microsoft
/
caffe
зеркало из https://github.com/microsoft/caffe.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность
3 644 коммитов 3 Ветки 11 Теги 51 MiB
Граф коммитов

309 Коммитов

Автор SHA1 Сообщение Дата
Evan Shelhamer cb6b09b46b [pycaffe] import newline cleanup 2015-02-18 20:46:41 -08:00
philkr d1238e14fa Changing CMAKE_SOURCE/BINARY_DIR to PROJECT_SOURCE/BINARY_DIR 2015-02-18 09:27:13 -08:00
Evan Shelhamer d2934aee4c [pycaffe] give phase to Net 
- expose `caffe.{TRAIN,TEST}` constants
- instantiate `caffe.Net`s with phase
- drop singleton phase interface `caffe.set_phase_{train,test}`
 2015-02-17 11:35:51 -08:00
Evan Shelhamer 9b38744f22 switch to V2 proto definitions for pytest 2015-02-16 22:47:30 -08:00
Jonathan L Long 3c94d96c08 [pytest] basic test of Python layer 2015-02-16 22:47:30 -08:00
Jonathan L Long da044c9b84 [pytest] test that get_solver runs 2015-02-16 22:47:30 -08:00
Jonathan L Long d94f107ad9 [pycaffe] allow Layer to be extended from Python 
This is done by adding PythonLayer as a boost::python HeldType.
 2015-02-16 22:47:30 -08:00
Jonathan L Long 86646f77be [pycaffe] re-expose SGDSolver, and expose other solvers 2015-02-16 22:46:13 -08:00
Jonathan L Long 2660977b50 [pycaffe] re-expose Layer 2015-02-16 22:46:13 -08:00
Jonathan L Long 8a510f7f14 [pycaffe] re-expose Blob 2015-02-16 22:46:13 -08:00
Jonathan L Long 74a584ca23 [pycaffe] re-expose Net 2015-02-16 22:46:13 -08:00
Jonathan L Long 798a65b442 [pycaffe] expose global ("Caffe::") functions 2015-02-16 22:46:13 -08:00
Jonathan L Long c2fe27b002 [pycaffe] strike down wrappers, momentarily gut all functionality 2015-02-16 22:46:12 -08:00
Jonathan L Long d6df2e09eb [pycaffe] enable numpy API deprecation warnings 2015-02-16 22:45:21 -08:00
Evan Shelhamer c5a6a8c1fe Merge pull request #1667 from Nerei/feature/cmake_well_done 
Improve CMake build with automation, options, and more.
 2015-02-16 20:55:15 -08:00
Anatoly Baksheev aea5dd674a cmake 2.8.7. support 2015-02-16 20:48:41 -08:00
Anatoly Baksheev 9358247014 improve CMake build 2015-02-16 20:48:16 -08:00
Takuya Narihira 02c99fb9c7 Fix `draw` to support new protobuf format 2015-02-16 16:10:47 -08:00
James Steven Supancic III 1344d1bc51 Fix Draw Net Problem #1709 
Introduced by  Layer type is a string #1694
 2015-02-16 11:39:14 -08:00
Evan Shelhamer 2f0f5dbac2 Merge pull request #1555 from drdan14/draw-net-improvements 
Improvements to network drawing via draw_net.py
 2015-01-25 23:13:52 -08:00
Evan Shelhamer 47e8831ac0 [pycaffe] de-dupe imports 2015-01-25 20:39:13 -08:00
Evan Shelhamer c44bfb38fb [fix] align pytest for solver with #1728 2015-01-22 19:54:57 -08:00
Evan Shelhamer 64b68b274c Merge pull request #1473 from longjon/pytest 
Python testing
 2015-01-22 00:29:35 -08:00
Evan Shelhamer 43bae95d7f Back-merge fixes + docs 
and other fixes and documentation updates.
 2015-01-21 15:54:21 -08:00
Evan Shelhamer 6f71db5480 Merge pull request #1662 from seanbell/fix-python-resize_image 
Fix caffe.io.resize_image for the case of constant images
 2015-01-15 22:31:25 -08:00
Evan Shelhamer 4041e4d011 set mode, phase, device in pycaffe; fix #1700 
Attach mode, phase, and device setters to caffe module itself
so that these can be set before making nets. This is needed to properly
initialize layers with the right device and phase configuration.

