Tacotron2: Install instructions for baremetal linux that don't suck?

Created on 23 Aug 2019 · 3Comments · Source: NVIDIA/tacotron2

This is an issue with every instance of tensorflow + nvidia cuda. Why do all the install instructions fail and fail and fail again? Version control and dependencies are too complex, correcting versions in one area ends up breaking some in others. No one is going to use this software if the only people using it managed to install it by mistake. There has to be a better way.

So this is a shout out to anyone who has installed tacotron on an Ubuntu 18.04 with an Nvidia gpu. Any help? Does anyone have a step-by-step bash history they wouldn't mind sharing? Thanks!

oh and if you want some error code see below:

nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2017 NVIDIA Corporation
Built on Fri_Nov__3_21:07:56_CDT_2017
Cuda compilation tools, release 9.1, V9.1.85

(venv3) merlin@merlin-desktop:~/opfoghorn/tacotron2$ python3 train.py --output_directory=/home/merlin/opfoghorn/tacotron2/outputdir/ --log_directory=/home/merlin/opfoghorn/tacotron2/logoutdir/
/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
np_resource = np.dtype([("resource", np.ubyte, 1)])
Traceback (most recent call last):
File "train.py", line 17, in
from hparams import create_hparams
File "/home/merlin/opfoghorn/tacotron2/hparams.py", line 1, in
import tensorflow as tf
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorflow/__init__.py", line 35, in
from tensorflow._api.v1 import compat
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorflow/_api/v1/compat/__init__.py", line 22, in
from tensorflow._api.v1.compat import v2
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorflow/_api/v1/compat/v2/__init__.py", line 303, in
from tensorboard.summary._tf import summary
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorboard/summary/__init__.py", line 25, in
from tensorboard.summary import v1
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorboard/summary/v1.py", line 24, in
from tensorboard.plugins.audio import summary as _audio_summary
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorboard/plugins/audio/summary.py", line 36, in
from tensorboard.plugins.audio import metadata
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorboard/plugins/audio/metadata.py", line 21, in
from tensorboard.compat.proto import summary_pb2
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorboard/compat/proto/summary_pb2.py", line 15, in
from tensorboard.compat.proto import tensor_pb2 as tensorboard_dot_compat_dot_proto_dot_tensor__pb2
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorboard/compat/proto/tensor_pb2.py", line 15, in
from tensorboard.compat.proto import resource_handle_pb2 as tensorboard_dot_compat_dot_proto_dot_resource__handle__pb2
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/tensorboard/compat/proto/resource_handle_pb2.py", line 22, in
serialized_pb=_b('n.tensorboard/compat/proto/resource_handle.protox12x0btensorboard"rnx13ResourceHandleProtox12x0enx06x64x65vicex18x01 x01(tx12x11ntcontainerx18x02 x01(tx12x0cnx04namex18x03 x01(tx12x11nthash_codex18x04 x01(x04x12x17nx0fmaybe_type_namex18x05 x01(tBnnx18org.tensorflow.frameworkBx0eResourceHandlePx01Z=github.com/tensorflow/tensorflow/tensorflow/go/core/frameworkxf8x01x01x62x06proto3')
File "/home/merlin/opfoghorn/apex/venv3/lib/python3.6/site-packages/google/protobuf/descriptor.py", line 879, in __new__
return _message.default_pool.AddSerializedFile(serialized_pb)
TypeError: Couldn't build proto file into descriptor pool!
Invalid proto descriptor for file "tensorboard/compat/proto/resource_handle.proto":
tensorboard.ResourceHandleProto.device: "tensorboard.ResourceHandleProto.device" is already defined in file "tensorboardX/src/resource_handle.proto".
tensorboard.ResourceHandleProto.container: "tensorboard.ResourceHandleProto.container" is already defined in file "tensorboardX/src/resource_handle.proto".
tensorboard.ResourceHandleProto.name: "tensorboard.ResourceHandleProto.name" is already defined in file "tensorboardX/src/resource_handle.proto".
tensorboard.ResourceHandleProto.hash_code: "tensorboard.ResourceHandleProto.hash_code" is already defined in file "tensorboardX/src/resource_handle.proto".
tensorboard.ResourceHandleProto.maybe_type_name: "tensorboard.ResourceHandleProto.maybe_type_name" is already defined in file "tensorboardX/src/resource_handle.proto".
tensorboard.ResourceHandleProto: "tensorboard.ResourceHandleProto" is already defined in file "tensorboardX/src/resource_handle.proto".

