 | |
| 2016-05-04 | Fix typo (indecies->indices) | Achal Dave | 1 | -2/+2 | |
| 2016-05-04 | Reformat to fit in 79 columns | Achal Dave | 1 | -2/+4 | |
| 2016-05-04 | Remove trailing spaces | Achal Dave | 1 | -5/+5 | |
| 2016-05-04 | Exit on error and report argument error details. | Achal Dave | 1 | -3/+5 | |
| The statement 'exit' has no effect in Python scripts. Use 'sys.exit()' instead. | |||||
| 2016-04-12 | Merge pull request #3982 from mnogu/fix-typo-model-option | Evan Shelhamer | 1 | -1/+1 | |
| [docs] typo in help text for "-model" option | |||||
| 2016-04-12 | Fix typo in help text for "-model" option | Muneyuki Noguchi | 1 | -1/+1 | |
| 2016-03-07 | Merge pull request #3292 from dgolden1/log-parse-no-test | Evan Shelhamer | 1 | -0/+5 | |
| parse_log.py bugfix: don't attempt to write CSV if there are no lines to write | |||||
| 2016-03-05 | split p2psync::run() | Jun Shi | 1 | -1/+1 | |
| 2016-02-29 | check all net upgrade conditions | Evan Shelhamer | 1 | -5/+0 | |
| check all conditions all the time; V0 -> V1 and V1 -> V2 do not suffice. | |||||
| 2016-02-29 | output all logging from upgrade net tools | Evan Shelhamer | 3 | -4/+7 | |
| output info, warnings, and errors for fuller description of the upgrade | |||||
| 2016-02-25 | drop Net inputs + Forward with bottoms | Evan Shelhamer | 2 | -5/+3 | |
| Drop special cases for `input` fields, the `Net` input members, and the `Net` interface for Forward with bottoms along with Forward() / ForwardPrefilled() distinction. | |||||
| 2016-02-02 | Nicely prints GPU names | Sergei Nikolaev | 1 | -1/+12 | |
| 2016-01-22 | Version 1.0.0-rc3 | Luke Yeager | 1 | -0/+3 | |
| 2015-12-28 | remove extra space before + | Fisher Yu | 1 | -1/+1 | |
| 2015-12-08 | Merge pull request #3090 from longjon/summarize-tool | Yangqing Jia | 1 | -0/+140 | |
| A Python script for at-a-glance net summary | |||||
| 2015-12-04 | Correct type of device_id; disambiguate shared_ptr | T.E.A de Souza | 1 | -9/+8 | |
| 2015-12-01 | dismantle layer headers | Evan Shelhamer | 1 | -1/+0 | |
| No more monolithic includes: split layers into their own headers for modular inclusion and build. | |||||
| 2015-11-26 | replace snprintf with a C++98 equivalent | Tea | 2 | -12/+7 | |
| 2015-11-05 | Don't attempt to write CSV if there are no lines to write | Daniel Golden | 1 | -0/+5 | |
| This can happen if, e.g., testing never occurs in the log | |||||
| 2015-10-30 | Update plot_training_log.py.example | Kai Li | 1 | -1/+1 | |
| I find there is no plot_log.sh file | |||||
| 2015-10-16 | Add automatic upgrade for solver type | Ronghang Hu | 2 | -1/+51 | |
| 2015-10-16 | Change solver type to string and provide solver registry | Ronghang Hu | 1 | -1/+1 | |
| 2015-09-25 | Fix parse_log.sh against "prefetch queue empty" messages | Dmytro Mishkin | 1 | -1/+6 | |
| 2015-09-19 | [tools] add Python script for at-a-glance prototxt summary | Jonathan L Long | 1 | -0/+140 | |
| 2015-09-17 | Merge pull request #3074 from ronghanghu/show-use-cpu | Ronghang Hu | 1 | -0/+1 | |
| Get back 'USE CPU' print for caffe train | |||||
| 2015-09-17 | Separate IO dependencies | Tea | 2 | -0/+8 | |
| OpenCV, LMDB, LevelDB and Snappy are made optional via switches (USE_OPENCV, USE_LMDB, USE_LEVELDB) available for Make and CMake builds. Since Snappy is a LevelDB dependency, its use is determined by USE_LEVELDB. HDF5 is left bundled because it is used for serializing weights and solverstates. | |||||
| 2015-09-16 | Get back 'USE CPU' print for caffe train | Ronghang Hu | 1 | -0/+1 | |
| 2015-09-04 | Update extract_features.cpp | Lumin Zhou | 1 | -1/+1 | |
| 2015-09-01 | Show output from convert_imageset tool | Luke Yeager | 1 | -2/+4 | |
| 2015-08-22 | Add signal handler and early exit/snapshot to Solver. | J Yegerlehner | 1 | -1/+31 | |
| Add signal handler and early exit/snapshot to Solver. Add signal handler and early exit/snapshot to Solver. Also check for exit and snapshot when testing. Skip running test after early exit. Fix more lint. Rebase on master. Finish rebase on master. Fixups per review comments. Redress review comments. Lint. Correct error message wording. | |||||
| 2015-08-09 | Multi-GPU | Cyprien Noel | 1 | -31/+80 | |
| - Parallelize batches among GPUs and tree-reduce the gradients - The effective batch size scales with the number of devices - Batch size is multiplied by the number of devices - Split batches between GPUs, and tree-reduce the gradients - Detect machine topology (twin-GPU boards, P2P connectivity) - Track device in syncedmem (thanks @thatguymike) - Insert a callback in the solver for minimal code change - Accept list for gpu flag of caffe tool, e.g. '-gpu 0,1' or '-gpu all'. Run on default GPU if no ID given. - Add multi-GPU solver test - Deterministic architecture for reproducible runs | |||||
| 2015-08-07 | Merge pull request #2634 from mlopezantequera/patch-2 | Jeff Donahue | 1 | -1/+1 | |
| Update parse_log.py | |||||
| 2015-08-06 | [pycaffe,build] include Python first in caffe tool | Evan Shelhamer | 1 | -5/+5 | |
| 2015-08-06 | Merge pull request #2462 from longjon/correct-python-exceptions | Evan Shelhamer | 1 | -1/+15 | |
| Handle Python layer exceptions correctly | |||||
| 2015-06-22 | Update parse_log.py | Manuel | 1 | -1/+1 | |
| Correct parsing (exponential notation learning rates were not being interpreted correctly) | |||||
 |
index : platform/upstream/caffeonacl | |
| Domain: Machine Learning / ML Framework; Licenses: BSD-2-Clause; | Inki Dae <inki.dae@samsung.com> |
| summaryrefslogtreecommitdiff |