I'm trying to train ssd mobileNet v3 with tensorflow,but the error( ImportError: cannot import name 'estimator' from 'tensorflow.compat.v1' (/usr/local/lib/python3.11/dist-packages/tensorflow/_api/v2/compat/v1/__init__.py)) happens

TensorFlow
, , ,
1

hello, i’m using goole colab and i download (tensorflow==2.18.0) someone said downgrade the tensorflow version so i run this (!pip uninstall -y tensorflow tensorflow-estimator
!pip install tensorflow==2.12 tensorflow-estimator==2.12) but this error occurs :
(ERROR: pip’s dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
albumentations 2.0.5 requires numpy>=1.24.4, but you have numpy 1.23.5 which is incompatible.
albucore 0.0.23 requires numpy>=1.24.4, but you have numpy 1.23.5 which is incompatible.
treescope 0.1.9 requires numpy>=1.25.2, but you have numpy 1.23.5 which is incompatible.
pymc 5.21.1 requires numpy>=1.25.0, but you have numpy 1.23.5 which is incompatible.
tf-keras 2.18.0 requires tensorflow<2.19,>=2.18, but you have tensorflow 2.12.0 which is incompatible.
imbalanced-learn 0.13.0 requires numpy<3,>=1.24.3, but you have numpy 1.23.5 which is incompatible.
chex 0.1.89 requires numpy>=1.24.1, but you have numpy 1.23.5 which is incompatible.
bigframes 1.40.0 requires numpy>=1.24.0, but you have numpy 1.23.5 which is incompatible.
tensorflow-text 2.18.1 requires tensorflow<2.19,>=2.18.0, but you have tensorflow 2.12.0 which is incompatible.
scikit-image 0.25.2 requires numpy>=1.24, but you have numpy 1.23.5 which is incompatible.
orbax-checkpoint 0.11.8 requires jax>=0.5.0, but you have jax 0.4.30 which is incompatible.
xarray 2025.1.2 requires numpy>=1.24, but you have numpy 1.23.5 which is incompatible.
blosc2 3.2.0 requires numpy>=1.26, but you have numpy 1.23.5 which is incompatible.
Successfully installed gast-0.4.0 google-auth-oauthlib-1.0.0 jax-0.4.30 jaxlib-0.4.30 keras-2.12.0 numpy-1.23.5 tensorboard-2.12.3 tensorflow-2.12.0 tensorflow-estimator-2.12.0 wrapt-1.14.1
WARNING: The following packages were previously imported in this runtime:
[numpy]
You must restart the runtime in order to use newly installed versions.)
i tried so many things but the dependency conflicts always occurs. what should i do?

3

Hi @Leena, Thanks for posting this issue.

Downgrading to TensorFlow to 2.12 is causing conflicts with other pre-installed packages that expect TensorFlow 2.18.0 and newer versions of NumPy.

I would recommend upgrading to tensorflow 2.18 and work with packages that are compatible with this version. You install tensorflow 2.18 using

!pip install tensorflow==2.18
print(tensorflow.__version__)

Let me know if that works for you.