experiments/STT
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Merge pull request #2686 from lissyx/ensure-r1.15

Browse Source 
Ensure properly link to TensorFlow r1.15
This commit is contained in:
lissyx 2020-01-22 12:31:27 +01:00 committed by GitHub
commit d6ca542722
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
7 changed files with 10 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
.compute
View File

@ -7,7 +7,7 @@ python3 -m venv /tmp/venv
source /tmp/venv/bin/activate
pip install -r <(grep -v tensorflow requirements.txt)
pip install tensorflow-gpu==1.14.0
pip install tensorflow-gpu==1.15.0
# Install ds_ctcdecoder package from TaskCluster
pip install $(python3 util/taskcluster.py --decoder)

6
doc/TRAINING.rst
View File

@ -64,7 +64,7 @@ If you have a capable (NVIDIA, at least 8GB of VRAM) GPU, it is highly recommend
.. code-block:: bash
pip3 uninstall tensorflow
pip3 install 'tensorflow-gpu==1.14.0'
pip3 install 'tensorflow-gpu==1.15.0'
Please ensure you have the required `CUDA dependency <USING.rst#cuda-dependency>`_.
@ -75,7 +75,7 @@ It has been reported for some people failure at training:
tensorflow.python.framework.errors_impl.UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.
[[{{node tower_0/conv1d/Conv2D}}]]
Setting the ``TF_FORCE_GPU_ALLOW_GROWTH`` environment variable to ``true`` seems to help in such cases. This could also be due to an incorrect version of libcudnn. Double check your versions with the `TensorFlow 1.14 documentation <USING.rst#cuda-dependency>`_.
Setting the ``TF_FORCE_GPU_ALLOW_GROWTH`` environment variable to ``true`` seems to help in such cases. This could also be due to an incorrect version of libcudnn. Double check your versions with the `TensorFlow 1.15 documentation <USING.rst#cuda-dependency>`_.
Common Voice training data
^^^^^^^^^^^^^^^^^^^^^^^^^^
@ -185,7 +185,7 @@ TensorFlow has tooling to achieve this: it requires building the target ``//tens
.. code-block::
$ python3 util/taskcluster.py --source tensorflow --artifact convert_graphdef_memmapped_format --branch r1.14 --target .
$ python3 util/taskcluster.py --source tensorflow --artifact convert_graphdef_memmapped_format --branch r1.15 --target .
Producing a mmap-able model is as simple as:

2
doc/USING.rst
View File

@ -23,7 +23,7 @@ Please refer to your system's documentation on how to install these dependencies
CUDA dependency
^^^^^^^^^^^^^^^
The GPU capable builds (Python, NodeJS, C++, etc) depend on the same CUDA runtime as upstream TensorFlow. Currently with TensorFlow 1.14 it depends on CUDA 10.0 and CuDNN v7.5. `See the TensorFlow documentation <https://www.tensorflow.org/install/gpu>`_.
The GPU capable builds (Python, NodeJS, C++, etc) depend on the same CUDA runtime as upstream TensorFlow. Currently with TensorFlow 1.15 it depends on CUDA 10.0 and CuDNN v7.5. `See the TensorFlow documentation <https://www.tensorflow.org/install/gpu>`_.
Getting the pre-trained model
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

4
native_client/README.rst
View File

@ -5,7 +5,7 @@ Building DeepSpeech Binaries
If you'd like to build the DeepSpeech binaries yourself, you'll need the following pre-requisites downloaded and installed:
* `Mozilla's TensorFlow r1.14 branch <https://github.com/mozilla/tensorflow/tree/r1.14>`_
* `Mozilla's TensorFlow r1.15 branch <https://github.com/mozilla/tensorflow/tree/r1.15>`_
* `General TensorFlow requirements <https://www.tensorflow.org/install/install_sources>`_
* `libsox <https://sourceforge.net/projects/sox/>`_
@ -32,7 +32,7 @@ Clone our fork of TensorFlow and checkout the correct version:
.. code-block::
git clone https://github.com/mozilla/tensorflow.git
git checkout origin/r1.14
git checkout origin/r1.15
Bazel: Download & Install
^^^^^^^^^^^^^^^^^^^^^^^^^

2
native_client/dotnet/README.rst
View File

@ -51,7 +51,7 @@ We need to clone ``mozilla/DeepSpeech`` and ``mozilla/tensorflow``.
.. code-block:: bash
git clone --branch r1.14 https://github.com/mozilla/tensorflow
git clone --branch r1.15 https://github.com/mozilla/tensorflow
Configuring the paths
---------------------

2
taskcluster/test-training_16k-linux-amd64-py36m-opt.yml
View File

@ -7,7 +7,7 @@ build:
apt-get -qq -y install ${python.packages_trusty.apt}
args:
tests_cmdline: "${system.homedir.linux}/DeepSpeech/ds/taskcluster/tc-train-tests.sh 3.6.4:m 16k"
convert_graphdef: "https://index.taskcluster.net/v1/task/project.deepspeech.tensorflow.pip.r1.14.351a98ab6e60c2bf257f05e515a420aba3027d8b.cpu/artifacts/public/convert_graphdef_memmapped_format"
convert_graphdef: "https://community-tc.services.mozilla.com/api/index/v1/task/project.deepspeech.tensorflow.pip.r1.15.ceb46aae5836a0f648a2c3da5942af2b7d1b98bf.cpu/artifacts/public/convert_graphdef_memmapped_format"
metadata:
name: "DeepSpeech Linux AMD64 CPU 16kHz training Py3.6"
description: "Training a DeepSpeech LDC93S1 model for Linux/AMD64 16kHz Python 3.6, CPU only, optimized version"

2
taskcluster/test-training_8k-linux-amd64-py36m-opt.yml
View File

@ -7,7 +7,7 @@ build:
apt-get -qq -y install ${python.packages_trusty.apt}
args:
tests_cmdline: "${system.homedir.linux}/DeepSpeech/ds/taskcluster/tc-train-tests.sh 3.6.4:m 8k"
convert_graphdef: "https://index.taskcluster.net/v1/task/project.deepspeech.tensorflow.pip.r1.14.351a98ab6e60c2bf257f05e515a420aba3027d8b.cpu/artifacts/public/convert_graphdef_memmapped_format"
convert_graphdef: "https://community-tc.services.mozilla.com/api/index/v1/task/project.deepspeech.tensorflow.pip.r1.15.ceb46aae5836a0f648a2c3da5942af2b7d1b98bf.cpu/artifacts/public/convert_graphdef_memmapped_format"
metadata:
name: "DeepSpeech Linux AMD64 CPU 8kHz training Py3.6"
description: "Training a DeepSpeech LDC93S1 model for Linux/AMD64 8kHz Python 3.6, CPU only, optimized version"