Update examples to new usage.
 2015-01-14 22:42:03 -08:00
Jonathan L Long 3f6a85c23c [pycaffe] basic, partial testing of Net and SGDSolver 2015-01-06 20:39:14 -08:00
Jonathan L Long 10c23643bb [pycaffe] expose SGDSolver.step 2015-01-02 19:37:48 -08:00
Sean Bell 2326bed7a6 fixed resize_image for the case of constant images 2014-12-30 20:06:31 -05:00
Daniel Golden 646a204c7d Improvements to network drawing 
* Use ArgumentParser in draw_net.py to parse input arguments

* Plot network horizontally (left-right) by default; yes, this confuses the names "top" and "bottom", but since everyone's monitor is wider than it is tall, you'll be able to see over 50% more of the network on your screen at once. The top-bottom functionality is available via the "--rankdir BT" command line argument to draw_net.py

* Display number of outputs on edge for data, convolution and inner product layers

* In node names, add newline between layer name and layer type, to save horizontal space (only if graph rank direction is left-right or right-left)

* Include convolution and pooling parameters within node names

* Distinct colors for convolution, pooling and inner product nodes
 2014-12-10 10:41:22 -08:00
Daniel Golden a5b7b2d7ed Update mean file help 
Numpy array dimensions should be [Channels x Height x Width], NOT [Height x Width x Channels]

If an image mean file is supplied with channels as the last dimension, classify.py throws errors like:

ValueError: operands could not be broadcast together with shapes (3,227,227) (256,0,225) (3,227,227)
 2014-12-03 14:31:46 -08:00
Michele Pratusevich 8271e6bd0c Update python requirements.txt 
You need pyyaml to run the download scripts for getting the BVLC model references.
 2014-11-18 17:47:13 -05:00
Sergey Karayev 1ed11d5611 back-merge 2014-10-17 11:51:27 -07:00
Evan Shelhamer 5fb32067e0 [docs] pip install harder 2014-10-11 00:25:27 -07:00
Evan Shelhamer b992de2afe [docs] pip install packages in order for dependencies 2014-10-11 00:05:57 -07:00
Evan Shelhamer 7eecdf9c80 Merge pull request #1195 from longjon/python-copy-from 
Expose Net.copy_from and Net.share_with in pycaffe
 2014-10-10 00:16:31 -07:00
max argus b9c1eec5a6 Fix CMake build of pycaffe 
- generate right shared library name
- fix CMake linking with the new layer factory
- find numpy

