I’ve recently been diving into some cutting-edge tools for model optimization and it’s been a game changer for my workflow. For example, using AutoML has significantly reduced the time I spend on hyperparameter tuning, allowing me to focus more on model architecture and data quality. I’m curious if anyone else has similar experiences or recommends specific tools that have boosted your training efficiency.
I totally get that! AutoML sped up my hyperparameter tuning too, but I still find manual tweaking on smaller datasets can enhance model performance. Have you tried experimenting with Bayesian optimization for that? 
.
It’s interesting to hear how AutoML has saved you time on hyperparameter tuning. I’ve found that using transfer learning on smaller datasets can really help, too, especially to kickstart model performance without extensive tuning. Have you thought about blending that approach with your current workflow?