use clap::builder::styling::{AnsiColor, Effects, Styles};
use clap::{Parser, Subcommand};
fn styles() -> Styles {
clap::builder::styling::Styles::styled()
.header(AnsiColor::Red.on_default() | Effects::BOLD)
.usage(AnsiColor::Red.on_default() | Effects::BOLD)
.literal(AnsiColor::Blue.on_default() | Effects::BOLD)
.error(AnsiColor::Red.on_default() | Effects::BOLD)
.placeholder(AnsiColor::Green.on_default())
}
#[derive(Parser, Debug)]
#[clap(
author,
version,
name = "lmm",
propagate_version = true,
styles = styles(),
help_template = r#"{before-help}{name} {version}
{about}
{usage-heading} {usage}
{all-args}{after-help}
AUTHORS:
{author}
"#,
about = r#"
▄▄▄ ▄▄▄ ▄▄▄ ▄▄▄ ▄▄▄
▀██▀ ███▄ ▄███ ███▄ ▄███
██ ██ ▀█▀ ██ ██ ▀█▀ ██
██ ██ ██ ██ ██
██ ██ ██ ██ ██
████████ ▀██▀ ▀██▄ ▀██▀ ▀██▄
Large Mathematical Model · Equation-Based Intelligence
The `lmm` CLI enables interaction with the Large Mathematical Model (LMM).
It provides advanced equation discovery, physics simulation, causal
inference, and unified sequence processing features.
"#
)]
pub struct Cli {
#[command(subcommand)]
pub command: Commands,
}
#[derive(Subcommand, Debug)]
pub enum Commands {
#[command(about = "Simulate continuous logical pathways")]
Simulate {
#[arg(short, long, default_value_t = 0.01)]
step: f64,
#[arg(short = 't', long, default_value_t = 100)]
steps: usize,
},
#[command(about = "Discover governing equations from data")]
Discover {
#[arg(short, long, default_value = "synthetic")]
data_path: String,
#[arg(short = 'i', long, default_value_t = 100)]
iterations: usize,
},
#[command(about = "Evaluate conscious state coherence")]
Consciousness {
#[arg(short, long, default_value_t = 3)]
lookahead: usize,
},
#[command(about = "Run harmonic and chaotic physical models")]
Physics {
#[arg(short, long, default_value = "harmonic")]
model: String,
#[arg(short, long, default_value_t = 200)]
steps: usize,
#[arg(short = 'z', long, default_value_t = 0.01)]
step_size: f64,
},
#[command(about = "Perform causal interventions and counterfactuals")]
Causal {
#[arg(short = 'n', long, default_value = "x")]
intervene_node: String,
#[arg(short = 'v', long, default_value_t = 1.0)]
intervene_value: f64,
},
#[command(about = "Compute tensor field gradients and divergences")]
Field {
#[arg(short, long, default_value_t = 10)]
size: usize,
#[arg(short = 'o', long, default_value = "gradient")]
operation: String,
},
#[command(about = "Encode continuous truth into discrete text")]
Encode {
#[arg(short, long, default_value = "-")]
input: String,
#[arg(short, long, default_value = "Hello, LMM!")]
text: String,
#[arg(long, default_value_t = 80)]
iterations: usize,
#[arg(long, default_value_t = 4)]
depth: usize,
},
#[command(about = "Decode text back into dynamic equations")]
Decode {
#[arg(short, long)]
equation: String,
#[arg(short, long)]
length: usize,
#[arg(short, long, default_value = "", allow_hyphen_values = true)]
residuals: String,
},
#[command(about = "Predict next sequence based on pattern logic")]
Predict {
#[arg(short, long, default_value = "-")]
input: String,
#[arg(short, long, default_value = "The Pharaohs encoded reality in")]
text: String,
#[arg(short = 'w', long, default_value_t = 32)]
window: usize,
#[arg(short = 'p', long, default_value_t = 16)]
predict_length: usize,
#[arg(long, default_value_t = 80)]
iterations: usize,
#[arg(long, default_value_t = 4)]
depth: usize,
#[arg(long)]
dictionary: Option<String>,
},
#[command(about = "Extract key meaning via GP scoring")]
Summarize {
#[arg(short, long, default_value = "-")]
input: String,
#[arg(
short,
long,
default_value = "Equations are the universe. Mathematical models compress reality into compact symbolic forms. Simulation is more powerful than description. The world is not made of words but of structure and force."
)]
text: String,
#[arg(short = 'n', long, default_value_t = 2)]
sentences: usize,
#[arg(long, default_value_t = 40)]
iterations: usize,
#[arg(long, default_value_t = 3)]
depth: usize,
},
#[command(about = "Generate a single structural sentence")]
Sentence {
#[arg(short, long, default_value = "-")]
input: String,
#[arg(short, long, default_value = "Mathematical equations")]
text: String,
#[arg(long, default_value_t = 60)]
iterations: usize,
#[arg(long, default_value_t = 3)]
depth: usize,
},
#[command(about = "Generate a cohesive paragraph from a seed")]
Paragraph {
#[arg(short, long, default_value = "-")]
input: String,
#[arg(
short,
long,
default_value = "Equations encode the structure of reality"
)]
text: String,
#[arg(short = 'n', long, default_value_t = 3)]
sentences: usize,
#[arg(long, default_value_t = 60)]
iterations: usize,
#[arg(long, default_value_t = 3)]
depth: usize,
},
#[command(about = "Structure a full essay with intro and conclusion")]
Essay {
#[arg(short, long, default_value = "-")]
input: String,
#[arg(
short,
long,
default_value = "Mathematical models and the structure of reality"
)]
text: String,
#[arg(short = 'n', long, default_value_t = 2)]
paragraphs: usize,
#[arg(short = 's', long, default_value_t = 3)]
sentences: usize,
#[arg(long, default_value_t = 60)]
iterations: usize,
#[arg(long, default_value_t = 3)]
depth: usize,
},
#[cfg(feature = "net")]
#[command(about = "Ask a question and get an equation-scored answer from the web")]
Ask {
#[arg(short, long)]
prompt: String,
#[arg(short, long, default_value_t = 5)]
limit: usize,
#[arg(short = 'n', long, default_value_t = 3)]
sentences: usize,
#[arg(long, default_value = "wt-wt")]
region: String,
#[arg(long, default_value_t = 40)]
iterations: usize,
#[arg(long, default_value_t = 3)]
depth: usize,
},
}