FAIF - Fast Artificial Intelligence Framework

The C++ header-only artificial intelligence and bioinformatics algorithms library. The C++ template library for bioinformatics and artificial intelligence. Basic abstractions (value nomian, domain, point, space, state, etc.), DNA primitives (Nucleotide, Chain, EnergyNucleo, SecStruct, FoldedChain, FoldedPair, Codon, CodonAminoTable, Nussinov algorithm), Naive Bayesian Classifier, Decision Tree Classifier, K Nearest Neighbour Classifier, cross - validator, population genetics primitives (Allele, Locus, Loci), Timeseries primitives (TimeSeriesDigit, TimeSeriesReal, linear resampling, autoregressive predictor), search in tree like structures (depth-first, breadth-first, unified cost, A*), Evolutionary Algorithm, EM (Expectaiton-Maximization) algorithm, Hill Climbing algorithm, Random generators, Gaussian eliminator. Serialization to text or XML based on boost::serialization.

Attention! The active object design pattern (classes Command, Scheduler, Queue) are moved to the new project mt4cpp

Description

manual, algorithms description (pdf)

Quick start - look also at examples and tests.

• DNA: (quick start)
• timeseries:
  • discretizer (quick start )
• learning:
  • naive Bayesian classifier (quick start , load/save internal state)
  • decision tree classifier (quick start , load/save internal state)
  • random forest
  • K Nearest Neighbors
  • Support Vector Machine (SVM) using sequential minimal optimization
• search
  • uninformed search in tree like structures: searchBreadthFirst, searchDepthFirst (quick start)
  • informed search: searchUnifiedCost (quick start)
  • heuristic search: searchAStar (quick start)
  • Evolutionary Algorithm (quick start)
  • Hill Climbing Algirhthm (quick start)
• utils
  • Random - wrapper on boost::random (quick start)