From 786510cd6ac87032a2d78b8644e108393a085b61 Mon Sep 17 00:00:00 2001 From: Eren Golge Date: Sat, 23 Mar 2019 17:33:47 +0100 Subject: [PATCH] loss functions updates --- config.json | 2 +- layers/losses.py | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/config.json b/config.json index 7b9e2719..837aa167 100644 --- a/config.json +++ b/config.json @@ -29,7 +29,7 @@ "url": "tcp:\/\/localhost:54321" }, - "reinit_layers": ["model.decoder.attention_layer"], + "reinit_layers": ["model.decoder.attention_layer"], //set which layers to be reinitialized in finetunning. Only used if --restore_model is provided. "model": "Tacotron2", // one of the model in models/ "grad_clip": 0.02, // upper limit for gradients for clipping. diff --git a/layers/losses.py b/layers/losses.py index 4e8cea81..0597bdf9 100644 --- a/layers/losses.py +++ b/layers/losses.py @@ -23,6 +23,7 @@ class L1LossMasked(nn.Module): loss: An average loss value masked by the length. """ # mask: (batch, max_len, 1) + target.requires_grad = False mask = sequence_mask( sequence_length=length, max_len=target.size(1)).unsqueeze(2).float() mask = mask.expand_as(input) @@ -51,6 +52,7 @@ class MSELossMasked(nn.Module): loss: An average loss value masked by the length. """ # mask: (batch, max_len, 1) + target.requires_grad = False mask = sequence_mask( sequence_length=length, max_len=target.size(1)).unsqueeze(2).float() mask = mask.expand_as(input)