Fix finding protobuf.
 2014-10-08 22:44:28 +00:00
Sergio Guadarrama 0ba046bc3e Merge pull request #1070 from sguada/move_data_mean 
Refactor data_transform to allow datum, cv:Mat and Blob transformation
 2014-10-03 19:18:48 -07:00
Evan Shelhamer aee2efb001 [fix] include Python.h instead of re-ordering for pycaffe build on OS X 2014-10-03 13:53:52 -07:00
Evan Shelhamer 562ecba364 [fix] bend pycaffe to the whims of OS X header order 
To build pycaffe on OS X the boost headers need to be in this order.
OS X is a bewildering platform.
 2014-10-03 13:44:36 -07:00
Evan Shelhamer 1e9681f915 [fix] set cmake static link command for clang++ and g++ globally 2014-10-03 13:30:21 -07:00
Sergio Guadarrama e9c08e5a8c Fixed MemoryDataLayer to make it work with pycaffe 2014-10-03 11:45:58 -07:00
Jonathan L Long 0dfda37c7f [pycaffe] expose Net::SharedTrainedLayersWith as Net.share_with 2014-10-02 12:51:11 -07:00
Evan Shelhamer 77199ddc9d Merge pull request #1196 from longjon/python-solver-improvements 
Python solver improvements
 2014-10-02 12:50:46 -07:00
Jonathan L Long ab5e86e2ba [pycaffe] expose Net::CopyTrainedLayersFrom as Net.copy_from 
This allows finetuning from Python, for example.
 2014-10-02 12:47:59 -07:00
Jonathan L Long ae3e17ea13 Back-merge to include #1198 2014-10-02 11:53:55 -07:00
Bert 4e0868d091 Update detect.py 
This generator causes an error in detect_windows because it is run twice there.
Better store it as a data structure instead.
 2014-10-02 11:51:16 -07:00
Jonathan L Long 5be4d7681c [pycaffe] expose SGDSolver.iter 2014-09-30 20:29:21 -07:00
Jonathan L Long 34e661b27e [pycaffe] expose SGDSolver.test_nets 2014-09-30 20:28:35 -07:00
Jonathan L Long 2f29ec4672 [pycaffe] add converter for vectors of Nets 2014-09-30 20:23:01 -07:00
Jonathan L Long 08ccc947f1 [pycaffe] fix comment typo 2014-09-30 19:17:17 -07:00
Yangqing Jia c9fb1f4a0a cmake. 2014-09-28 17:45:17 -07:00
Jonathan L Long fdf2de1672 [pycaffe] expose Net::Reshape 2014-09-18 13:17:43 -07:00
Sergey Karayev da715ea07a removed mention of getting_pretrained_models page and old paths 2014-09-04 03:59:14 +01:00
Jonathan L Long 3e12d49324 [pycaffe] use _blob_names, _layer_names instead of removed .name 2014-09-01 19:52:57 -07:00
Jonathan L Long afd8f370b4 [pycaffe] expose Net.blob_names and Net.layer_names 2014-09-01 19:52:38 -07:00
Jonathan L Long ed13a61990 [pycaffe] add converter for vector<string> used by _*_names 2014-09-01 19:52:17 -07:00
Jonathan L Long 3e92496935 [pycaffe] expose Blob.Reshape 
While this function should never be called in the current interface, it
will be necessary to have when embedding Python.
 2014-08-31 01:14:58 -07:00
Jonathan L Long 88221c9a45 [pycaffe] remove name property from PyBlob and PyLayer 
Attaching names to pycaffe Blobs and Layers isn't necessary anymore,
since they are now accessed by name. This change brings PyBlob and
PyLayer closer to their C++ counterparts, and makes it easier to pass
around blobs without knowing their names, e.g., when embedding Python
in a layer.
 2014-08-31 01:13:50 -07:00
Jonathan L Long 7907a56b86 [pycaffe] declare the _caffe module init function 2014-08-31 01:13:50 -07:00
Jonathan L Long f4f508e5f2 [pycaffe] split _caffe into source and header files 
This brings pycaffe more in line with usual C++ style, and will allow
the wrapper classes to be used within caffe when embedding Python.
 2014-08-31 01:13:39 -07:00
Jonathan L Long 6be5b6ed7a [pycaffe] make PyBlob a template over Dtype 
We will not actually support double nets in Python with this commit, but
this change will be necessary to create layers that embed Python.
 2014-08-31 00:07:43 -07:00
Jonathan L Long 3535b4731a [pycaffe] make PyNet a class, not a struct 2014-08-31 00:07:11 -07:00
Jonathan L Long 128be26dad [pycaffe] use a namespace alias instead of using directives for boost::python 2014-08-30 23:55:57 -07:00
Jonathan L Long 9d541499a9 [pycaffe] live in caffe namespace instead of opening it 2014-08-30 23:49:44 -07:00
Jonathan L Long e7f712eb97 [pycaffe] use class names of the form Py* instead of Caffe* 2014-08-30 23:48:41 -07:00
Adam Kosiorek a2a4d636c9 cmake build configuration for travis-ci 2014-08-17 01:07:16 -07:00
Adam Kosiorek ececfc0c0c cmake build system 2014-08-17 01:07:11 -07:00
Evan Shelhamer f8b612d8cd default raw_scale in python scripts to ImageNet model value 2014-08-09 20:43:40 -07:00
Evan Shelhamer b97b88f0c2 LICENSE governs the whole project so strip file headers 2014-08-06 18:35:29 -07:00
Evan Shelhamer 0db94786a7 drop np.asarray() in favor of declaration (~1.75x speedup) 2014-08-05 23:17:59 -07:00
Evan Shelhamer 4f7726916c fix pycaffe context cropping with or without mean 2014-08-05 23:17:59 -07:00
Evan Shelhamer e1b3413708 take array in pycaffe `Net.set_mean()` instead of file path 2014-08-05 23:17:59 -07:00
Evan Shelhamer d5c3cef471 fix pycaffe input processing 
- load an image as [0,1] single / np.float32 according to Python convention
- fix input scaling during preprocessing:
  - scale input for preprocessing by `raw_scale` e.g. to map an image
    to [0, 255] for the CaffeNet and AlexNet ImageNet models
  - scale feature space by `input_scale` after mean subtraction
  - switch examples to raw scale for ImageNet models
  - fix #525
- preserve type after resizing.
- resize 1, 3, or K channel images with special casing between
  skimage.transform (1 and 3) and scipy.ndimage (K) for speed
 2014-08-05 23:17:59 -07:00
Evan Shelhamer 99f27fc082 define caffe.Net input preprocessing members by boost::python 
define `Net.{mean, input_scale, channel_swap}` on the boost::python side
so that the members always exist. drop ugly initialization logic.
 2014-08-05 15:55:45 -07:00
Evan Shelhamer d842f4a24f Merge pull request #733 from longjon/pycaffe-tweaks 
pycaffe fixes
 2014-07-28 14:14:03 -07:00
Jonathan L Long fb2f7c1c27 pycaffe: test channel_order and input_scale against None 
An input_scale of zero should produce warnings/errors/infinities rather
than being the same as an input_scale of one. A channel_order of () or 0
should not be equivalent to the default ordering.
 2014-07-18 17:08:49 -07:00
Jonathan L Long 32fefa96f9 pycaffe: reorder exceptions 
It doesn't make sense to complain about input not being batch-sized if
it isn't even 4-d in the first place.
 2014-07-18 17:08:49 -07:00
Jonathan L Long fc48d5de5c pycaffe: allow unspecified mean. Fixes #671. 2014-07-18 17:08:49 -07:00
Jonathan L Long 3ef43e41e7 pycaffe: expose Forward/Backward From/To as kwargs start and end 2014-07-18 15:39:55 -07:00
Evan Shelhamer 506e476047 Merge pull request #455 from shelhamer/pycaffe-save 
Save from python for net surgery
 2014-06-12 16:00:49 -07:00
Evan Shelhamer 6a5b2c19c7 save from python for net surgery 
0. Scheme desired parameters.
1. Do surgery on the net through `net.params['name'][idx].data[...] = `.
2. Save post-operation net params by `net.save('fname')`.

