ALLY Tech AI is a system for processing structured and unstructured data sets. Based on them, the platform allows you to create, train and pre-train artificial intelligence models using the Drag&Drop principle without coding, creating flowcharts in BPMN 2.0 notation. The platform allows you to perform the full cycle of Data Science.
The system can be used to solve the tasks of knowledge base creation, predictive analytics in industry and medicine. For example, you can create a model for classifying moles, predicting forest fires, etc.
RAZUM AI can handle these types of data:
- numerical data (tables);
- analyzing time series, images, videos, texts.
AlLY Tech AI already has pre-designed models of data processing and AI software modules (machine learning and neural networks). Some of the models are already customized and pre-trained.
Solving problems using the platform does not require coding, a specialist just needs to create a flowchart with a sequence of transformations over the initial data, run it and get the result. ALLY Tech AI provides a wide range of possible visualizations of the results: various graphs, charts, tables, and images. This makes it easier to analyze the process and visualize the results.
The container-oriented architecture of the platform allows you to maintain modularity, with functional blocks packaged in appropriate docker containers.
The platform infrastructure supports rapid deployment. It takes one day to unpack and load user data and start training models. The system supports an optimal set of libraries and rules for data preprocessing “out of the box” and various hardware acceleration modules guarantee fast solution of all typical tasks. The ALLY Tech AI platform is user-centric: starting with theclear interface, ending with a large list of possible functions used for data processing. For example, data analysis (preprocessing, preprocessing, various tests, etc.), machine learning (prediction, classification, clustering, etc.), deep learning (classification, regression, object recognition, etc.).