Update Colab Base image to colab_20250219-060225_RC01 #1475
Conversation
calderjo
changed the title
| self.assertEqual(0, result.returncode) | ||
| self.assertIsNone(result.stderr) | ||
| self.assertIn("Download a model from the Hugging Face Hub or Kaggle Model Hub.", result.stdout.decode("utf-8")) |
There was a problem hiding this comment.
Thanks! @neshdev Does this look sufficient to protect against loading the wrong tune lib in the future?
There was a problem hiding this comment.
For context: This test was passing due to that fact we only check that the tune --help command went to completion with no specifics.
Since ray and torch both have --help flags :lol-sob: , it passed.
hence the added output check.
| # See https://github.com/keras-team/keras/commit/dcefb139863505d166dd1325066f329b3033d45a | ||
| # Colab base is on Keras 3.8, we have to install the package separately | ||
| RUN uv pip install --system google-cloud-automl==1.0.1 google-cloud-aiplatform google-cloud-translate==3.12.1 \ | ||
| google-cloud-videointelligence google-cloud-vision google-genai "keras<3.6" |
There was a problem hiding this comment.
@rosbo Any concerns here, do we need a newer Keras?
|
We can follow up on "upgrading" Keras version when Vincent is back from ooo in subsequent PRs Would be nice to get a head start and see if any additional changes are needed for learn tools as a result up base image upgrade. which is only run in SDG build. So i will merge this in as is, but keeping the above topic open! |
We are upgrading the base image to the latest release image by colab: colab_20250219-060225_RC01
Which includes the following upgrades:
TF 2.18
Python 3.11
Cuda 12.5
This PR includes a handful of fixes to resolve conflicts related to these upgrade. Notably issues pertaining torch and cudnn.
We also bumped lightgbm version as well
We also included a fix to tune cli package conflict.