Handwoven deep nets, anyone?
 2014-06-12 14:27:36 -07:00
Evan Shelhamer a7e397abbd Merge pull request #482 from shelhamer/rcnn-detector-example 
Make R-CNN the Caffe detection example
 2014-06-11 15:22:04 -07:00
Evan Shelhamer 8dc270e691 pycaffe: leave grayscale images gray according to arg 2014-06-11 10:17:17 -07:00
Evan Shelhamer 206bf86872 make selective search proposals with R-CNN configuration 2014-06-09 21:06:18 -07:00
Evan Shelhamer 4992abecec pycaffe Detector crops with surrounding context 
- caffe.Detector learned how to crop windows with context in the R-CNN
  style s.t. the bordero of the network input is a given amount of
  context.
- add --context_pad arg to detect.py for amount of context. Default is
  16, as in R-CNN.
 2014-06-09 18:14:25 -07:00
Evan Shelhamer a57b8d5b40 fix old detect.py default 2014-06-09 18:14:25 -07:00
Evan Shelhamer 8c609da736 caffe.Net preprocessing members belong to object, not class 2014-05-26 21:50:39 -07:00
Evan Shelhamer 211c3c29bb follow-up on #443 to invert k channels (instead of 3) 2014-05-23 09:21:17 -07:00
James Thewlis e5fcf030fb Correctly invert the swapping of colour channels 
In the 'deprocess' method, get back the image with the original channel order
by inverting the original transform, rather than reversing the tuple which is
incorrect.
 2014-05-23 09:21:17 -07:00
Evan Shelhamer 1cb6feaf5a fix draw_net python script 
include caffe.draw for drawing functions.
 2014-05-22 01:02:56 -07:00
Evan Shelhamer 8830dc5b81 preprocess single inputs instead of lists 
For compositionality and expectations.
 2014-05-20 11:42:41 -07:00
Evan Shelhamer 02ecf1dacf windowed detection in python 2014-05-20 11:42:41 -07:00
Evan Shelhamer 111df0ecfc squash infuriating loop assignment bug in batching 2014-05-19 23:56:48 -07:00
Evan Shelhamer 2fc32d558a image classification in python 2014-05-19 23:56:48 -07:00
Evan Shelhamer bf4d7262bd fix padding for the last batch 2014-05-19 23:56:16 -07:00
Evan Shelhamer 6b85fd006d split drawnet into module code and script 
Don't run scripts in the module dir to avoid import collisions between
io and caffe.io.
 2014-05-19 23:55:22 -07:00
Evan Shelhamer 50d0b6d9c6 add caffe.io submodule for conversions, image loading and resizing 2014-05-19 23:55:21 -07:00
Evan Shelhamer 738c8758e3 fix python mean subtraction 2014-05-19 23:54:25 -07:00
Evan Shelhamer 37123a5126 drop cute names in favor of Net.{pre,de}process() for input formatting 
...and refer to inputs as inputs and not images since general vectors
and matrices are perfectly fine.
 2014-05-16 16:10:25 -07:00
Evan Shelhamer 51024131cc Net.caffeinate() and Net.decaffeinate() format/unformat lists 2014-05-16 16:10:25 -07:00
Evan Shelhamer 5d584c27f0 take blob args as ndarrays and assign on the python side 
Take blob args and give blob returns as single ndarrays instead of lists
of arrays.

