From the course: AWS Certified Machine Learning Engineer Associate (MLA-C01) Cert Prep
Unlock this course with a free trial
Join today to access over 24,700 courses taught by industry experts.
Hyperparameter tuning (automatic model tuning)
From the course: AWS Certified Machine Learning Engineer Associate (MLA-C01) Cert Prep
Hyperparameter tuning (automatic model tuning)
- [Instructor] Hello guys, and welcome again. So in today's lesson, we are going to talk about the hyperparameter tuning. And the hyperparameter tuning is squeezing your accuracy of your model. So you just want those few extra percentage increase in the accuracy or the precision order recall. So if you do so, if you want so, you need to perform the hyperparameter tuning. So hyperparameter tuning is also known as automatic model tuning. It kind of finds the best version of your model by running many training jobs using the hyperparameter ranges that you specify. So you would want to use a hyperparameter tuning job if you want to get these extra scores in a specific metric, let's say the accuracy, the MSE, the precision, the recall, and any of these metrics. So the hyperparameter tuning search for the best combination of metrics from the ranges that you specify in order to output the best score, which is the specific metric that you specify. And before you start hyperparameter tuning…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
-
-
(Locked)
Intro: Modelling (SageMaker built-in algorithms)1m 3s
-
Amazon SageMaker, SageMaker Studio12m 10s
-
(Locked)
Hands-on learning: Amazon SageMaker walkthrough2m 54s
-
(Locked)
Hands-on learning: Create an Amazon SageMaker notebook instance4m 35s
-
(Locked)
Built-in algorithms overview4m 19s
-
(Locked)
Linear Learner8m 27s
-
(Locked)
XGBoost5m 1s
-
(Locked)
LightGBM7m 5s
-
(Locked)
K-Nearest Neighbours4m
-
(Locked)
Factorization Machines4m 38s
-
(Locked)
DeepAR5m 13s
-
(Locked)
Image classification6m 4s
-
(Locked)
Object detection3m 38s
-
Semantic segmentation4m 13s
-
(Locked)
Seq2Seq3m 49s
-
(Locked)
BlazingText5m 8s
-
(Locked)
Neural Topic Model (NTM)2m 38s
-
(Locked)
Latent Dirichlet Allocation (LDA)1m 55s
-
(Locked)
Random Cut Forest (RCF)3m 27s
-
(Locked)
K-means clustering3m 24s
-
(Locked)
Hierarchical clustering8m 36s
-
Object2Vec5m 59s
-
(Locked)
Principal Component Analysis (PCA)2m 22s
-
(Locked)
IP Insights4m
-
(Locked)
Reinforcement learning4m 13s
-
(Locked)
Built-in algorithms recap4m 27s
-
(Locked)
Hyperparameter tuning (automatic model tuning)6m 6s
-
(Locked)
Hands-on learning: Hyperparameter tuning job3m 22s
-
(Locked)
Exam cram6m 58s
-
(Locked)
-
-
-
-
-
-