Python Environment and Package Versions
we will try to install the most current packages as of December 2024 to run this example
In order to get keras version 3, we need to install a tensorflow version 2.16 or later.
Creating a defined environment
conda create -n vision python=3.10
Then activate this environment before installing packages!
conda activate vision
Install specific packages:
conda install -c conda-forge tensorflow=2.17.0 -y
conda install -c conda-forge matplotlib=3.10.0 -y
conda install -c conda-forge pandas=2.2.3 -y
conda install -c conda-forge progress
conda install -c conda-forge pydot -y
conda install -c conda-forge tensorflow-datasets -y
Notes:
When I install python version 3.13.1, then conda will not let me install tensorflow 2.17.
(Error LibMambaUnsatisfiableError: Encountered problems while solving:
- nothing provides __cuda needed by tensorflow-2.17.0 )
However, python 3.10 works with tensorflow 2.17
