Last active
January 13, 2017 16:30
-
-
Save Vishruit/25925b29bef787ac6b9a9531452334f1 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Following instructions @ http://mi.eng.cam.ac.uk/projects/segnet/tutorial.html | |
1) Git cloning @ https://github.com/alexgkendall/caffe-segnet | |
-- conda new env create | |
2) proceed to caffe installation @ http://caffe.berkeleyvision.org/installation.html | |
-- NO change to Makefile.config | |
-- corrected path to cuda | |
-- OMG!! Its installing perfectly fine now without any errors yet! | |
-- make all successful | |
-- heading towards make test | |
-- make test success! Awesome | |
--make runtest ... Success -- 2 disabled tests :| | |
3) Test run a program | |
3.2) Pre-req | |
-- pycaffe dowsnt work now | |
-- git clone @ https://github.com/alexgkendall/SegNet-Tutorial | |
-- edit vim Models/segnet_train.prototxt && vim Models/segnet_inference.prototxt | |
changed data source and batch size | |
same for segnet basic | |
-- home path [/home/rise/Vishruit/soft/SegNet-Tutorial] | |
-- Done | |
3.2) Running the test program | |
-- Command --> ./SegNet/caffe-segnet/build/tools/caffe train -gpu 0 -solver /SegNet/Models/segnet_solver.prototxt # This will begin training SegNet on GPU 0 | |
./SegNet/caffe-segnet/build/tools/caffe train -gpu 0 -solver /SegNet/Models/segnet_solver.prototxt # This will begin training SegNet on GPU 0 | |
./SegNet/caffe-segnet/build/tools/caffe train -gpu 0 -solver /SegNet/Models/segnet_basic_solver.prototxt # This will begin training SegNet-Basic on GPU 0 | |
./SegNet/caffe-segnet/build/tools/caffe train -gpu 0 -solver /SegNet/Models/segnet_solver.prototxt -weights /SegNet/Models/VGG_ILSVRC_16_layers.caffemodel # This will begin training SegNet on GPU 0 with a pretrained encoder | |
3.3) Made pycaffe | |
-- install all the dependencies. Make sure of that! | |
-- success | |
-- issue with $PYTHONPATH - automatically gets reset to no value | |
4) Preparing dataset | |
-- need to figure out the age old issue of doubling size upon conversion problem | |
-- shall go ahead with .png format [test TODO] | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment