ResolutionImpossible error while installing tfx library in macos Ventura with Apple M2

[e-compagno]

System information

• Have I specified the code to reproduce the issue (Yes, No): Yes
• Environment in which the code is executed (e.g., Local(Linux/MacOS/Windows),
  Interactive Notebook, Google Cloud, etc): MacOS
• TensorFlow version:
• TFX Version: any
• Python version: 3.9
• Python dependencies (from pip freeze output):

Describe the current behavior

Cannot install tfx on MacOS 13.4.1 (Ventura) with Apple M2 silicon processor. I have tried both with python 3.9 from brew and from miniconda.

Describe the expected behavior

The tfx library should be installed via pip. What I get is the following error

ERROR: Cannot install tfx==0.13.0, tfx==0.14.0, tfx==0.15.0, tfx==0.21.0, tfx==0.21.1, tfx==0.21.2, tfx==0.21.3, tfx==0.21.4, tfx==0.21.5, tfx==0.22.0, tfx==0.22.1, tfx==0.22.2, tfx==0.23.0, tfx==0.23.1, tfx==1.10.0, tfx==1.11.0, tfx==1.12.0, tfx==1.13.0, tfx==1.8.0, tfx==1.8.1, tfx==1.9.0 and tfx==1.9.1 because these package versions have conflicting dependencies.

The conflict is caused by:
    tfx 1.13.0 depends on ml-metadata<1.14.0 and >=1.13.1
    tfx 1.12.0 depends on tensorflow<2.12 and >=2.11.0
    tfx 1.11.0 depends on tensorflow!=2.0.*, !=2.1.*, !=2.2.*, !=2.3.*, !=2.4.*, !=2.5.*, !=2.6.*, !=2.7.*, !=2.8.*, !=2.9.*, <2.11 and >=1.15.5
    tfx 1.10.0 depends on tensorflow!=2.0.*, !=2.1.*, !=2.2.*, !=2.3.*, !=2.4.*, !=2.5.*, !=2.6.*, !=2.7.*, !=2.8.*, <2.10 and >=1.15.5
    tfx 1.9.1 depends on tensorflow!=2.0.*, !=2.1.*, !=2.2.*, !=2.3.*, !=2.4.*, !=2.5.*, !=2.6.*, !=2.7.*, !=2.8.*, <2.10 and >=1.15.5
    tfx 1.9.0 depends on tensorflow!=2.0.*, !=2.1.*, !=2.2.*, !=2.3.*, !=2.4.*, !=2.5.*, !=2.6.*, !=2.7.*, !=2.8.*, <2.10 and >=1.15.5
    tfx 1.8.1 depends on tensorflow!=2.0.*, !=2.1.*, !=2.2.*, !=2.3.*, !=2.4.*, !=2.5.*, !=2.6.*, !=2.7.*, <2.9 and >=1.15.5
    tfx 1.8.0 depends on tensorflow!=2.0.*, !=2.1.*, !=2.2.*, !=2.3.*, !=2.4.*, !=2.5.*, !=2.6.*, !=2.7.*, <2.9 and >=1.15.5
    tfx 0.23.1 depends on tfx-bsl<0.24 and >=0.23
    tfx 0.23.0 depends on tfx-bsl<0.24 and >=0.23
    tfx 0.22.2 depends on ml-metadata<0.23 and >=0.22.1
    tfx 0.22.1 depends on ml-metadata<0.23 and >=0.22.1
    tfx 0.22.0 depends on tfx-bsl<0.23 and >=0.22
    tfx 0.21.5 depends on tfx-bsl<0.22 and >=0.21.3
    tfx 0.21.4 depends on tfx-bsl<0.22 and >=0.21.3
    tfx 0.21.3 depends on tfx-bsl<0.22 and >=0.21.3
    tfx 0.21.2 depends on ml-metadata<0.22 and >=0.21.2
    tfx 0.21.1 depends on tfx-bsl<0.22 and >=0.21.3
    tfx 0.21.0 depends on tensorflow-data-validation<0.22 and >=0.21
    tfx 0.15.0 depends on ml-metadata<0.16 and >=0.15.0rc0
    tfx 0.14.0 depends on ml-metadata<0.15 and >=0.14
    tfx 0.13.0 depends on ml-metadata<0.14 and >=0.13.2

To fix this you could try to:
1. loosen the range of package versions you've specified
2. remove package versions to allow pip attempt to solve the dependency conflict

ERROR: ResolutionImpossible: for help visit https://pip.pypa.io/en/latest/topics/dependency-resolution/#dealing-with-dependency-conflicts

Standalone code to reproduce the issue

python3.9 -m venv venv
source venv/bin/activate
pip install tfx

[singhniraj08]

@e-compagno,

Apple M1 & M2 support is a work for progress. Team is working on priority for this, but there are internal infrastructure reasons for why this has proceeded slowly.

Similar request #5804 is already present to track this feature. Requesting you to close this issue and follow similar thread for updates. Thank you!

[e-compagno]

Hello @singhniraj08, thank you for your reply. Of course, given it's a duplicate issue I am closing this ticket

[singhniraj08]

@e-compagno, Thank you for confirmation. Closing this issue, Please take a look into the answers provided above, feel free to reopen and post your comments(if you still have queries on this). Thank you!

[github-actions]

Are you satisfied with the resolution of your issue?
Yes
No