| Age | Commit message (Collapse) | Author | Files | Lines | |
|---|---|---|---|---|---|
| 2016-02-29 | NetSpec: allow setting blob names by string | Anatoly Baksheev | 1 | -0/+6 | |
| 2016-02-29 | Use 'six' library to ensure python3 compliance. | Thibault Deregnaucourt | 1 | -10/+12 | |
| Use '//' instead of '/' for entire division. | |||||
| 2016-02-23 | removing all references to Blob.num property (that assumes Blob is 4D). ↵ | shai | 1 | -3/+3 | |
| Replacing it with accessing Blob.shape[0] - for Blobs with num_axes() != 4 | |||||
| 2016-01-26 | Merge pull request #2144 from tishibas/load_image-improved | Jeff Donahue | 1 | -1/+1 | |
| improved to load RGB image as grayscale image | |||||
| 2016-01-26 | Merge pull request #3581 from Austriker/draw-python3 | Jeff Donahue | 1 | -1/+10 | |
| Updated draw.py to make it work under Python 3 | |||||
| 2016-01-25 | Updated import to make it work with pydotplus | Hugo Serrat | 1 | -1/+10 | |
| 2016-01-22 | Version 1.0.0-rc3 | Luke Yeager | 2 | -0/+4 | |
| 2016-01-21 | add register Net and Solver | biluochun | 1 | -0/+2 | |
| 2016-01-20 | fixbug #issues/3494 No to_python (by-value) converter found for C++ type: ↵ | biluochun | 1 | -0/+1 | |
| boost::shared_ptr<caffe::Blob<float> > | |||||
| 2016-01-04 | Exposing layer top and bottom names to python | philkr | 2 | -0/+22 | |
| 2015-12-01 | dismantle layer headers | Evan Shelhamer | 1 | -1/+2 | |
| No more monolithic includes: split layers into their own headers for modular inclusion and build. | |||||
| 2015-11-18 | Skip python layer tests if WITH_PYTHON_LAYER unset | Luke Yeager | 2 | -0/+4 | |
| 2015-11-14 | Fix outs and diffs being overwritten in forward_backward_all. | Alex Lee | 1 | -2/+2 | |
| 2015-11-05 | Correct transposition & channel_swap in deprocess | ernest-tg | 1 | -2/+2 | |
| The deprocess( ) function should invert the preprocess( ) function, however it only worked when the permutation of your channel_swap is of order 2 and the permutation of your transpose were of order 3. This is usually the case, which is why this bug went unnoticed for a long time. To reproduce it (on former version), try to preprocess and then deprocess with transformer.set_transpose('data', (0,2,1)) (or (1,0,2) or (2,1,0)) Or with transformer.set_channel_swap('data', (2,0,1)) (or (1,2,0) ) Indeed, we had L152 (in preprocess) caffe_in = caffe_in[channel_swap, :, :] L181 (in deprocess) decaf_in = decaf_in[channel_swap, :, :] So we applied [channel_swap,:,:] twice to the initial data => not always the identity L154 (in preprocess) caffe_in = caffe_in.transpose(transpose) L183 (in deprocess) decaf_in = decaf_in.transpose([transpose[t] for t in transpose]) The transposition [transpose[t] for t in transpose] is (tranpsose)² so we applied transpose[t] three times which is not always the identity. | |||||
| 2015-11-03 | fix detect.py (invalid model path) | René Scheibe | 1 | -1/+1 | |
| 2015-10-30 | Merge pull request #3082 from gustavla/pycaffe-snapshot | Evan Shelhamer | 2 | -2/+12 | |
| Expose `Solver::Snapshot` to pycaffe | |||||
| 2015-10-22 | diff.ndim != 4 is outdated | Rodrigo Benenson | 1 | -2/+0 | |
| this code seems not to apply to the caffe head. ``` if diff.ndim != 4: raise Exception('{} diff is not 4-d'.format(top)) ``` | |||||
| 2015-10-16 | Add automatic upgrade for solver type | Ronghang Hu | 1 | -2/+2 | |
| 2015-10-16 | Split solver code into one file per solver class | Ronghang Hu | 1 | -0/+1 | |
| 2015-10-16 | Merge pull request #3190 from DrSleep/master | Evan Shelhamer | 1 | -3/+3 | |
| Fix drawing problems with repeated convolution -- close #3201 | |||||
| 2015-10-15 | Allow old-style shape in blobproto_to_array | Luke Yeager | 2 | -2/+50 | |
| Fixes #3199 Bug introduced in #3170 | |||||
| 2015-10-14 | Fixed drawing problems with repeated convolution | Vladimir | 1 | -3/+3 | |
| 2015-10-09 | Remove the 4D constraint of blobproto IO in python | sh1r0 | 1 | -8/+4 | |
