Keras Library installation (Error Metaclass)

[Deschwaba]

Hi Community,

I would like to use deep learning nodes in KNIME. I have installed ANACONDA with respective libraries. However, KNIME states that the keras and tensorflow are not properly installed:

Does anyone know, how I could fix this?

[mlauber71]

@Deschwaba depending on your operating system and KNIME version installing Deep Learning nodes can be a challenge. I can offer up this information and article:

KNIME and Python — Setting up Deep Learning Environments for Keras and...

Deep Learning Python Environment Mac OS (M2)

@trj I have not yet found a suitable way to get KNIME DL integration and Python Deep Learning packages (Keras, Tensorflow) up and running on an Apple Silicon machine. The current KNIME DL nodes seem to expect older versions of the Python packages which are not there for M1 and M2 systems. You can read about my general efforts here: I also tried to replace several nodes that would want old TF packages with ‘pure’ Python nodes - which kind of is not what a low-code would look like, but did not…

[aichampionshub]

I am facing the same problem on Windows. Keep getting one or the other error. Now getting error around ‘Metaclass’ when installing Keras as a new environment.

Have been trying for few days. Tried the above linked article as well - used the YAML files on a clean install and still facing issues.

I seriously doubt if KNIME is suitable for deep learning. Works like a charm for ML Libraries but basic Keras integration seems to have so many challenges.

[marzukim]

my knime DL installation on my machine for your reference
conda version 23.5.0
keras:
-python 3.5.4
-keras 2.2.2
-tensorflow 1.10.0
-numpy 1.14.2
-pandas 0.20.3
-h5py 2.8.0

tensorflow 2:
python 3.10.11

[PAI]

I have the same problem on Win system, multiple reinstallations of Anaconda and KNIME, but still no go…

Here are the packages installed by the KNIME “create new environment”:

Anaconda version 2.5.0

Packages:

abseil-cpp: 20210324.2
absl-py 2.0.0
arrow-cpp 5.0.0
pastor 0.8.1
aws-c-cal 0.5.11
aws-c-common 0.6.2
aws-c-event-stream 0.2.7
aws-c-io 0.10.5
aws-checksums 0.1.11
aws-sdk-cpp 1.8.186
backcall 0.2.0
backports 1.0
backports.functools_lru_cache 1.6.5
bzip2 1.0.8
c-ares 1.21.0
ca-certificates 2023.11.17
cairo 1.14.12
colorama 0.4.5
cycler 0.11.0
dataclasses 0.8
decorator 5.1.1
freetype 2.12.1
gast 0.5.4
gflags 2.2.2
glog 0.5.0
google-pasta 0.2.0
grpc-cpp 1.40.0
grpcio 1.38.1
h5py 2.8.0
hdf5 1.10.4
icu 64.2
]importlib-metadata 4.8.1
intel-openmp 2023.2.0
ipython 7.1.1
ipython_genutils 0.2.0
jedi 0.13.3
jpeg 9e
jpype1 0.6.3
jsonschema 4.1.2
jupyter_core 4.8.1
keras 2.2.4
keras-applications 1.0.8
keras-preprocessing 1.1.2
kiwisolver 1.3.1
krb5 1.20.1
lerc 3.0
libblas 3.9.0
libbrotlicommon 1.0.9
libbrotlidec 1.0.9
libbrotlienc 1.0.9
libcblas 3.9.0
libclang 9.0.1
libcurl 8.1.2
libdeflate 1.10
libgpuarray 0.7.6
libhwloc 2.9.3
libiconv 1.17
liblapack 3.9.0
libpng 1.6.39
libprotobuf 3.18.0
libsqlite 3.44.0
libssh2 1.10.0
libthrift 0.15.0
libtiff 4.3.0
libutf8proc 2.8.0
libxml2 2.11.6
libzlib 1.2.13
lz4-c 1.9.3
m2w64-gcc-libgfortran 5.3.0
m2w64-gcc-libs 5.3.0
m2w64-gcc-libs-core 5.3.0
m2w64-gmp 6.1.0
m2w64-libwinpthread-git 5.0.0.4634.697f757
mako 1.3.0
markdown 3.5.1
markupsafe 2.0.1
matplotlib 3.0.3
matplotlib-base 3.0.3
mkl 2023.2.0
msys2-conda-epoch 20160418
nbformat 4.4.0
numpy 1.19.5
olefile 0.46
onnx 1.4.1
onnx-tf 1.2.1
openssl 1.1.1w
pandas 0.23.4
parquet-cpp 1.5.1
parso 0.7.1
pickleshare 0.7.5
pillow 5.3.0
pip 21.3.1
pixman 0.38.0
prompt-toolkit 3.0.36
prompt_toolkit 3.0.36
protobuf 3.18.0
pthreads-win32 2.9.1
py4j 0.10.9.7
pyarrow 5.0.0
pygments 2.14.0
pygpu 0.7.6
pyparsing 3.1.1
pyqt 5.12.3
pyqt5-sip 4.19.18
pyqtwebengine 5.12.1
pyreadline 2.1
persistent 0.17.3
python 3.6.15
python-dateutil 2.7.5
python_abi 3.6
pytz 2023.3.post1
pywin32 301
pyyaml 5.4.1
qt 5.12.5
re2 2021.09.01
scipy 1.5.3
setuptools 58.0.4
six 1.16.0
snappy 1.1.10
sqlite 3.44.0
tbb 2021.10.0
tensorboard 1.14.0
tensorflow 1.14.0
tensorflow-base 1.14.0
tensorflow-estimator 1.14.0
termcolor 1.1.0
theano 1.0.5
tk 8.6.13
tornado 6.1
traitlets 4.3.3
typing 3.7.4.3
typing_extensions 4.1.1
ucrt 10.0.22621.0
vc 14.3
vc14_runtime 14.36.32532
vs2015_runtime 14.36.32532
vs2017_win-64 19.16.27033
vswhere 3.1.4
wcwidth 0.2.10
werkzeug 2.0.2
wheel 0.37.1
wrapt 1.13.1
xz 5.2.6
yaml 0.2.5
zipp 3.6.0
zlib 1.2.13
zstd 1.5.5

[amsamms]

I have the same problem

No solution yet ???

[mlauber71]

@amsamms @PAI I would encourage you to read my suggested article and try the steps suggested there. Currently there is no easy solution if the default installation of deep learning libraries does not work.

Keras Library installation (Error Metaclass)

@Deschwaba depending on your operating system and KNIME version installing Deep Learning nodes can be a challenge. I can offer up this information and article:

You can try and use my preconfigured nodes but there is no guarantee this will work out of the box:

AutoML - All Models Post Process Component Failing

It should work on windows. I also have two meta nodes with Conda Environment Propagation for Keras and TensorFlow for Windows. I have not tested them for some time though.

[amsamms]

I encountered the same error as described here when I opted to install a new environment using the CPU.

However, when I switched to installing a new environment with the GPU, the error was resolved, and KNIME DL started working flawlessly

[Deschwaba]

Thanks for your reference, it’s working!