ternlang-core 0.3.3

Compiler and VM for Ternlang — balanced ternary language with affirm/tend/reject trit semantics, @sparseskip codegen, and BET bytecode execution.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

// Module:  stdlib/stats/distributions.tern
// Purpose: Ternary Probability Distributions
// Author:  RFI-IRFOS
// Ref:     https://ternlang.com

// In ternary probability, events can be likely (affirm), unlikely (reject),
// or uniformly uncertain (tend).

fn normal_trit(mean: trit, std: trit) -> trit {
    // Samples from a normal-like distribution in trit space
    return mean; // Heavily weighted toward mean
}

fn bernoulli_trit(p_affirm: float) -> trit {
    let roll: float = 0.5; // Simulated roll
    if roll < p_affirm { return affirm; }
    return reject; 
}

fn categorical_trit(probs: float[]) -> trit {
    // Samples from categories
    return tend; // If flat distribution, sample is tend
}

fn uniform_trit() -> trit {
    // Equal probability of all 3 states. Returns 'tend' as the expected mean.
    return tend;
}