AutoML - bobbae/gcp GitHub Wiki
AutoML is a general term describing automated AI driven methods to enhance ML application and processes. AutoML is also name used by multiple projects and companies in different contexts and cloud based services.
Google Cloud AutoML helps you build custom machine learning models with limited machine learning expertise needed.
https://cloud.google.com/automl
AI Platform with AutoML
AI Platform brings AutoML and AI Platform (Classic) together into a unified API, client library, and user interface.
https://cloud.google.com/vertex-ai/docs
AutoML Vision
Derive insights from your images in the cloud or at the edge with AutoML Vision or use pre-trained Vision API models to detect emotion, understand text, and more.
https://cloud.google.com/vision/automl/docs
AutoML Vision enables you to train machine learning models to classify your images according to your own defined labels.
AutoML Vision Edge allows you to export your custom trained models.
https://cloud.google.com/vision/automl/docs#videos
AutoML Video intelligence
Video Intelligence includes several options that you can use to integrate machine learning video intelligence models into your applications.
AutoML Natural Language
AutoML Natural Language enables you to build and deploy custom machine learning models that analyze documents, categorizing them, identify entities within them, or assessing attitudes within them.
AutoML Translation
AutoML Translation lets you to create your own translation models so that translation queries return results specific to your domain.
Cloud Translation - Advanced is part of the Cloud Translation service that provides support for glossaries, batch requests, and AutoML models.
AutoML Tables
AutoML Tables helps to build and deploy machine learning models on structured data at speed and scale.
AutoML Tables help you create clean, effective training data by providing information about missing data, correlation, cardinality, and distribution for each of your features.
How-to guides
https://cloud.google.com/automl-tables/docs/how-to
AutoML Tables Beginner's Guide
https://cloud.google.com/automl-tables/docs/beginners-guide
Evaluating and Deploying AutoML Tables model
https://cloud.google.com/automl-tables/docs/beginners-guide#deploy_your_model_and_make_predictions
AutoML Tables samples
https://cloud.google.com/automl-tables/docs/samples
AutoML Tables Quickstart
https://cloud.google.com/automl-tables/docs/quickstart
AutoML Tables hyperparameters
https://cloud.google.com/automl-tables/docs/logging#hps
AutoML Feature Engineering
When you kick off training, AutoML Tables automatically performs common feature engineering tasks for you, including:
- Normalize and bucketize numeric features.
- Create one-hot encoding and embeddings for categorical features.
- Perform basic processing for text features.
- Extract date- and time-related features from Timestamp columns.
Data Prepping
https://developers.google.com/machine-learning/data-prep
Model training
When you kick off training for your model, AutoML Tables takes your dataset and starts training for multiple model architectures at the same time. This approach enables AutoML Tables to determine the best model architecture for your data quickly, without having to serially iterate over the many possible model architectures.
Using AutoML Tables from BigQuery
You can use AutoML Tables directly from BigQuery SQL.
https://towardsdatascience.com/the-best-of-both-worlds-calling-auto-ml-from-bigquery-9dfd433a45d6
Choosing between AutoML Tables and BigQueryML
You might want to use BigQuery ML if you are more focused on rapid experimentation or iteration with what data to include in the model and want to use simpler model types for this purpose, such as logistic regression.
You might want to work directly in the AutoML Tables interface if you have already finalized the data, and you are optimizing for maximizing model quality.
Transparency and Logging
You can view the structure of your AutoML Tables model using Cloud Logging.
Using Cloud Logging to view AutoML Tables model
Using Cloud Logging, you can see the final model hyperparameters as well as the hyperparameters and object values used during model training and tuning.
AutoML Tables model logs have, among other fields:
- automl_tables as the value of the log_type field
- jsonPayload that contains the specific details of the log entry
- timestamp that describes when the model was created
Payload contents
The contents of a log entry are provided in JSON object format and are stored in the jsonPayload field.
- modelStructure
- trainingObjectivePoint
List of Hyperparameters
The hyperparameter data provided in the logs differ for each type of model.
https://cloud.google.com/automl-tables/docs/logging#hps
Working with long-running operations
When you make an API call that takes a long time to complete, the initial call returns right away, even though the operation is still running.
AutoML Tables sends project owners an email when long-running operations are complete.
Explainability
https://cloud.google.com/automl-tables/docs/features#ai-explanations
Test data export
https://cloud.google.com/automl-tables/docs/features#export
When to use AutoML vs others
Use BigQuery ML for easy, low-cost machine learning and quick viable ML problem using BQML, use Auto ML for code-free, state-of-the-art models.
Measuring and Improving Speech-to-Text Accuracy
Pycaret
https://towardsdatascience.com/build-pycaret-deploy-gcp-521415a6c330
Examples
Classify images of clouds using AutoML
https://www.cloudskillsboost.google/focuses/8406?parent=catalog
Aircraft classification using AutoML
https://www.lucidchart.com/blog/cloud/building-an-image-classification-system-part-2
AutoML Tables notebooks
https://cloud.google.com/automl-tables/docs/notebooks
Music recommendations
Handwritten recipes using AutoML and Vision API
Deploy AutoML model in Cloud Run
Links
- AutoML Cloud Natural language Classification Tutorial
- AutoML Cloud Natural Language Samples
- AutoML Tables Quickstart
- AutoML Tables notebooks
- AutoML Cloud Translation API tutorial
- AutoML Cloud Video Intelligence API Quickstarts
- AutoML Vision
- Cloud Vision API
- Cloud Natural language
- Cloud Translation
- Cloud Text-to-Speech
- Cloud Speech-to-Text
- Dialogflow
- Cloud Inference API