Bigquery Ml - Machine Learning In Sql Using Google Bigquery

Create Machine Learning models in Google Cloud Big Query using standard SQL. Big query ML course for ML & Data engineers

What you'll learn
BigQuery ML - Learn Machine Learning in Google Cloud using BigQuery.
Learn to Train, Evaluate, Inference, Tune and Explain Machine leaning models using standard SQL with Big Query.
Theory + BigQuery ML implementation of many Machine learning algorithms.
Detailed theory for each of the ML algorithm with a Real-world example implementation in BigQuery ML.
Linear regression, Logistic regression, K-means clustering, Boosted Tree.
Deep neural networks, ARIMA+ Time series Forecasting, Matrix Factorization, PCA.
Hyperparameter tuning of models, Model Explainability functions, Feature pre-processing functions, model management operations in BigQuery ML.
Requirements
Basic knowledge of SQL.
Description
"BigQuery ML lets you create and execute machine learning models in BigQuery using standard SQL queries."Big Query ML is a blessing for engineers who want to work in Machine Learning domain but lack programming language like Python, R. With Big Query ML, they can use their existing SQL knowledge to build operational production-grade Machine learning models.What's included in the course ?Brief introduction to various Machine Learning services of Google Cloud.Fundamentals of BigQuery ML and challenges which it solves. All of the Machine Learning algorithms are explained in 2 Steps :Step 1 : Theoretical explanation of working of an ML algorithm. Step 2 : Practical implementation of the ML algorithm in BigQuery ML.Each and every Machine learning algorithm is explained with HANDS-ON examples.Hyperparameter tuning of models, Model Explainability functions, Feature pre-processing functions.Model management operations using bq commands.BigQuery ML pricing (Flat rate & On-demand pricing models).Assignment for each Machine learning algorithm for self Hands-On in Big Query ML.Learn Best practices and Optimization techniques for BigQuery ML.Machine Learning algorithms explained:Linear regressionLogistic regressionK-means clustering Boosted TreeDeep neural networks ARIMA+ Time series Forecasting Product Component Analysis (PCA)Matrix FactorizationAfter completing this course, you can confidently start creating production-grade Machine Learning models in Real-world corporate projects using BigQuery ML.Add-OnsQuestions and Queries will be answered very quickly.Queries, datasets and references used in lectures are attached in the course for your convenience.I am going to update it frequently, every time adding new components of Bigquery ML.

Overview

Section 1: Introduction to GCP

Lecture 1 Introduction to Google Cloud Platform

Lecture 2 GCP vs AWS vs Azure - Why choose GCP

Lecture 3 AI & ML services in Google Cloud

Section 2: BigQuery ML (BQML) introduction

Lecture 4 What is BigQuery ML

Lecture 5 Conventional ML challenges and How Big query is addressing them

Lecture 6 BigQuery ML Features

Lecture 7 Advantages of BigQuery ML

Lecture 8 Lifecycle/Workflow of a BigQuery ML Project

Lecture 9 BQML supported models

Section 3: BigQuery Basics - Crash course

Lecture 10 Announcement

Lecture 11 Setup a GCP account

Lecture 12 Important Note

Lecture 13 Create a Project

Lecture 14 BigQuery UI Tour

Lecture 15 Create a Dataset

Lecture 16 Create a Table

Section 4: Linear Regression

Lecture 17 What is Linear regression - Part 1

Lecture 18 What is Linear regression - Part 2

Lecture 19 High-level view of Create Model query

Lecture 20 Limitations of Create model query

Lecture 21 Linear regression Example Use case

Lecture 22 Basic Options in Create model query

Lecture 23 Overfitting problem

Lecture 24 L2/Ridge regularization

Lecture 25 L1/Lasso regularization

Lecture 26 Gradient Descent Optimize Strategy

Lecture 27 Types of Gradient Descent

Lecture 28 Learn rate Option

Lecture 29 Other Options in Create model query

Lecture 30 Model Training - Write Create model Query for Linear regression

Lecture 31 Exploring Model details

Lecture 32 Model Evaluation Query (ML.EVALUATE)

Lecture 33 Model Training - Optimize Create Model Query

Lecture 34 ML.TRAINING_INFO Function

Lecture 35 Model Prediction (ML.PREDICT)

Section 5: Hyperparameter Tuning in BigQuery

Lecture 36 What is Hyperparameter Tuning ?

Lecture 37 Hyperparameter Tuning Options in BigQuery

Lecture 38 Tune the Linear regression model

Lecture 39 ML.TRIAL_INFO Function

Section 6: Model Explainability Functions

Lecture 40 Why Model Explainability is important ?

Lecture 41 Model Explainability Functions in BigQuery

Lecture 42 ML.WEIGHTS Function

Lecture 43 List of functions supported by all models

Section 7: Logistic regression

Lecture 44 What is Logistic regression ?