Assign the net blobs and diffs as needed on the python side, which
reduces copies and simplifies the C++ side of the wrapper.

Thanks @longjon for the suggestion.
 2014-05-16 16:10:25 -07:00
Evan Shelhamer 025c64e71d resize to input dimensions when formatting in python 2014-05-15 13:52:53 -07:00
Evan Shelhamer 459c8c1dac replace iterator with indices for consistency 2014-05-15 13:06:17 -07:00
Evan Shelhamer a7f6750cc3 python style 2014-05-15 12:35:32 -07:00
Evan Shelhamer a3b307a385 fix accidental revert of Init() from f5c28581 2014-05-15 10:20:43 -07:00
Evan Shelhamer 1b23680244 batch inputs in python by forward_all() and forward_backward_all() 2014-05-14 23:48:01 -07:00
Evan Shelhamer 8af33e8ca2 don't squeeze blob arrays for python 
Preserve the non-batch dimensions of blob arrays, even for singletons.

The forward() and backward() helpers take lists of ndarrays instead of a
single ndarray per blob, and lists of ndarrays are likewise returned.

Note that for output the blob array could actually be returned as a
single ndarray instead of a list.
 2014-05-14 20:17:16 -07:00
Evan Shelhamer af0b857d54 python forward() and backward() extract any blobs and diffs 2014-05-14 20:14:55 -07:00
Evan Shelhamer ac5e6fa1ce python Net.backward() helper and Net.BackwardPrefilled() 2014-05-14 14:37:33 -07:00
Evan Shelhamer 9d4324e5e7 bad forward/backward inputs throw exceptions instead of crashing python 2014-05-14 14:02:54 -07:00
Evan Shelhamer 0e5a5cf50e pycaffe Net.forward() helper 
Do forward pass by prefilled or packaging input + output blobs and
returning a {output blob name: output list} dict.
 2014-05-14 13:44:02 -07:00
Evan Shelhamer 96cd02dd53 set input preprocessing per blob in python 2014-05-14 13:44:02 -07:00
Evan Shelhamer 56ca978c4e expose input and output blob names to python as lists 2014-05-14 13:44:02 -07:00
Evan Shelhamer 872ddf3f81 pycaffe comments, lint 2014-05-13 18:10:42 -07:00
Evan Shelhamer 8da2a3209c add python io getters, mean helper, and image caffeinator/decaffeinator 2014-05-13 18:10:41 -07:00
Evan Shelhamer 47ec9ace41 make python wrapper mean match binaryproto dimensions 
ilsvrc_2012_mean.npy has dims K x H x W.

Code written for the old D x D x K mean needs to be rewritten!
 2014-05-13 18:10:41 -07:00
Evan Shelhamer 51f276e6b2 match existing python formatting 2014-05-13 18:10:41 -07:00
Jonathan L Long d2fac2d6eb pycaffe: allow 1d labels to be passed to set_input_arrays 2014-05-02 13:27:07 -07:00
Jonathan L Long 76c255449f pycaffe: add Net.set_input_arrays for input from numpy 
This requires a net whose first layer is a MemoryDataLayer.
 2014-05-02 13:25:51 -07:00
Jonathan L Long e1072a66d4 pycaffe: store a shared_ptr<CaffeNet> in SGDSolver 
Doing this, rather than constructing the CaffeNet wrapper every time,
will allow the wrapper to hold references that last at least as long as
SGDSolver (which will be necessary to ensure that data used by
MemoryDataLayer doesn't get freed).
 2014-05-02 13:25:51 -07:00
Jonathan L Long 634a382bce pycaffe: let boost pass shared_ptr<CaffeNet> 2014-05-02 13:25:51 -07:00
Jonathan L Long f5c285813c pycaffe: add unary CaffeNet constructor for uninitialized nets 
Note that parameters are uninitialized, not zero-filled.
 2014-04-25 15:49:06 -07:00
Evan Shelhamer e9ce6cf4de note pydot dependency of model visualization 2014-04-25 09:23:00 -07:00
ZhiHeng NIU 65015e3712 Update the drawnet.py to reflect the recent revised net definition. 2014-04-25 17:43:56 +08:00
Jeff Donahue f45a94660a fix lint errors by adding 'explicit' to new single arg pycaffe 
constructors
 2014-04-11 16:10:15 -07:00
Jonathan L Long 584abd3832 pycaffe: expose SGDSolver.solve 2014-04-04 23:39:56 -07:00
Jonathan L Long 8d44507a30 pycaffe: introduce CheckFile helper 2014-04-04 23:39:56 -07:00
Jonathan L Long 2915f4bef3 pycaffe: expose SGDSolver.net 2014-04-04 23:39:56 -07:00
Jonathan L Long a8a0191089 switch from inheritance to directly overriding methods for caffe.Net 
Also, internally drop the "Caffe" prefix for classes exported by the
internal _caffe module. This change means that SGDSolver will
automatically return a Net with the "nice" Python interface rather than
the internal one, (and avoids the extra layer of method wrapping that
results from switching to composition).
 2014-04-04 23:39:56 -07:00
Jonathan L Long 7823d66d14 pycaffe: expose SGDSolver 2014-04-04 23:39:56 -07:00
Evan Shelhamer 2c0c5b03dc Revert "Fix segfault." (python) 
Revert 88507032de from #287.

