Optimizing Model Compilers
Luca discusses the evolution of model compilers, highlighting Thunder as a less mature option compared to established tools like Torch compile and Jax. He emphasizes the importance of granular control in optimization, which allows practitioners to efficiently manage resources without excessive abstraction. The balance between flexibility and user autonomy is crucial for the success of these tools.In this clip
From this podcast

Super Data Science: ML & AI Podcast with Jon Krohn
831: PyTorch Lightning, Lit-Serve and Lightning Studios — with Dr. Luca Antiga
Related Questions