Lecture 45 Sigmoid Function

Lecture 46 Logistic regression Example Use case

Lecture 47 Model Training - Write Create model Query for Logistic regression

Lecture 48 Evaluation metrics Fundamentals explained

Lecture 49 Precision, Recall, Accuracy, F1 score

Lecture 50 Evaluation Functions in BigQuery

Lecture 51 Prediction Function (ML.PREDICT)

Lecture 52 Applications of Logistic regression

Section 8: Feature Pre-processing

Lecture 53 Automatic Feature Pre-processing

Lecture 54 Manual Feature Pre-processing - Part 1

Lecture 55 Manual Feature Pre-processing - Part 2

Lecture 56 FEATURE_INFO Function

Section 9: K-means Clustering

Lecture 57 What is Clustering

Lecture 58 K-means algorithm working

Lecture 59 Advantages & Disadvantages of K-means

Lecture 60 Applications of K-means algorithm

Lecture 61 Options in Create model query

Lecture 62 K-means Example in BigQuery - Create model

Lecture 63 K-means Example in BigQuery - Evaluation

Lecture 64 K-means Example in BigQuery - Prediction

Lecture 65 K-means Example in BigQuery - Anomaly detection

Section 10: Boosted Trees

Lecture 66 What is Boosting and Why it is needed

Lecture 67 Boosted Tree working explained

Lecture 68 Types of Boosting

Lecture 69 Options in Create model query - Part 1

Lecture 70 Options in Create model query - Part 2

Lecture 71 Boosted Tree Example - Use Case Intro & EDA

Lecture 72 Boosted Tree Example - Feature Engineering Part 1

Lecture 73 Boosted Tree Example - Feature Engineering Part 2

Lecture 74 Boosted Tree Example - Create model

Lecture 75 Boosted Tree Example - Hyperparameter Tuning

Lecture 76 Boosted Tree Example - Evaluation

Section 11: Model management Operations in BigQuery

Lecture 77 Introduction

Lecture 78 Operations on Models - Part 1

Lecture 79 Operations on Models - Part 2

Section 12: Deep Neural Network (DNN)

Lecture 80 What is Artificial Neural Network

Lecture 81 Working of Artificial Neural Network

Lecture 82 DNN working explained

Lecture 83 Activation Functions - Sigmoid, TanH

Lecture 84 Activation Functions - RELU

Lecture 85 Which Activation Function to choose?

Lecture 86 Dropout technique to avoid Overfitting

Lecture 87 HIDDEN_UNITS Option

Lecture 88 Optimizer in DNN

Lecture 89 Other Options in DNN

Lecture 90 DNN Example Use Case

Lecture 91 DNN Example Implementation in BigQuery ML

Section 13: Principal Component Analysis (PCA)

Lecture 92 What is dimensionality reduction ?

Lecture 93 Working logic of PCA

Lecture 94 Mathematics behind PCA (4 step procedure)

Lecture 95 BigQuery Model Options in PCA

Lecture 96 Creating PCA Model

Lecture 97 Evaluating and Inferencing PCA model

Section 14: BigQuery ML Pricing

Lecture 98 Free operations in BigQuery ML

Lecture 99 What is Flat rate pricing model

Lecture 100 Costs involved in Flat rate pricing model

Lecture 101 Reservations

Lecture 102 BigQuery ML On-demand pricing model

Lecture 103 Calculate price for Create Model query

Section 15: Matrix Factorization (Collaborative Filtering)

Lecture 104 Introduction to Matrix Factorization (Recommendation systems)

Lecture 105 Content Filtering

Lecture 106 Model based Collaborative Filtering

Lecture 107 Memory based Collaborative Filtering

Lecture 108 Create model Options

Lecture 109 Purchase BigQuery slots for Matrix Factorization

Lecture 110 Create Recommendation Model using Matrix Factorization

Lecture 111 Evaluation and Recommendation

Section 16: ARIMA+ for Time series Forecasting

Lecture 112 What is Time series Forecasting ?

Lecture 113 Components of Time series

Lecture 114 Stationarity in Time series

Lecture 115 Auto regression (AR) in ARIMA

Lecture 116 Moving Average (MA) in ARIMA

Lecture 117 ARIMA+ Options - Part 1

Lecture 118 ARIMA+ Options - Part 2

Lecture 119 ARIMA+ Example - Use case & EDA

Lecture 120 ARIMA+ Example - Create Model

Lecture 121 ARIMA+ Example - Evaluation

Lecture 122 ARIMA+ Example - Inferencing Functions

Lecture 123 ARIMA+ Example - Model Explainability

Section 17: Additional Learnings

Lecture 124 Google Cloud SDK setup

Section 18: BONUS

Lecture 125 Bonus

Machine Learning Engineers,Data analysts,Data scientists,Data Engineers