The call to `Py_Initialize()` is unnecessary and masked a build
configuration issue.
 2014-04-04 22:10:47 -07:00
Evan Shelhamer a9516a4d43 Merge pull request #277 from jeffdonahue/makefile-improvements 
Makefile improvements
 2014-04-04 15:25:38 -07:00
Rob Hess 88507032de Fix segfault. 2014-04-02 17:50:18 -07:00
Jeff Donahue 39ca2def35 cleanup python build and clean 2014-04-02 10:58:49 -07:00
Evan Shelhamer 6db92e838b Standardize copyright, add root-level CONTRIBUTORS credit 2014-03-27 11:20:35 -07:00
Evan Shelhamer d124b8ffe8 drop models/ in favor of examples/ 
Move the Caffe reference imagenet model and script to fetch it to
examples/imagenet.

Caffe bundles reference models, but it makes more sense to keep them
close to examples.
 2014-03-18 14:03:05 -07:00
Jonathan L Long 895cd777e5 Raise Python exceptions if CaffeNet input files don't exist 
This is for convenience only; if the input files are moved or deleted at
the wrong time, pycaffe could still crash (as it did before). However,
in most cases this will make interactive use easier.
 2014-03-17 22:59:02 -07:00
Jonathan L Long 82c38b25cc Add comment explaining placement of system headers in C++ Python module 2014-03-17 22:59:01 -07:00
Evan Shelhamer f890a18b49 Merge pull request #199 from longjon/pycaffe-ordereddict 
Update Python interface, using OrderedDict for blobs and layers to expose activations, params, and diffs.
 2014-03-17 17:07:36 -07:00
Jonathan L Long 51d27d14e3 Move semicolon to appease lint 2014-03-17 16:46:47 -07:00
Jonathan L Long b9452c4831 Fix indentation in C++ Python module 2014-03-15 02:40:16 -07:00
Jonathan L Long 97186375f5 Add extra comment on vector_indexing_suite to _caffe.cpp 2014-03-15 02:28:07 -07:00
Jonathan L Long ec424a57e8 Remove spurious constructors from CaffeBlob and CaffeBlobWrap 
This prevents accidental construction of partially-specified blobs,
which could break the Python layer.
 2014-03-15 02:28:07 -07:00
Jonathan L Long 922c4adfd1 Add names to the blobs returned by CaffeLayer 
Even though these are redundant, this keeps the meaning of
CaffeBlob.name consistent for now, and will allow CaffeBlob construction
to be simplified.
 2014-03-15 02:28:07 -07:00
Jonathan L Long 8db1400f02 Expose caffe.Net.params as an OrderedDict 2014-03-15 02:27:35 -07:00
Jonathan L Long 91cd44b899 Expose layers and remove now-redundant params in Python interface 
For now, only layer names and blobs are accessible, but more
functionality (layer Forward and Backward) could be added later.
 2014-03-15 02:24:37 -07:00
Evan Shelhamer 0816a94803 fix wrapper example paths 2014-03-13 20:15:27 -07:00
Jonathan L Long 2f12de35c9 Update imagenet/wrapper.py to use the new Net interface 2014-03-13 11:15:57 -07:00
Jonathan L Long ac1bc6fcdf Update detector.py to use the new Net/blobs interface 2014-03-13 11:15:57 -07:00