Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more
Hey,
is there a way to set layer Specific Learning Rate for the R version?
I found two different posts, one naming: Optimizer.set_lr_scale(). and one mx.sym.YOURSYMBOL(attr={'lr_mult': 0.01}).
which both make sense but I haven't found a way to do this in R.
Thanks for any help.
Hey, is there a way to set layer Specific Learning Rate for the R version? I found two different posts, one naming: Optimizer.set_lr_scale(). and one mx.sym.YOURSYMBOL(attr={'lr_mult': 0.01}).
which both make sense but I haven't found a way to do this in R. Thanks for any help.