Skip to content

Latest commit

 

History

History
20 lines (15 loc) · 768 Bytes

README.md

File metadata and controls

20 lines (15 loc) · 768 Bytes

###Digit Recognition via CNN
####to train:

  • download our dataset and put it into data/;
  • run examples/fdigit/convert.py for converting data into lmdb
  • run solve.py to start training

####to test:

  • download pre-trained model or train your own model as metioned above.
  • run examples/fdigit/test_fcn11_full.m (need Matlab and matcaffe)
  • python bindings are on the go.

Code base on caffe


Contributors: