NNTrack Neural Network Modeling and Training Software

About NNTrack

NNTrack is a visual, block-based environment for modeling neural network architectures, training networks, and exporting them for use on hardware modules.

The environment is designed for creating and training neural network models without programming experience. It also supports working with pre-trained neural networks that recognize video streams and Russian speech, as well as processing big data. The software helps users study how neural networks work, implement popular architectures such as VGG19, ResNet, and YOLO, adjust hyperparameters, and visualize training progress with charts and metrics.

The software supports integration with external devices, including Arduino-compatible controllers such as Trackduino, mobile and stationary robots, visual programming environments such as TrackyCode, drones, and game engines. NNTrack can be used to create projects ranging from smart games and robots to systems for recognizing faces, voices, and emotions, as well as analyzing big data.

Integration with unmanned aerial vehicles (UAVs) lets users explore neural network applications such as detecting fires and smoke, searching for people, monitoring infrastructure (power lines, roads, and pipelines), and monitoring crop conditions. The software is intended for school students, university students, engineers, and people beginning careers in artificial intelligence.

NNTrack includes 13 ready-made models for studying neural networks. They can be used to control robotic models built around a Trackduino or Arduino-compatible controller.

NNTrack and its teaching and learning materials are intended to support the advanced-level Informatics curriculum, section 1, “Machine Learning and Artificial Intelligence.” The source article describes this alignment with reference to a draft order of the Russian Ministry of Education on the federal state educational standard for upper secondary education. The listed learning outcomes include understanding the role of machine learning in solving AI problems; learning about data analysis methods (classification, clustering, and regression); and developing practical skills in collecting, cleaning, and assessing data quality, selecting and building models, transforming and visualizing data, and interpreting results.

The environment lets users select a model type based on the task and data characteristics, study how intelligent systems work, build them in practice, and use existing libraries and modern algorithms, including those for implementing neural networks.

The software is licensed indefinitely for one workstation (PC). The package includes technical support, teaching materials, and updates at no additional charge.

The package also includes a digital teaching and learning course on Python programming and a collection of exercises with solutions. An electronic course introducing C++ programming is also provided after a network cooperation agreement is signed, along with online teacher training.

Watch the NNTrack software video

Computer system requirements

Minimum system requirements
ProcessorArchitecture: x64 with AVX support
Clock speed: ≥ 3 GHz
RAM≥ 4 GB
Operating systemMicrosoft Windows 10
Recommended system requirements
ProcessorArchitecture: x64 with AVX support
Clock speed: ≥ 4 GHz
RAM≥ 8 GB
Operating systemMicrosoft Windows 11

NNTrack interface

Main window

The main window is the starting point of the application.

The image shows: 1 — main menu; 2 — link to the main window; 3 — link to the modeling window; 4 — link to the training window; 5 — link to the export window; 6 — link to settings; 7 — create a new project; 8 — open an existing project; 9 — status bar; 10 — buttons for opening recently used projects.

The top of the window contains the quick-access menu. A navigation panel on the left switches between the main window, modeling, training, export, and settings. Hover over an item to see a tooltip with the window name; the active window is highlighted with a white background. The status bar at the bottom displays information and error messages. The main area contains buttons for creating and opening a project.

Download