Take your machine learning projects to production
AI Platform makes it easy for machine learning developers, data scientists, and data engineers to take their ML projects from ideation to production and deployment, quickly and cost-effectively. From data engineering to “no lock-in” flexibility, AI Platform’s integrated tool chain helps you build and run your own machine learning applications.
AI Platform supports Kubeflow, Google’s open-source platform, which lets you build portable ML pipelines that you can run on-premises or on Google Cloud without significant code changes. And you’ll have access to cutting-edge Google AI technology like TensorFlow, TPUs, and TFX tools as you deploy your AI applications to production.
You can use Cloud Storage or BigQuery to store your data. Then use the built-in data labeling service to label your training data by applying classification, object detection, and entity extraction, etc., for images, videos, audio, and text. You can also import the labeled data to AutoML and train a model directly.
Build and run
You can build your ML applications on GCP with a managed Jupyter Notebook service that provides fully configured environments for different ML frameworks using Deep Learning VM Image. Then you can use AI Platform Training and Prediction services to train your models and deploy them to production on GCP in a serverless environment, or do so on-premises using the training and prediction microservices provided by Kubeflow.
You can manage your models, experiments, and end-to-end workflows using the AI Platform interface within the GCP console, or do so on-premises using Kubeflow Pipelines. AI Platform offers advanced tooling to help you understand your model results and explain them to business users.
Machine learning development: the end-to-end cycle
Kubeflow, AI Hub, and notebooks can be used for no charge. You can learn about the pricing of our managed services like AI Platform Training, AI Platform Predictions, Compute Engine, Google Kubernetes Engine, BigQuery, and Cloud Storage here. You can also use our pricing calculator to estimate the costs of running your workloads.
In retail, it’s important to provide customers with easy access to alternative products or recommended add-ons. We train our own machine learning models with TensorFlow on Google Cloud ML, and we automate the periodic retraining of these models with Kubeflow Pipelines. Together with AI Hub, useful for sharing models between data scientists, we can now iterate faster on our models, and automatically deploy them to staging and production.Lucas Ngoo, co-founder, CTO, Carousell
The ability to use Google Cloud Platform to perform image analysis on Cloud ML Engine for epidemiologic breast cancer studies represents a huge step forward. Applying analysis to human pathology is a very new field, and we’re excited about what we’ll find. If we hadn’t taken a machine learning approach, it would have taken us three years instead of three months to analyze over 1,700 tissue samples stored in Cloud Storage, even with a team of dedicated pathologists, and because people bring their own bias to any analysis, we’re also achieving better consistency and quality.Mia M. Gaudet, PhD, Scientific Director of Epidemiology Research, American Cancer Society
Google Cloud Machine Learning Partners come with deep AI expertise and can help you incorporate ML for a wide range of use cases across every stage of model development and serving.
Highlights from Next ’19
Learn and build
New to GCP? Get started with any GCP product for free with a $300 credit.
Need more help?
Our experts will help you build the right solution or find the right partner for your needs.