From cd06a4c1e5761d118c2d9d3aec7b38c512d74229 Mon Sep 17 00:00:00 2001 From: Eren Golge Date: Tue, 12 Nov 2019 13:51:22 +0100 Subject: [PATCH] linter fix --- layers/common_layers.py | 3 --- layers/tacotron2.py | 2 +- speaker_encoder/train.py | 2 +- 3 files changed, 2 insertions(+), 5 deletions(-) diff --git a/layers/common_layers.py b/layers/common_layers.py index 6a1e4b6e..006aa57a 100644 --- a/layers/common_layers.py +++ b/layers/common_layers.py @@ -175,9 +175,6 @@ class GravesAttention(nn.Module): phi_t = g_t * torch.exp(-0.5 * inv_sig_t * (mu_t_ - j)**2) alpha_t = self.COEF * torch.sum(phi_t, 1) - if alpha_t.max() > 1e+3: - breakpoint() - # apply masking if mask is not None: alpha_t.data.masked_fill_(~mask, self._mask_value) diff --git a/layers/tacotron2.py b/layers/tacotron2.py index 4046b06b..78bdd10d 100644 --- a/layers/tacotron2.py +++ b/layers/tacotron2.py @@ -254,7 +254,7 @@ class Decoder(nn.Module): memories = torch.cat((memory, memories), dim=0) memories = self._update_memory(memories) if speaker_embeddings is not None: - memories = torch.cat([memories, speaker_embeddings], dim=-1) + memories = torch.cat([memories, speaker_embeddings], dim=-1) memories = self.prenet(memories) self._init_states(inputs, mask=mask) diff --git a/speaker_encoder/train.py b/speaker_encoder/train.py index 962bb881..19067401 100644 --- a/speaker_encoder/train.py +++ b/speaker_encoder/train.py @@ -11,7 +11,7 @@ from TTS.speaker_encoder.dataset import MyDataset from TTS.speaker_encoder.loss import GE2ELoss from TTS.speaker_encoder.model import SpeakerEncoder from TTS.speaker_encoder.visual import plot_embeddings -from TTS.speaker_encoder.generic_utils import save_best_model, save_checkpoint +from TTS.speaker_encoder.generic_utils import save_best_model from TTS.utils.audio import AudioProcessor from TTS.utils.generic_utils import (NoamLR, check_update, copy_config_file, count_parameters,