Hyperparameter Optimization Using Dask with ORÍON | Xavier Bouthillier | Dask Summit 2021

preview_player
Показать описание
Oríon is a framework for asynchronous hyperparameter optimization (HPO) built around two main principles: 1) HPO should be effortless to execute in common machine learning workflows 2) New HPO algorithms should be readily available for practitioners. In this talk we will present Oríon and its core design principles, followed by an integration example with Dask demonstrating its simplicity of use.

===

The Dask Distributed Summit is where users, contributors, and newcomers can share experiences to learn from one another and grow together. The Dask Distributed Summit provides content, information, and learning opportunities for attendees of all levels of Dask familiarity and expertise.

  1. Dask
Рекомендации по теме
Hyperparameter Optimization Using 0:28:35

Hyperparameter Optimization Using Dask with ORÍON | Xavier Bouthillier | Dask Summit 2021

Better and Faster 0:27:33

Better and Faster Hyper Parameter Optimization with Dask | SciPy 2019 | Scott Sievert

Dask and Optuna 0:13:53

Dask and Optuna for Hyper Parameter Optimization

PyData Madison Meetup: 1:16:14

PyData Madison Meetup: Better and Faster Hyperparameter Optimization with Dask

An Introduction to 0:29:42

An Introduction to Distributed Hybrid Hyperparameter Optimization- Jun Liu | SciPy 2022

Scaling Hyperparameter Tuning 0:49:21

Scaling Hyperparameter Tuning with Dask-ML on Clusters

XGBoost and HyperParameter 0:07:42

XGBoost and HyperParameter Optimization

Dan Ryan: Efficient 0:26:03

Dan Ryan: Efficient and Flexible Hyperparameter Optimization | PyData Miami 2019

Dask Use Case 0:01:55

Dask Use Case | Who Uses Dask: CapitalOne

Data Pipeline Hyperparameter 0:30:34

Data Pipeline Hyperparameter Optimization - Alex Quemy

Mastering Hyperparameter Tuning 0:28:15

Mastering Hyperparameter Tuning with Optuna: Boost Your Machine Learning Models!

Optuna: A Define 0:26:37

Optuna: A Define by Run Hyperparameter Optimization Framework | SciPy 2019 |

Training Pytorch Models 0:30:09

Training Pytorch Models Faster with Dask | Scott Sievert | Dask Summit 2021

Tom Augspurger: Scalable 0:38:28

Tom Augspurger: Scalable Machine Learning with Dask | PyData New York 2019

Lexalytics - Hyper 0:03:35

Lexalytics - Hyper Parameter Optimization

Hyperparameter Optimization (HPO) 0:16:40

Hyperparameter Optimization (HPO) with RAPIDS on Microsoft AzureML and NVIDIA GPUs

Accelerating MLFlow Hyper-parameter 0:28:56

Accelerating MLFlow Hyper-parameter Optimization Pipelines with RAPIDS

Autotune: A Derivative-free 0:04:45

Autotune: A Derivative-free Optimization Framework for Hyperparameter Tuning

Tuning ML Models: 0:23:43

Tuning ML Models: Scaling, Workflows, and Architecture

Dask and XGBoost 0:05:53

Dask and XGBoost Experiment

Optuna: A Define-by-Run 0:19:24

Optuna: A Define-by-Run Hyperparameter Optimization Framework | SciPy Japan | Shotaro Sano, et al

Fugue Tune: Distributed 0:36:56

Fugue Tune: Distributed Hybrid Hyperparameter Tuning

Data Science + 0:31:33

Data Science + ML | Clusters of Clusters | Grant Gelven | Walmart | Dask Summit 2021 Keynote

Pandas Parallel Transformations 0:26:56

Pandas Parallel Transformations with Dask and Pandarallel