pip list
Package Version

absl-py 0.7.1
apex 0.1
asn1crypto 0.24.0
astor 0.8.0
atomicwrites 1.3.0
attrs 19.1.0
audioread 2.1.8
certifi 2018.1.18
chardet 3.0.4
command-not-found 0.3
cryptography 2.1.4
cupshelpers 1.0
cxxfilt 0.2.0
cycler 0.10.0
decorator 4.4.0
defer 1.0.6
distro-info 0.18ubuntu0.18.04.1
gast 0.2.2
google-pasta 0.1.7
grpcio 1.23.0
h5py 2.9.0
httplib2 0.9.2
idna 2.6
importlib-metadata 0.19
inflect 0.2.5
joblib 0.13.2
Keras-Applications 1.0.8
Keras-Preprocessing 1.1.0
keyring 10.6.0
keyrings.alt 3.0
language-selector 0.1
launchpadlib 1.10.6
lazr.restfulclient 0.13.5
lazr.uri 1.0.3
librosa 0.6.0
lightdm-gtk-greeter-settings 1.2.2
llvmlite 0.29.0
Markdown 3.1.1
matplotlib 2.1.0
more-itertools 7.2.0
netifaces 0.10.4
numba 0.45.1
numpy 1.17.0
oauth 1.0.1
packaging 19.1
pexpect 4.2.1
Pillow 5.1.0
pip 19.2.2
pluggy 0.12.0
protobuf 3.9.1
py 1.8.0
pycairo 1.16.2
pycrypto 2.6.1
pycups 1.9.73
pygobject 3.26.1
pyparsing 2.4.2
pytest 5.1.1
python-apt 1.6.4
python-dateutil 2.6.1
python-debian 0.1.32
pytz 2019.2
pyxdg 0.25
PyYAML 5.1.2
reportlab 3.4.0
requests 2.18.4
requests-unixsocket 0.1.5
resampy 0.2.2
scikit-learn 0.21.3
scipy 1.0.0
screen-resolution-extra 0.0.0
SecretStorage 2.3.1
setuptools 41.2.0
simplejson 3.13.2
six 1.11.0
tensorboard 1.14.0
tensorboardX 1.1
tensorflow 1.14.0
tensorflow-estimator 1.14.0
termcolor 1.1.0
torch 1.2.0
torchvision 0.4.0
tqdm 4.34.0
ubuntu-drivers-common 0.0.0
ufw 0.36
Unidecode 1.0.22
urllib3 1.22
usb-creator 0.3.3
vboxapi 1.0
virtualenv 16.7.3
wadllib 1.3.2
wcwidth 0.1.7
Werkzeug 0.15.5
wheel 0.33.6
wrapt 1.11.2
xkit 0.0.0
zipp 0.5.2
zope.interface 4.3.2

+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 1909 G /usr/lib/xorg/Xorg 50MiB |
| 1 1909 G /usr/lib/xorg/Xorg 128MiB |
| 1 5236 G ...quest-channel-token=9884100064965360199 94MiB |
+-----------------------------------------------------------------------------+

Source

danindiana

Most helpful comment

I was having that tensorboard proto error too (after a ton of other errors one at a time). Ended up forcing it to a newer version and it worked:
pip install tensorboardX==1.8

ryangrahamnc on 25 Aug 2019

👍3

All 3 comments

I was having that tensorboard proto error too (after a ton of other errors one at a time). Ended up forcing it to a newer version and it worked:
pip install tensorboardX==1.8

ryangrahamnc on 25 Aug 2019

👍3

See issue #257
Or see this pull request
https://github.com/NVIDIA/tacotron2/commit/0f348efed49dea2a3a40e6f318862f0d34264400