llama-optimus is a lightweight Python tool to automatically optimize llama.cpp performance flags for maximum throughput. Maximize your tokens/s for prompt processing (pp) & token generation (tg).
A Python machine learning pipeline that downloads a dataset, prepares it with transformations and a train/test split, trains a LightGBM binary classifier using Optuna for hyperparameter tuning, and ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results