From f19896ccca23f091abb82d77a2f281a9c954a147 Mon Sep 17 00:00:00 2001 From: Muneyuki Noguchi Date: Sun, 20 Dec 2015 19:12:09 +0900 Subject: [PATCH] Replace blobs_lr with lr_mult in readme.md. models/finetune_flickr_style/deploy.prototxt uses lr_mult now. --- examples/finetune_flickr_style/readme.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/examples/finetune_flickr_style/readme.md b/examples/finetune_flickr_style/readme.md index ecb9d3d..4e9d41f 100644 --- a/examples/finetune_flickr_style/readme.md +++ b/examples/finetune_flickr_style/readme.md @@ -22,10 +22,10 @@ Because we are predicting 20 classes instead of a 1,000, we do need to change th Therefore, we change the name of the last layer from `fc8` to `fc8_flickr` in our prototxt. Since there is no layer named that in the `bvlc_reference_caffenet`, that layer will begin training with random weights. -We will also decrease the overall learning rate `base_lr` in the solver prototxt, but boost the `blobs_lr` on the newly introduced layer. +We will also decrease the overall learning rate `base_lr` in the solver prototxt, but boost the `lr_mult` on the newly introduced layer. The idea is to have the rest of the model change very slowly with new data, but let the new layer learn fast. Additionally, we set `stepsize` in the solver to a lower value than if we were training from scratch, since we're virtually far along in training and therefore want the learning rate to go down faster. -Note that we could also entirely prevent fine-tuning of all layers other than `fc8_flickr` by setting their `blobs_lr` to 0. +Note that we could also entirely prevent fine-tuning of all layers other than `fc8_flickr` by setting their `lr_mult` to 0. ## Procedure -- 2.7.4