Docker installation error #646
Comments
|
Could you please post the complete logs? |
|
ubuntu@ip-10-12-2-14:/fsx_new/Alex/tmp1/icefall/docker/Ubuntu18.04-pytorch1.12.1-cuda11.3-cudnn8$ docker build -t icefall/pytorch1.12.1 . -- Found Torch: /opt/conda/lib/python3.7/site-packages/torch/lib/libtorch.so But you are using CUDA 11.3 to compile k2. Please try to use the same CUDA version for PyTorch and k2. You can remove this check if you are sure this will not cause problems Call Stack (most recent call first): -- Configuring incomplete, errors occurred! Traceback (most recent call last): |
|
I think there might be some issues with the docker image. What is the output of the following command when it is run from within the docker container before installing k2? |
|
Unfortunately I cannot start the docker. I do following: docker start funny_einstein Now, I am trying to run docker ps And I do not see that the docker funny_enstein was started. |
|
Please try docker pull pytorch/pytorch:1.12.1-cuda11.3-cudnn8-devel
docker run -it pytorch/pytorch:1.12.1-cuda11.3-cudnn8-develand it should present you a terminal. Please enter the following command in the terminal python3 -m torch.utils.collect_env |
|
Before I doing that. Could you please look on the output of nvidia-smi. Maybe I should use another pytorch docker (something suitable for 11.4) ? ubuntu@ip-10-12-2-14:/fsx_new$ nvidia-smi |
|
In any case following is the output that you requested ubuntu@ip-10-12-2-14:/fsx_new/Alex/tmp1/icefall/docker/Ubuntu18.04-pytorch1.12.1-cuda11.3-cudnn8$ docker exec -it icefall /bin/bash OS: Ubuntu 18.04.6 LTS (x86_64) Python version: 3.7.13 (default, Mar 29 2022, 02:18:16) [GCC 7.5.0] (64-bit runtime) Versions of relevant libraries: |
|
It is fine to use CUDA 11.3 as it is lower than the version displayed by nvidia-smi. I think there are no PyTorch versions that support CUDA 11.4. |
|
Is there some problem with the following output? ubuntu@ip-10-12-2-14:/fsx_new/Alex/tmp1/icefall/docker/Ubuntu18.04-pytorch1.12.1-cuda11.3-cudnn8$ docker exec -it icefall /bin/bash OS: Ubuntu 18.04.6 LTS (x86_64) Python version: 3.7.13 (default, Mar 29 2022, 02:18:16) [GCC 7.5.0] (64-bit runtime) Versions of relevant libraries: |
|
Please pay attention to the output Is CUDA available: False That may explain why it complained when installing k2. |
|
Can you run the container that was built successfully before the dockerfile hit the error?
Get the container ID of the top or second top container that should be dead.
then share your conda list output. |
|
How can I solve the problem of Is CUDA available: False ? |
|
One of the steps might have uninstalled the cuda available PyTorch and installed the CPU PyTorch instead. The image by default ships with cuda available PyTorch. It could be the installation of torchaudio as part of Lhotse's requirement that have uninstalled the cuda version of PyTorch. In this dockerfile I have hard-coded torchaudio version to 0.12. If you still can't fix it, can you try this? FROM pytorch/pytorch:1.12.1-cuda11.3-cudnn8-devel
# install normal source
RUN apt-get update && \
apt-get install -yq --no-install-recommends \
g++ \
make \
automake \
autoconf \
bzip2 \
unzip \
wget \
sox \
libtool \
git \
subversion \
zlib1g-dev \
gfortran \
ca-certificates \
patch \
ffmpeg \
valgrind \
libssl-dev \
vim \
curl
# cmake
RUN wget -P /opt https://cmake.org/files/v3.18/cmake-3.18.0.tar.gz && \
cd /opt && \
tar -zxvf cmake-3.18.0.tar.gz && \
cd cmake-3.18.0 && \
./bootstrap && \
make && \
make install && \
rm -rf cmake-3.18.0.tar.gz && \
find /opt/cmake-3.18.0 -type f \( -name "*.o" -o -name "*.la" -o -name "*.a" \) -exec rm {} \; && \
cd -
# flac
RUN wget -P /opt https://downloads.xiph.org/releases/flac/flac-1.3.2.tar.xz && \
cd /opt && \
xz -d flac-1.3.2.tar.xz && \
tar -xvf flac-1.3.2.tar && \
cd flac-1.3.2 && \
./configure && \
make && make install && \
rm -rf flac-1.3.2.tar && \
find /opt/flac-1.3.2 -type f \( -name "*.o" -o -name "*.la" -o -name "*.a" \) -exec rm {} \; && \
cd -
RUN conda install -y -c pytorch torchaudio=0.12 && \
pip install kaldiio graphviz
#install k2 from source
RUN git clone https://github.com/k2-fsa/k2.git /opt/k2 && \
cd /opt/k2 && \
python3 setup.py install
# install lhotse
RUN pip install git+https://github.com/lhotse-speech/lhotse
RUN git clone https://github.com/k2-fsa/icefall /workspace/icefall && \
cd /workspace/icefall && \
pip install -r requirements.txt
RUN git clone https://github.com/Minami-Lab-UEC/sherpa.git /workspace/sherpa && \
cd /workspace/sherpa && \
pip install -r ./requirements.txt && \
python3 setup.py install
ENV PYTHONPATH=/workspace/icefall:$PYTHONPATH
WORKDIR /workspace/icefall |
|
@teowenshen , thanks a lot. It seems that your upper script solved the problem. Thank you so much. |
|
Great. I will change the dockerfile tomorrow to close this issue. There's another typo on the README.md that I've been meaning to correct too. The correct syntax for |
|
@csukuangfj Kindly add to the Dockerfile |
Could you please make a PR to add kaldifeat? |
Hi All,
After the following check list I have a below error. Could you help please
During the running I got following warning and error:
CMake Warning at /opt/conda/lib/python3.7/site-packages/torch/share/cmake/Torch/TorchConfig.cmake:22 (message):
static library kineto_LIBRARY-NOTFOUND not found.
Call Stack (most recent call first):
/opt/conda/lib/python3.7/site-packages/torch/share/cmake/Torch/TorchConfig.cmake:127 (append_torchlib_if_found)
cmake/torch.cmake:11 (find_package)
CMakeLists.txt:292 (include)
-- Found Torch: /opt/conda/lib/python3.7/site-packages/torch/lib/libtorch.so
-- PyTorch version: 1.13.0
-- PyTorch cuda version: None
CMake Error at cmake/torch.cmake:52 (message):
PyTorch 1.13.0 is compiled with CUDA None.
But you are using CUDA 11.3 to compile k2.
Please try to use the same CUDA version for PyTorch and k2.
You can remove this check if you are sure this will not cause problems
Call Stack (most recent call first):
CMakeLists.txt:292 (include)
The text was updated successfully, but these errors were encountered: