TIM Studio is an intuitive web interface to Tangent Works' end-to-end platform for time-series projects. It brings the power of Machine Learning to the fingertips of even the least technical users, without compromising on speed, accuracy or configurability.
TIM InstantML is a Predictive analytics service for time series data, enabling anyone – not just a data scientist – to create value from business data.
You’ll be able to predict the future and detect anomalies smoother than ever.
InstantML is extremely fast, hyper automated, highly accurate and, most importantly, creates human readable models.
So no more headache of dealing with the complexity of Time Series forecasting, but instead direct impact on your results!
Get started with TIM InstantML in seconds: combine optimal features, create models and bring your data to life!
Use the TIM Studio tool to manage your machine learning projects. Thanks to TIM’s shared repository and architecture, TIM Studio is now elevated from another interface to consume TIM’s analytics capabilities, to a platform bringing all of your work together in a structured and intuitive way.
TIM Studio is built on three pillars:
A user can start working through TIM Studio with precanned solutions, leaning on demos and use case templates that are tailored to the industry of their interest. When confident with the way of working, the user can continue on to experimentation.
When trying out unfamiliar technology, or searching for inspiration for new business challenges to overcome, a precanned example can go a long way in making you succesful. We recognize this, and work to provide a variety of real-world use case templates that range across industry, across business area and across Machine Learning challenge. These templates can be accessed and explored from right within TIM Studio.
Additionally, this list of templates is continuously updated and expanded upon. In our Use Case Library, you can find an up-to-date overview of the templates that are provided.
No complexity of traditional auto ML model selection. No time-consuming data science. No manual feature engineering.
To support every user on their journey to incorporate Machine Learning in their day-to-day business, TIM Studio strives to offer a logical and intuitive process all the way from first dataset upload to interpretation of final results. This process can be fairly direct, or involve many intermediate steps, depending on the use case and the goals of the person using TIM Studio. Regardless of your desired use, you will find TIM Studio to be intuitive in helping to accomplish what you set out to do. Nevertheless, you can find the answer to any question that comes up in this documentation.
Experimentation can happen directly in TIM Studio, as well as across a range of data platforms that TIM integrates with. All experiments – regardless of where they were performed – will eventually be available from within TIM Studio too.
The integration with other platforms has become even stronger with TIM Studio v2.1 and TIM v5.0; the shared repository will make it possible to consult any analytics done with TIM in TIM Studio, regardless of the original platform used.
For more information about the supported platforms, check out the section on TIM on platforms.
When satisfied with an experiment, the user can move on to production, using the results for decision making and monitoring model behavior and model performance over time.
In productions, the goal is for the user to start using the predictions (forecasts, detections) that come out of the models and allow them to help in the decision making process, thereby setting themselves up to reap the rewards of embracing analytics.
Once a user is scaling up these efforts, it will be vital to keep track of the models that are running in production and how their performance evolves. Here MLOps comes into play, and monitoring of the models over time. TIM Studio is built to help you in this process and give you an intuitive overview of your use cases, experiments and jobs, including the evolution of metrics over iterations.
The TIM Repository organizes all your Machine Learning projects. ML experiments are grouped in workspace and user groups. It is easy to keep track of all versions of your Data Sets.
Define different Use cases and Experiments and set the business objective you have.
Create models for forecasting in seconds and see the impact of predictor candidates on your model and the predictor target.
Perform Anomaly detection. Model the normal behaviour of process and detect anomalies on the fly.
Get inspired by the TIM Use Case Library providing you example data sets and experiments in various industries to explore the usability and results for your business.
Use TIM INSTANTML on your favourite
Data Integration Platform,
Analytics & BI Platform,
or Data Science and Machine Learning Platform.
Worried about deploying TIM Instant ML?
TIM is a containerised solution that can be used as a SAAS Service in seconds, in your own cloud, on premise and on-the-edge
Get more value out of your data as of today!