| 2015-10-05 | Add pycaffe test for solver.snapshot() | Gustav Larsson | 1 | -1/+10 | |
| 2015-09-29 | Fix a typo | Yang Song | 1 | -1/+1 | |
| Fix a typo in the message. | |||||
| 2015-09-17 | Expose `Snapshot` to pycaffe | Gustav Larsson | 1 | -1/+2 | |
| - Solver::Snapshot is made public - It is also added as `snapshot` to pycaffe Addressing #3077 | |||||
| 2015-09-17 | Separate IO dependencies | Tea | 1 | -0/+1 | |
| 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. | |||||
| 2015-09-15 | removed bug in caffe.io.resize_image when applied to Nd images | Mohamed Omran | 1 | -1/+1 | |
| 2015-09-04 | Merge pull request #3024 from danielgordon10/python-solver-fix | Evan Shelhamer | 3 | -2/+12 | |
| [pycaffe] expose all solvers for direct instantiation (although note get_solver) | |||||
| 2015-09-04 | enabling the alternate solvers to be accessed by the python interface | Daniel Gordon | 3 | -2/+12 | |
| 2015-09-03 | NetSpec: don't require lists to specify single-element repeated fields | Jeff Donahue | 2 | -4/+9 | |
| 2015-09-01 | Cleanup: Fixup capitalisation of Caffe_POSTFIX. | Darren Garvey | 1 | -1/+1 | |
| Replaces CAffe_POSTFIX -> Caffe_POSTFIX. | |||||
| 2015-08-26 | No need to squeeze the output of the network | Marco Castelluccio | 1 | -1/+1 | |
| 2015-08-26 | Merge pull request #2944 from philkr/python_layer_param | Jon Long | 2 | -1/+71 | |
| Give the python layer parameter/weight blobs. | |||||
| 2015-08-26 | Draw Deconvolution layers like Convolution layers | Jonas Maaskola | 1 | -3/+3 | |
| 2015-08-25 | Python parameter test added | philkr | 2 | -2/+60 | |
| 2015-08-25 | Allow the python layer have weight/parameter blobs. | philkr | 1 | -2/+14 | |
| 2015-08-20 | Merge pull request #2930 from lukeyeager/pycaffe-layer_type_list | Jeff Donahue | 3 | -1/+13 | |
| Expose LayerFactory::LayerTypeList in pycaffe | |||||
| 2015-08-20 | Expose LayerFactory::LayerTypeList in pycaffe | Luke Yeager | 3 | -1/+13 | |
| Useful for validating NetParameters without crashing on SIGABRT | |||||
| 2015-08-14 | Exposing blob loss weight to python | philkr | 2 | -0/+14 | |
| 2015-08-10 | Update net_spec.py | Russell Stewart | 1 | -1/+1 | |
| 2015-08-07 | Merge pull request #2813 from longjon/net-spec-imp | Evan Shelhamer | 2 | -13/+38 | |
| Python net spec cleanups and top-less layers | |||||
| 2015-08-07 | Merge pull request #2877 from jeffdonahue/pycaffe-shape-accessor | Jon Long | 1 | -0/+5 | |
| [pycaffe] add Blob shape accessor | |||||
| 2015-08-07 | pycaffe: add shape accessor | Jeff Donahue | 1 | -0/+5 | |
| 2015-08-06 | PythonLayer takes parameters by string | Takuya Narihira | 1 | -0/+59 | |
| 2015-08-06 | [pytest] open exception file with mode for python3 | Evan Shelhamer | 1 | -1/+1 | |
| 2015-08-06 | Merge pull request #2462 from longjon/correct-python-exceptions | Evan Shelhamer | 1 | -0/+22 | |
| Handle Python layer exceptions correctly | |||||
| 2015-07-23 | [pytest] simple test of top-less layers | Jonathan L Long | 1 | -0/+15 | |
| 2015-07-23 | [pycaffe] net spec layers can have ntop=0 | Jonathan L Long | 1 | -1/+3 | |
| In this case, the Function is returned instead of a Top, which can be assigned a name if desired. Thanks @philkr for an earlier implementation of this. | |||||
| 2015-07-23 | [pycaffe] allow layers to have names different from their first tops | Jonathan L Long | 1 | -3/+11 | |
| Previously, net spec only allowed names to be assigned to Tops, giving layers the names of their first tops. Now, names can be assigned to Functions, which become layer names in serialization. Unnamed Functions still get named after their first top, if present, or autogenerated, if not. (This will allow top-less layers in a natural way.) | |||||
 |
index : platform/upstream/caffeonacl | |
| Domain: Machine Learning / ML Framework; Licenses: BSD-2-Clause; | Inki Dae <inki.dae@samsung.com> |
| summaryrefslogtreecommitdiff |