incompatible dependency versions after installing v4.2.0

[ShahrzadZolghadr]

Hello, thanks for the great tf ranking project. I want to use TFR BERT in a project, but I have not able to use tensorflow-ranking yet. I am using colab. at first I used the simple !pip install tensorflow_ranking instruction but the running process continued a long time. It seems it is because of the pip new version. after that I used the below instruction:

!pip install --use-deprecated=legacy-resolver tensorflow_ranking This one took a short time but at the end this error happened:

ERROR: pip's legacy dependency resolver does not consider dependency conflicts when selecting packages. This behaviour is the source of the following dependency conflicts. tensorflow-text 2.6.0 requires tensorflow<2.7,>=2.6.0, but you'll have tensorflow 2.7.0 which is incompatible.

So, I cannot import the tfr and when I import it the colab session crashes. What should I do? Is there anyone who experience something like this on colab?