Issue with AlreadyExistsError: Another metric with the same name already exists. about analog-watch-recognition

[akucia]

I tried installing everything from scratch in a brand new venv and I didn't run into any issues.
There are a couple of things we can check:

• Which Python version are you using?

• Can you post the full list of installed packages? (output of pip freeze)

• Did you try installing watch_recognition module without building .whl ? (simple pip install ./watch_recognition/)

• Can you post a full error stack trace?

[ScriptHound]

Im using python 3.8.13
The list of installed packages:

absl-py==0.15.0
albumentations==1.0.3
asttokens==2.0.5
astunparse==1.6.3
backcall==0.2.0
build==0.7.0
cachetools==5.0.0
certifi==2021.10.8
charset-normalizer==2.0.12
clang==5.0
click==8.0.1
cycler==0.11.0
debugpy==1.5.1
decorator==5.1.1
distinctipy==1.1.5
efficientnet==1.0.0
entrypoints==0.4
executing==0.8.3
flatbuffers==1.12
gast==0.4.0
google-api-core==2.7.1
google-auth==2.6.2
google-auth-oauthlib==0.4.6
google-cloud-core==2.2.3
google-cloud-storage==2.2.1
google-crc32c==1.3.0
google-pasta==0.2.0
google-resumable-media==2.3.2
googleapis-common-protos==1.56.0
grpcio==1.45.0
h5py==3.1.0
idna==3.3
image-classifiers==1.0.0
imageio==2.15.0
importlib-metadata==4.11.3
ipykernel==6.9.2
ipython==8.1.1
jedi==0.18.1
joblib==1.1.0
jupyter-client==7.1.2
jupyter-core==4.9.2
keras==2.8.0
Keras-Applications==1.0.8
Keras-Preprocessing==1.1.2
kiwisolver==1.4.0
Markdown==3.3.6
matplotlib==3.4.3
matplotlib-inline==0.1.3
nest-asyncio==1.5.4
networkx==2.7.1
numpy==1.19.5
oauthlib==3.2.0
opencv-python-headless==4.5.5.64
opt-einsum==3.3.0
packaging==21.3
pandas==1.3.1
parso==0.8.3
pep517==0.12.0
pexpect==4.8.0
pickleshare==0.7.5
Pillow==8.3.2
prompt-toolkit==3.0.28
protobuf==3.19.4
psutil==5.9.0
ptyprocess==0.7.0
pure-eval==0.2.2
pyasn1==0.4.8
pyasn1-modules==0.2.8
pycocotools==2.0.4
Pygments==2.11.2
pyparsing==3.0.7
python-dateutil==2.8.2
pytz==2022.1
PyWavelets==1.3.0
PyYAML==6.0
pyzmq==22.3.0
requests==2.27.1
requests-oauthlib==1.3.1
rsa==4.8
scikit-image==0.18.2
scikit-learn==0.24.2
scipy==1.8.0
segmentation-models==1.0.1
six==1.16.0
stack-data==0.2.0
tensorboard==2.8.0
tensorboard-data-server==0.6.1
tensorboard-plugin-wit==1.8.1
tensorflow==2.6.0
tensorflow-estimator==2.8.0
termcolor==1.1.0
threadpoolctl==3.1.0
tifffile==2022.3.16
tomli==2.0.1
tornado==6.1
tqdm==4.62.3
traitlets==5.1.1
typing-extensions==3.7.4.3
urllib3==1.26.9
watch-recognition @ file:///home/scripthound/analog_watch_recognition/analog-watch-recognition/watch_recognition/dist/watch_recognition-0.0.1-py3-none-any.whl
wcwidth==0.2.5
Werkzeug==2.0.3
wrapt==1.12.1
zipp==3.7.0

I have also tried installing without building the wheel and got the same traceback.

Traceback:
traceback.txt

[akucia]

I think I saw similar errors when segmentation models had issues with multiple keras sources (you can directly import keras module or import it from Tensorflow via tf.keras). That's why there's an env var being set in the top of the notebook, but maybe it's not working correctly in your case.

os.environ['SM_FRAMEWORK'] = 'tf.keras'

In the 4th notebook cell, where the error happens you should see log message:

Segmentation Models: using `tf.keras` framework.

Does similar error happen if you just run the segmentation_models import in a clean notebook?
import segmentation_models as sm

[akucia]

Also, check out this Stack Overflow thread
https://stackoverflow.com/questions/58012741/error-importing-tensorflow-alreadyexistserror-another-metric-with-the-same-nam

[ScriptHound]

Ive just changed version of Keras to 2.6.0 and it worked. Thank you very much. Ill create a new issue if something happen again.