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

Merge pull request #3340 from lissyx/tf-1.15.4-r0.8

Browse Source 
Fix #3088: Use TensorFlow 1.15.4 with CUDNN fix
This commit is contained in:
lissyx 2020-09-25 16:21:49 +02:00 committed by GitHub
commit 3ebcbdea91
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Dockerfile.build.tmpl
View File

@ -64,7 +64,7 @@ ENV TF_CUDA_PATHS "/usr,/usr/local/cuda-10.1,/usr/lib/x86_64-linux-gnu/"
ENV TF_CUDA_VERSION 10.1 ENV TF_CUDA_VERSION 10.1
ENV TF_CUDNN_VERSION 7.6 ENV TF_CUDNN_VERSION 7.6
ENV TF_CUDA_COMPUTE_CAPABILITIES 6.0 ENV TF_CUDA_COMPUTE_CAPABILITIES 6.0
ENV TF_NCCL_VERSION 2.4 ENV TF_NCCL_VERSION 2.7
# Common Environment Setup # Common Environment Setup
ENV TF_BUILD_CONTAINER_TYPE GPU ENV TF_BUILD_CONTAINER_TYPE GPU

3
Dockerfile.train.tmpl
View File

@ -43,6 +43,9 @@ RUN pip3 install --upgrade native_client/ctcdecode/dist/*.whl
# Prepare deps # Prepare deps
RUN pip3 install --upgrade pip==20.2.2 wheel==0.34.2 setuptools==49.6.0 RUN pip3 install --upgrade pip==20.2.2 wheel==0.34.2 setuptools==49.6.0
# FIXME: Hack because no 1.15.4 Docker image
RUN pip3 install --upgrade tensorflow-gpu==1.15.4
# Install DeepSpeech # Install DeepSpeech
# - No need for the decoder since we did it earlier # - No need for the decoder since we did it earlier
# - There is already correct TensorFlow GPU installed on the base image, # - There is already correct TensorFlow GPU installed on the base image,

2
doc/TRAINING.rst
View File

@ -65,7 +65,7 @@ If you have a capable (NVIDIA, at least 8GB of VRAM) GPU, it is highly recommend
.. code-block:: bash .. code-block:: bash
pip3 uninstall tensorflow pip3 uninstall tensorflow
pip3 install 'tensorflow-gpu==1.15.2' pip3 install 'tensorflow-gpu==1.15.4'
Please ensure you have the required :ref:`CUDA dependency <cuda-deps>`. Please ensure you have the required :ref:`CUDA dependency <cuda-deps>`.

2
setup.py
View File

@ -74,7 +74,7 @@ def main():
] ]
tensorflow_pypi_dep = [ tensorflow_pypi_dep = [
'tensorflow == 1.15.2' 'tensorflow == 1.15.4'
] ]
# Due to pip craziness environment variables are the only consistent way to # Due to pip craziness environment variables are the only consistent way to