Web Neural is a neural computation framework designed to give developers explicit, observable control over artificial neural network execution. Rather than hiding training and inference behind opaque abstractions, Web Neural exposes the full lifecycle of neural processing - from blocking training loops to structured message passing between execution layers.
This framework is ideal for developers who want: * Deterministic control over neural training and inference * Transparent execution flow without hidden background behavior * Chained request-response patterns for managing complex neural workflows * Clear separation between computation engines and orchestration layers
In this guide you'll learn how to: 1. Initialize and configure neural engines 2. Execute blocking training cycles with progress reporting 3. Chain lambda-style requests and responses accross layers 4. Observe, control, and evolve neural systems with confidence
Whether you're experimenting with neural architectures, building adaptive systems, or integrating machine learning into larger applications, Web Neural provides a disciplined, transparent foundation for neural computation - allowing you to understand, trust, and extend every step of the process.
You can instantiate the framework by including the following in your html source.
<script type = "module" src = "https://www.javaika.com/frame/web-neural/[version]/api.js"></script>
Replace '[version]' with the framework version you intend to use (i.e. 1.0.0).