next generation Machine intelligence Ecosystem
SliceMatrix-IO is an end-to-end (e2e) computational eco-system which enables users to build machine intelligence systems with speed, agility and scalability on demand. Furthermore, users can build end-to-end machine intelligent solutions with minimal code.
start machine learning immediately
Watch short videos on how to build Intelligent Trading Models
a machine learning ecosystem
SliceMatrix-IO is a computational software platform that delivers powerful machine learning algorithms. SliceMatrix is extensible and can run on any platform such as a drone, Android device, iOS device, cluster of servers, etc.
Furthermore SliceMatri-IO delivers:
- Real-Time Persistence, Build Once & Repurpose.
- Once a model or pipeline is built, it is stored in the cloud, meaning it can be re-used multiple times by different processes in parallel. This saves the end-user valuable time and effort when building models & analyzing multiple datasets across many projects.
- Shrink API Deployment Time
- Move seamlessly from prototyping to production. Scale up or down on demand, based on business needs and requirements.
- SliceMatrix-IO enables applications with the right permission level to securely access your models in real-time across the firm.
build an anomaly detection model with a few lines of python code
process multiple datasets simultaneously
build powerful machine intelligent applications with SliceMatrix-IO models
scale with cost effective pricing
SliceMatrix-IO(beta) product and pricing options, offer the flexibility to effectively manage your costs and still scale as your business requires. With SliceMatrix-IO(beta), you can easily right size your services.
Real-World price Calculation + Anomaly Detector + trading model
Suppose you want to use SliceMatrix-IO's suite of anomaly detection algorithms to monitor a list of stocks in real time. How many stock symbols could you follow?Let's assume you have the Neo Matrix usage plan, i.e. 5,000 calls a month.
You want your program to calculate an anomaly score for each symbol every hour during the trading day, let's say 8am to 5pm to pick up pre-market and after-hours activity!
That means we're making 1 call to the anomaly detectors "score" method every hour for 9 hours a day, i.e. 9 calls per stock per day.
Assuming there are 20 trading days in a month that adds up to 180 calls per stock.
5k / 180 = ~ 27.78
That means you could monitor approximately 27 stock symbols in realtime for only $25 / month. Furthermore, you don't have to set up any additional infrastructure to train and run machine learning models in real-time!