0.4 scalar tensor padding update

mask utils update for 0.4 cuda

Dockerfile

@@ -1,3 +1,4 @@
 FROM pytorch/pytorch:0.4_cuda9_cudnn7
 
-RUN pip install numpy scipy matplotlib librosa==0.6.0 tensorflow tensorboardX inflect==0.2.5 Unidecode==1.0.22
+RUN pip install numpy scipy matplotlib librosa==0.6.0 tensorflow tensorboardX
+inflect==0.2.5 Unidecode==1.0.22

README.md

@@ -21,10 +21,10 @@ Distributed and FP16 support relies on work by Christian Sarofeen and NVIDIA's
 3. CD into this repo: `cd tacotron2`
 4. Update .wav paths: `sed -i -- 's,DUMMY,ljs_dataset_folder/wavs,g' filelists/*.txt`
 5. Install [pytorch 0.4](https://github.com/pytorch/pytorch)
-6. Install python requirements or use docker container (tbd)
+6. Install python requirements or build docker image 
     - Install python requirements: `pip install requirements.txt`
     - **OR**
-    - Docker container `(tbd)` 
+    - Build docker image: `docker build --tag tacotron2 .` 
 
 ## Training
 1. `python train.py --output_directory=outdir --log_directory=logdir`

model.py

@@ -470,8 +470,8 @@ class Tacotron2(nn.Module):
         text_padded, input_lengths, mel_padded, gate_padded, \
             output_lengths = batch
         text_padded = to_gpu(text_padded).long()
+        max_len = int(torch.max(input_lengths.data).numpy())
         input_lengths = to_gpu(input_lengths).long()
-        max_len = torch.max(input_lengths.data)
         mel_padded = to_gpu(mel_padded).float()
         gate_padded = to_gpu(gate_padded).float()
         output_lengths = to_gpu(output_lengths).long()

requirements.txt

@@ -1,6 +1,6 @@
 torch==0.2.0.post3
 matplotlib==2.1.0
-tensorflow==1.5.0
+tensorflow
 numpy==1.13.3
 inflect==0.2.5
 librosa==0.6.0

utils.py

@@ -5,7 +5,7 @@ import torch
 
 def get_mask_from_lengths(lengths):
     max_len = torch.max(lengths)
-    ids = torch.arange(0, max_len, out=torch.LongTensor(max_len)).cuda()
+    ids = torch.arange(0, max_len).long().cuda()
     mask = (ids < lengths.unsqueeze(1)).byte()
     return mask