综合技术

Hyperparameter Tuning with MLflow

微信扫一扫,分享到朋友圈

Hyperparameter Tuning with MLflow
0

Apache Spark MLlib users often tune hyperparameters using MLlib’s built-in tools CrossValidator
and  TrainValidationSplit
.  These use grid search to try out a user-specified set of hyperparameter values; see the  Spark docs on tuning
for more info.

Databricks Runtime 5.3 and 5.3 ML and above support automatic MLflow tracking for MLlib tuning in Python.

With this feature, PySpark
CrossValidator
and  TrainValidationSplit
will automatically log to MLflow, organizing runs in a hierarchy and logging hyperparameters and the evaluation metric.  For example, calling  CrossValidator.fit()
will log one parent run.  Under this run,  CrossValidator
will log one child run for each hyperparameter setting, and each of those child runs will include the hyperparameter setting and the evaluation metric.  Comparing these runs in the MLflow UI helps with visualizing the effect of tuning each hyperparameter.

展开阅读全文

微信扫一扫,分享到朋友圈

Hyperparameter Tuning with MLflow
0

AdTech Sucks: This Time It’s Personal IDs

上一篇

融资117亿,如今Telegram的lEO却是一次大鲸出货?

下一篇

你也可能喜欢

评论已经被关闭。

插入图片