Skip to content

Conversation

@yongwww
Copy link
Member

@yongwww yongwww commented Feb 27, 2025

The ml_dtypes>=0.5.1 was installed in the install python packages script, but the tensorflow 2.18 installation later will uninstall the ml_dtypes and install an old ml_dtypes version 0.4 instead. (https://ci.tlcpack.ai/blue/rest/organizations/jenkins/pipelines/tvm-docker/branches/main/runs/2062/nodes/50/steps/114/log/?start=0). We can install ml_dtypes right after tensorflow.

@yongwww
Copy link
Member Author

yongwww commented Feb 27, 2025

#17690 tried to run the ci with manually built images by the fix, CPU and GPU checks were passed.

@yongwww yongwww mentioned this pull request Feb 27, 2025
@yongwww yongwww changed the title [Docker] Fix ml dtypes downgrade issue introduced by TensorFlow [Docker] Fix ml_dtypes downgrade issue introduced by TensorFlow Feb 27, 2025
@MasterJH5574 MasterJH5574 merged commit 410cc2f into apache:main Feb 28, 2025
15 checks passed
@yongwww yongwww deleted the fix_ml_dtypes branch February 28, 2025 16:46
ShiboXing pushed a commit to ShiboXing/tvm that referenced this pull request Aug 10, 2025
…he#17691)

The ml_dtypes>=0.5.1 was installed in the install python packages
script, but the tensorflow 2.18 installation later will uninstall the
ml_dtypes and install an old ml_dtypes version 0.4 instead. (https://ci.tlcpack.ai/blue/rest/organizations/jenkins/pipelines/tvm-docker/branches/main/runs/2062/nodes/50/steps/114/log/?start=0).
We can install ml_dtypes right after tensorflow.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants