NVIDIA / caffe

Caffe: a fast open framework for deep learning.
http://caffe.berkeleyvision.org/
Other
672 stars 263 forks source link

io.py updated #465

Closed dwijaybane closed 6 years ago

dwijaybane commented 6 years ago

This error is solved in BVLC repo so please update reference: https://github.com/BVLC/caffe/pull/3555 https://github.com/BVLC/caffe/issues/5718

meanvalueer

Steps to reproduce: It is a procedure to train & test imagenet data set with ImageNet model

Train We have to run below command to get model and label data. (It tooks long time. Approximately, 10m, 232MB, 405KB/s ). We can download the data files that was used at 2012 ILSVRC (ImageNet Large-Scale Visual Recognition Challenge) with the below commands.

$ time python scripts/download_model_binary.py models/bvlc_reference_caffenet $ sh data/ilsvrc12/get_ilsvrc_aux.sh

Test $ python python/classify.py examples/images/cat.jpg foo

drnikolaev commented 6 years ago

Hi @dwijaybane thanks. This pre-release has the fix, could you give it a try? https://github.com/drnikolaev/caffe/tree/caffe-0.17