I wanted to understand how neural networks and model training actually worked, so before starting Tensorless, I built two AI models with PyTorch.
Those two projects were mainly for learning. I wanted to understand the basics by actually building things instead of only reading tutorials.
If I already have the training data, why can't I just give it to a framework and let it figure out most of the boring stuff?
Before Tensorless existed, I spent time learning PyTorch by actually building two AI models.
That gave me a basic understanding of how the whole process worked: Preparing data Creating a model Training Loss Optimizers Batching Saving models Loading models Inference And all the configuration around training
Building those two models was important because I wasn't starting Tensorless completely blind.
And after doing it twice, I started noticing how much of the setup was repetitive.
Give Tensorless your training data ↓ Tensorless figures out the setup ↓ Train ↓ Get your model
If you're a beginner, you shouldn't need to understand every training parameter before you can train a model.
But if you're an advanced programmer, you should still be able to configure everything yourself.
The configuration was mostly determined automatically using fixed conditions in the code.
So instead of manually deciding every parameter, Tensorless would make those decisions for you.
One of the reasons I started thinking differently about the architecture was the amount of stuff involved in a PyTorch-based workflow.
When I was experimenting with the project, the dependencies and modules could become 500+ MB.
"I'm trying to make training simpler and lightweight, but I'm bringing in a huge stack just to do it."
I didn't want Tensorless to become a massive package full of components that users might never need.
I wanted the package to contain the things actually required for its own workflow.
Instead of keeping Tensorless as a PyTorch wrapper forever, I started working toward my own neural/training system.