deviandice/DL-Art-School
1
0
Fork 0
forked from mrq/DL-Art-School
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix bug and allow position encodings to be trained separately from the rest of the model

Browse Source
This commit is contained in:
James Betker 2022-04-08 16:26:01 -06:00
parent 09ab1aa9bc
commit 032983e2ed
1 changed files with 12 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
codes/models/audio/tts/unified_voice2.py
View File

@ -241,7 +241,7 @@ class UnifiedVoice(nn.Module):
mel_length_compression=1024, number_text_tokens=256,
start_text_token=255, stop_text_token=0, number_mel_codes=8194, start_mel_token=8192,
stop_mel_token=8193, train_solo_embeddings=False, use_mel_codes_as_input=True,
checkpointing=True, average_conditioning_embeddings=False):
checkpointing=True, average_conditioning_embeddings=False, freeze_everything_but_position_embeddings=False):
"""
Args:
layers: Number of layers in transformer stack.
@ -272,10 +272,10 @@ class UnifiedVoice(nn.Module):
self.stop_mel_token = stop_mel_token
self.layers = layers
self.heads = heads
self.max_conditioning_inputs = max_conditioning_inputs
self.max_mel_tokens = -1 if max_mel_tokens == -1 else max_mel_tokens+2+self.max_conditioning_inputs
self.max_text_tokens = -1 if max_text_tokens == -1 else max_text_tokens+2
self.model_dim = model_dim
self.max_conditioning_inputs = max_conditioning_inputs
self.mel_length_compression = mel_length_compression
self.conditioning_encoder = ConditioningEncoder(80, model_dim, num_attn_heads=heads)
self.average_conditioning_embeddings = average_conditioning_embeddings
@ -304,6 +304,15 @@ class UnifiedVoice(nn.Module):
for module in embeddings:
module.weight.data.normal_(mean=0.0, std=.02)
if freeze_everything_but_position_embeddings:
for p in self.parameters():
p.requires_grad = False
p.DO_NOT_TRAIN = True
for m in [self.mel_pos_embedding, self.text_pos_embedding]:
for p in m.parameters():
del p.DO_NOT_TRAIN
p.requires_grad = True
def get_grad_norm_parameter_groups(self):
return {
'conditioning_encoder': list(self.conditioning_encoder.parameters()),
@ -566,7 +575,7 @@ def register_unified_voice2(opt_net, opt):
if __name__ == '__main__':
gpt = UnifiedVoice(model_dim=256, heads=4, train_solo_embeddings=True, use_mel_codes_as_input=True, max_conditioning_inputs=4)
gpt = UnifiedVoice(model_dim=256, heads=4, train_solo_embeddings=True, use_mel_codes_as_input=True, max_conditioning_inputs=4, freeze_everything_but_position_embeddings=True)
l = gpt(torch.randn(2, 3, 80, 800),
torch.randint(high=256, size=(2,120)),
torch.tensor([32, 120]),