From bd762101dba321146d2d9cb747c79c4c678cbfdb Mon Sep 17 00:00:00 2001 From: Achal Dave Date: Wed, 20 Apr 2016 17:34:29 -0400 Subject: Explicitly point out -weights flag in tutorial The -weights flag is somewhat easy to miss as it's only in one command, but is the crucial thing that anyone searching for 'how to finetune' is looking for. Hopefully this more clearly points out the '-weights' flag, which might otherwise be overlooked in this tutorial. --- examples/finetune_flickr_style/readme.md | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'examples') diff --git a/examples/finetune_flickr_style/readme.md b/examples/finetune_flickr_style/readme.md index 9ba4c921..188dedf1 100644 --- a/examples/finetune_flickr_style/readme.md +++ b/examples/finetune_flickr_style/readme.md @@ -57,7 +57,11 @@ The prototxts in this example assume this, and also assume the presence of the I We'll also need the ImageNet-trained model, which you can obtain by running `./scripts/download_model_binary.py models/bvlc_reference_caffenet`. -Now we can train! (You can fine-tune in CPU mode by leaving out the `-gpu` flag.) +Now we can train! The key to fine-tuning is the `-weights` argument in the +command below, which tells Caffe that we want to load weights from a pre-trained +Caffe model. + +(You can fine-tune in CPU mode by leaving out the `-gpu` flag.) caffe % ./build/tools/caffe train -solver models/finetune_flickr_style/solver.prototxt -weights models/bvlc_reference_caffenet/bvlc_reference_caffenet.caffemodel -gpu 0 -- cgit v1.2.3