Quick question with building my own model for hls4ml

[Viggo-Avila]

I'm planning to make my own model using QKeras. Is it fine if I make it using the environment.yml of the hls4ml-tutorial? Because I was wondering if the difference of the versions of the dependencies of hls4ml and the ones used when making my own model (let's say my python version for hls4ml is 3.10.14 and my python version for making the model is 3.14.0) will have an affect? Because if the packages need to be the same version both for making the model and converting to hls4ml, then wouldnt it be better to use the environment.yml of the tutorial?

[vloncar]

Python version shouldn't matter, just make sure TF version is the same since we rely on the format of the saved model to be able to parse it and convert it to firmware.