Skip to content

Lattice

Latest
Latest

Choose a tag to compare

View all tags
@arungeorgesaji arungeorgesaji released this 10 Nov 11:17
· 1 commit to main since this release
v1.0
43e514d

Overview

A Julia-based grid computation framework integrating AI, scientific computing, and data processing in a single system.

See It In Action

Watch the demo video to see a quick demo of some basic features without having to download it for yourself

Watch the Lattice Demo Video

Core Features

  • Grid Abstraction: Universal data structure for multi-dimensional operations
  • Neural Networks: Complete deep learning framework with layers, optimizers, and training utilities
  • Image Processing: Convolution operations, edge detection, and filtering
  • Audio Processing: Waveform generation and feature extraction
  • Text Processing: Encoding methods and similarity metrics
  • Physics Simulation: Fluid dynamics, heat diffusion, particle systems, and cellular automata
  • Computer Graphics: 2D/3D shape generation, ray marching, transformations, and morphological operations
  • File I/O: Binary and text-based grid serialization

Installation

From Julia REPL

using Pkg
Pkg.add(url="https://github.com/arungeorgesaji/lattice")

From Command Line

julia -e 'using Pkg; Pkg.add(url="https://github.com/arungeorgesaji/lattice")'

Running Examples

# Clone the repository if you haven't already
julia examples/example_name.jl

Components

Core

  • Grid — Base data abstraction supporting N-dimensional arrays
  • Arithmetic, transformation, and mapping operations
  • Broadcasting and iteration support
  • Grid creation helpers: zeros_grid, ones_grid, random_grid

I/O

  • save_grid, load_grid — Binary serialization
  • save_grid_text, load_grid_text — Human-readable text serialization

Neural Networks

  • Layers: DenseLayer, ConvLayer, RNNLayer, AttentionLayer, MaxPool, GlobalAvgPool
  • Activations: relu, sigmoid, tanh, softplus
  • Loss Functions: mse_loss, binary_cross_entropy, categorical_cross_entropy, huber_loss
  • Optimizers: SGD (momentum), Adam
  • Utilities: Sequential, get_parameters, count_parameters

Image Operations

  • Blur, edge detection, and custom convolution
  • Supports convolution modes :valid and :same
  • Visualization tools: show_ascii, show_comparison

Audio Processing

  • Sine wave generation, normalization, gain control
  • Feature extraction: RMS energy, zero-crossing rate
  • Compatible with Grid operations and transformations

Text processing

  • Character and one-hot encoding
  • Sliding window generation for sequence data
  • Text-to-grid and grid-to-text conversion
  • Similarity scoring between text grids

Physics Simulations

  • Fluid Dynamics: Density and velocity field simulation
  • Heat Diffusion: Temperature propagation over time
  • Wave Propagation: Basic wave equation solver
  • Particle Systems: Gravity-based particle simulation
  • Cellular Automata: Game of Life and custom rule support

Computer Graphics

  • 2D Shape Generation: Circles, rectangles, lines with customizable parameters
  • 3D Voxel Operations: Sphere and cube generation in 3D space
  • Ray Marching: Signed distance field rendering for 3D shapes
  • Transformations: Rotation and scaling operations
  • Pattern Generation: Checkerboards, gradients, and procedural textures
  • Morphological Operations: Dilation and erosion for shape processing

Visualization

  • ASCII renderers for console-based visualization
  • Comparative visual outputs for transformations
Assets 2
Loading