Cacher is the code snippet organizer for pro developers

import os.path as osp

# join path
pascal_root = osp.join(caffe_root, 'data/pascal/VOC2012')

# to determine a path point to a file
if not os.path.isfile(caffe_root + 'models/bvlc_reference_caffenet/bvlc_reference_caffenet.caffemodel'):
    print("Downloading pre-trained CaffeNet model...")
    
# to join path and open file
with open(osp.join(workdir, 'trainnet.prototxt'), 'w') as f:
	# provide parameters to the data layer as a python dictionary. Easy as pie!
    data_layer_params = dict(batch_size = 128, im_shape = [227, 227], split = 'train', pascal_root = pascal_root)
    f.write(caffenet_multilabel(data_layer_params, 'PascalMultilabelDataLayerSync'))