BigML Certifications
The breadth of intelligent applications the BigML platform can support spawn many new opportunities for BigML partners to get involved in delivering Machine Learning-based solutions. Our certifications are perfect for software developers, system integrators, technology consulting, and strategic consulting firms to rapidly get up to speed with Machine Learning and the BigML platform as they acquire and grow their customer base.
In order to be eligible to enroll into the BigML Certified courses you must show certain level of proficiency in Machine Learning, BigML Dashboard, BigML API, and WhizzML. The following getting started assets will get you up and running in no time: ML 101, Tutorials, API documentation, and WhizzML.
Certified Engineer
This certification course prepares analysts, scientists, and software developers to become BigML Certified Engineers. It consists of 8 online classes of 1.5 hours each. Evaluation will be based on solving a set of theoretical questions and exercises presented during the course. The modules listed below will consist of 2 sessions each to complete the 8 online classes.
CERTIFICATIONS CALENDAR
Modules
Objective
- Understand how to parameterize supervised and unsupervised methods to achieve better performance.
- Learn how to compose multiple methods together to better solve modeling problems.
Pre-requisites
- BigML sources and datasets.
- Supervised (Models, Ensembles, Linear Regressions, Logistic Regressions, Deepnets, Time Series, and OptiML) and Unsupervised (Cluster Analysis, Anomaly Detection, Association Discovery, and Topic Modeling) methods.
- 1-click Model, 1-click Ensemble, 1-click Linear Regression, 1-click Logistic Regression, 1-click Deepnet, 1-click Time Series, 1-click OptiML, 1-click Cluster, 1-click Anomaly, 1-click Association, 1-click Topic Model.
- Simple evaluations and metrics.
Syllabus
-
Modeling vs. Prediction
-
Supervised Learning
Decision Trees: Node threshold, Weights, Statistical Pruning, Modeling Missing Values.
Ensemble Classifiers: Bagging (Sample Rates, Number of Models), Random Decision Forests (Random Candidates), Boosting.
Linear Regression: Field Encodings.
Logistic Regression: L1 Normalization, L2 Normalization, Field Encodings, Scales.
Deepnets: Topologies, Gradient Descent Algorithms, Automatic Network Discovery.
Time Series: Error, Trend, Damped, Seasonality.
Evaluation: How to Properly Evaluate a Predictive Model, Cross-Validation, ROC Spaces and Curves.
OptiML: How to optimize the process for model selection and parametrization to automatically find the best model for a given dataset.
Fusion: Combination of models, ensembles, linear regressions, logistic regressions, and deepnets to balance out the individual weaknesses of single models.
-
Unsupervised Learning
Clustering: Number of Clusters, Dealing with Missing Values, Modeling Clusters, Scaling Fields, Weights, Summary Fields, K-means vs. G-means.
Association Discovery: Measures (Support, Confidence, Leverage, Significance Level, Lift), Search Strategies (Confidence, Coverage, Leverage, Lift, Support), Missing Items, Discretization.
Topic Modeling: Topics, Terms, Text analysis.
Anomaly Detection: Forest Size, Constraints, ID Fields.
-
Combination and Automation
Stacking.
Timing
-
The lecturer will be available between 08:00 AM and 08:00 PM PT. Please send an email to education@bigml.com for other time ranges.
Objective
- Proficiency in using BigML's API and client-side tools to create ML resources.
- Integration and automation of the workflows needed put a ML solution in production.
Pre-requisites
- Basic knowledge of BigML and its resources (UI-level familiarity is enough).
- Basic programming skills (some examples are in Python, so knowledge of the language will be a plus).
- Familiarity with REST APIs.
Syllabus
-
API description
Domains (bigml.io vs. Private Deployments).
Authentication.
Inputs and outputs.
Resources: Common information, Specifics, Listing and filtering.
-
First level wrappers
Bindings.
Methods mapping.
Field management.
Local resources.
-
Second level wrappers
BigMLer.
Resource management.
Field management.
Workflow automation.
Automated feature engineering.
