search

liuliu / ccv

C-based/Cached/Core Computer Vision Library, A Modern Computer Vision Library
http://libccv.org
Other
7.07k stars 1.71k forks source link

Caffe to CCV convnet import script? #121

Open geraldchao opened 9 years ago

geraldchao commented 9 years ago

Is there any interest in having a script that converts/imports pre-trained models from Caffe into CCV? Examples of these models are VGG's 2014 models and ones from MIT (https://github.com/BVLC/caffe/wiki/Model-Zoo). There's also an existing python script for importing into vlfeat here: https://github.com/vlfeat/matconvnet/blob/master/utils/import-caffe.py

I thought I check here to see if there's already a script for CCV, and whether there's interest in creating one if not.

Gerald

liuliu commented 9 years ago

No, not yet. I am expecting some contributions as I am distracted on other tasks for 0.7 / 0.8 release. It would look like this script (in reverse fashion: https://github.com/petewarden/ccv/commit/636531e9b1b6b3b250add0054a99b6e27853bf12).

mbertini commented 9 years ago

The latest 0.7 branch has a VGG-D model. Did you create a script to convert those Caffe Zoo models ?

liuliu commented 9 years ago

It is trained from scratch. Their model is distributed under CC NC 4 (non commercial use). I cannot simply redistribute under different license :) On Wed, Dec 24, 2014 at 4:12 AM Marco Bertini notifications@github.com wrote:

The latest 0.7 branch has a VGG-D model. Did you create a script to convert those Caffe Zoo models ?

— Reply to this email directly or view it on GitHub https://github.com/liuliu/ccv/issues/121#issuecomment-68048500.

ghost commented 9 years ago

@liuliu any update for parsing the caffemodel to ccv

craftzdog commented 9 years ago

+1