google-research / google-research/task_adaptation
tensorflow-datasets version problems
- Dominant language
- Python
- Stars
- 352
- Forks
- 37
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
I have some trouble running VTAB. Using recommended `pip install -e .` I end up with the following package versions:
- tensorflow==1.15.0
- tensorflow-datasets==4.0.1
After running `python cifar_test.py` I get the following error:
`
ImportError: This version of TensorFlow Datasets requires TensorFlow version >= 2.1.0; Detected an installation of version 1.15.0. Please upgrade TensorFlow to proceed.
`
From what I understand, the lastest `tensorflow-datasets` version that works with `tensorflow==1.15` is `tensorflow-datasets==3.2.1`, but downgrading results in other problems while downloading data.
[Here](https://colab.research.google.com/drive/19wsVq11138DBo8xo9L2QhAwckbAtOMxY?usp=sharing) is Google Colab notebook that reproduces problems with original installation, and [here](https://colab.research.google.com/drive/1lzgkECjcVxuRwOtiCr4TVdz4R-DiEJLl?usp=sharing) is notebook that reproduces problems with downgrading tensorflow-datasets to v3.2.1.
Contributor guide
Assessment
This issue has not been assessed yet.