-
Modeling strategies
-
Predicting strategies
Timing
-
The lecturer will be available between 01:00 AM and 01:00 PM PT. Please send an email to education@bigml.com for other time ranges.
Objective
- Data is typically: scattered, unclean, and imperfect. How to make it ML-Ready.
- Once data is ML-Ready, why/how to make better features.
- Not all features are good. How to choose and what to watch out for.
Pre-requisites
- Advanced Modeling Class.
- Familiarity with: SQL, Python / Pandas, CSV formatting.
Syllabus
-
ML-Ready Data
What is it?
Formats.
Structures for ML tasks.
Automating Labeling.
-
Data Transformations
Cleansing Missing Data, Cleaning Data, Better Data.
Transformations outside and inside BigML: SQL-style queries, Denormalizing, Aggregating, Pivoting, Time windows, Updates, Streaming Data, Images.
Principal Component Analysis (PCA): Dataset transformation and dimensionality reduction.
-
Feature Engineering
Auto Transformations: Date-time parsing, LR/cluster missing, LR/cluster auto-scaling, Bag-of-words (Language, Tokenization, etc).
Manual - Flatline: DSL for feature engineering, Basics (s-expressions/formulas, Literals, Counters, Field Values / Properties, Strings, Regex, Operators), Limitations.
Numerics: Discretization, Normalization, Z-score, Built-in math functions, Type-casting, Random, Shocks, Moving averages.
Date-times: UI timestamp, Epoch, Moon phase.
Text: JSON key/val, Topic distributions.
-
Feature Selection
Correlations.
Leakage.
Field Importance (ensembles).
Advanced Selection: Best-First, Boruta.
Timing
-
The lecturer will be available between 10:00 AM and 10:00 PM PT. Please send an email to education@bigml.com for other time ranges.
Objective
- Proficiency in using BigML's DSL language, WhizzML, as a server-side tool to automate ML-workflows in a scalable, replicable and shareable way.
Pre-requisites
- Basic knowledge of BigML and its resources (UI-level familiarity is enough).
- Familiarity with ML-workflows.
- Basic programming skills (knowledge of some language of the LISP-family and/or WhizzML will be a plus).
Syllabus
-
WhizzML directives
-
Directives mappings
-
Simple workflows in WhizzML
Batch Anomaly Score.
Evaluation.
Clustered dataset generation.
-
Advanced workflows in WhizzML
Cross-validation.
Covariate shift.
Stacked generalization.
Timing
-
The lecturer will be available between 03:00 PM and 09:00 PM PT. Please send an email to education@bigml.com for other time ranges.
Certifications calendar
Certified Architect
This certification course prepares BigML Certified Engineers to become BigML Certified Architects. Once you have successfully become a BigML Certified Engineer, you are eligible to enroll into the BigML Certified Architect course. The certification process consists of 8 online classes of 1.5 hours each. Evaluation will be based on solving a set of theoretical questions and exercises presented during the course.
Modules
Objective
- Get ready to design and build robust Machine Learning-based applications that operate in real-world environments.
Pre-requisites
- BigML Certified Engineer.
Syllabus
-
Real-world Machine Learning
-
Building end-to-end Machine Learning applications
-
How to size and address your project
Premature optimization is the root of all evil in Machine Learning as well.
Automating the automatable.
Syllabus
-
How to generate thousands of predictions per second
-
How to store predictions for further analyses
-
How to implement robust predictions
Syllabus
-
Local models vs. remote models
-
How to use and operate models
-
How to monitor your models
Syllabus
-
Automated model and parameter selection
-
When good is "good enough"
-
What your actual test set tells you about your model
Syllabus
-
Tracking models over time. You can learn from everything
-
Automating covariate shift detection
-
Active Learning
Syllabus
-
Diversity vs. volume
-
Detecting biases
-
Detecting blind spots
Syllabus
-
Choice of data engineering tools
-
Automating feature selection
-
Automating feature generation
Syllabus
-
Anatomy of a robust Machine Learning application
-
Lessons learned and best practices
-
Design patterns: beyond lessons learned and best practices
