shithub: opus

Info  •  Files  •  Log  •  Branches

Training options for sparse GRU B

--- a/dnn/training_tf2/train_lpcnet.py

+++ b/dnn/training_tf2/train_lpcnet.py

@@ -38,7 +38,10 @@

 parser.add_argument('--quantize', metavar='<input weights>', help='quantize model')

 parser.add_argument('--density', metavar='<global density>', type=float, help='average density of the recurrent weights (default 0.1)')

 parser.add_argument('--density-split', nargs=3, metavar=('<update>', '<reset>', '<state>'), type=float, help='density of each recurrent gate (default 0.05, 0.05, 0.2)')

+parser.add_argument('--grub-density', metavar='<global GRU B density>', type=float, help='average density of the recurrent weights (default 1.0)')

+parser.add_argument('--grub-density-split', nargs=3, metavar=('<update>', '<reset>', '<state>'), type=float, help='density of each GRU B input gate (default 1.0, 1.0, 1.0)')

 parser.add_argument('--grua-size', metavar='<units>', default=384, type=int, help='number of units in GRU A (default 384)')

+parser.add_argument('--grub-size', metavar='<units>', default=16, type=int, help='number of units in GRU B (default 16)')

 parser.add_argument('--epochs', metavar='<epochs>', default=120, type=int, help='number of epochs to train for (default 120)')

 parser.add_argument('--batch-size', metavar='<batch size>', default=128, type=int, help='batch size to use (default 128)')

@@ -51,6 +54,12 @@

 elif args.density is not None:

     density = [0.5*args.density, 0.5*args.density, 2.0*args.density];

+grub_density = (1., 1., 1.)

+if args.grub_density_split is not None:

+    grub_density = args.grub_density_split

+elif args.grub_density is not None:

+    grub_density = [0.5*args.grub_density, 0.5*args.grub_density, 2.0*args.grub_density];

+

 import importlib

 lpcnet = importlib.import_module(args.model)

@@ -138,9 +147,11 @@

     #Adapting from an existing model

     model.load_weights(args.quantize)

     sparsify = lpcnet.Sparsify(0, 0, 1, density)

+    grub_sparsify = lpcnet.SparsifyGRUB(0, 0, 1, args.grua_size, grub_density)

 else:

     #Training from scratch

     sparsify = lpcnet.Sparsify(2000, 40000, 400, density)

+    grub_sparsify = lpcnet.SparsifyGRUB(2000, 40000, 400, args.grua_size, grub_density)

 model.save_weights('{}_{}_initial.h5'.format(args.output, args.grua_size))

-model.fit([in_data, features, periods], out_exc, batch_size=batch_size, epochs=nb_epochs, validation_split=0.0, callbacks=[checkpoint, sparsify])

+model.fit([in_data, features, periods], out_exc, batch_size=batch_size, epochs=nb_epochs, validation_split=0.0, callbacks=[checkpoint, sparsify, grub_sparsify])

-- 

⑨