I was following the example python notebook 00-classification.ipynb and also this
comment about loading the image names into a list.
I tried following the classification example changing these two lines
images = [caffe_root + 'examples/images/cat.jpg',caffe_root + 'examples/images/fishbike.jpg']
net.blobs['data'].data[...] = transformer.preprocess('data', caffe.io.load_image(images)
but then I get the error :
AttributeError: 'list' object has no attribute 'read'
Is there a way to do this or do I need to first convert the files to LMDB or LEVELDB? Then specify the files in the prototxt and then call net.forward()?