A System so Advanced, the Military uses it to Guide Drones.

Extreme Trading Hardware

The Modulus NNPC-12 card was originally developed for the military to guide Unmanned Aerial Vehicles (UAVs). Then, we acquired the technology and adapted it to finance.

The Result? An A.I. Board which can "Blow Up" your Trading Counterparties.

Modulus is proud to bring you the new NNPC-12 neural network PCI card for financial applications. The NNPC-12 provides true MIMD processing for the fastest possible network training, evaluation, and implementation.

The heart of the NNP family is the MIMD Neural Network Processor, which provides the basic unit of processing. Multiple NNPs can operate in parallel through a unique bus and connector design.

PCI, ISA, and VME busses are supported with interface boards that have up to two built-in NNPs. Additional NNPs can be stacked on the baseboard. The VME board includes two TMS320C40 DSP Chips.

The boards are supported by all PCs and servers.

NNPC-12 Card

Features

  • True Multiple Instruction-Multiple Data (MIMD) architecture.
  • Stack to a Max of 10 Processors.
  • 8,192 Neurons per system.
  • 32,768 Connections per processor.
  • 32 million multiply-accumulate cycles per second, per processor.
  • Unique Neural Network Processor Language; neural network tools and compiler provided.
  • Program and run standard paradigms or your own unique paradigm.
  • Powerful architecture concept that avoids memory bottlenecks.

System Requirements

  • PC Platforms running Windows:
    • PC with ISA Bus
    • PC with PCI Bus
  • Silicon Graphics Platforms running IRIX 5.2 and above:
    • Indy using Bit3 Corporation GIO32 to VME bus adaptor
    • Indigo using Bit3 Corporation GIO32 to VME bus adaptor
    • Indigo 2 using internal EISA bus
    • Onyx using internal VME bus
    • Power Series with single IO3 board using internal VME bus
  • VME
    • VME 6U dual TMS320C40 DSP card with support for four Neural Network Processors

Training Techniques

  • Backpropagation
  • Resilient Propagation (RProp)
  • Scaled Conjugate Gradient (SCG)
  • Levenberg–Marquardt algorithm
  • Manhattan Update Rule Propagation
  • Competitive Learning
  • Hopfield Learning
  • Genetic Algorithm Training
  • Instar Training
  • Outstar Training
  • ADALINE Training

Neural Network Architectures

  • ADALINE Neural Network
  • Adaptive Resonance Theory 1 (ART1)
  • Bidirectional Associative Memory (BAM)
  • Boltzmann Machine
  • Counterpropagation Neural Network (CPN)
  • Elman Recurrent Neural Network
  • Neuroevolution of augmenting topologies (NEAT)
  • Feedforward Neural Network (Perceptron)
  • Hopfield Neural Network
  • Jordan Recurrent Neural Network
  • Radial Basis Function Network
  • Recurrent Self Organizing Map (RSOM)
  • Self-Organizing Map (Kohonen SOM)

Get Started with NNPC-12 >

Software Toolkit

Available programming tools include an assembler, linker, debugger, and diagnostics program which all allow developers to quickly design and use networks with NNPC-12.

These programming tools allow rapid prototyping of neural network solutions and integration in real-world applications. The programming tools support a variety of paradigms trained with supervised and/or unsupervised training algorithms.

NNPC-12 is available to US-based residents and companies only.

Why Choose Modulus?

Modulus is a financial technology company. While that may not sound like a real differentiator, it is. It means that our solutions come from our years of experience in the financial technology industry. Our products and services are provided by developers and engineers who have first-hand trading experience. Everyone here at Modulus speaks your language.

Read more about us.