Documentation

Commands, modules, and tuning

Everything you need to run the engine. Copy-paste the commands, tune the modules, and read the output.

Commands

Run the engine

usage

djehuti2  <datafile>  [options]

  <datafile>              NPY / HDF5 / VCF / FASTA / GenBank / JSON / ...
  --modules <spec>        comma-separated names, 'all', or 0xHEX  (default: all)
  --sims N                Monte Carlo simulation count (>= 1000)
  --module-config <json>  per-module parameter overrides (JSON string)
  --config <file>         run a JSON job file, prints combined JSON result
  --list-modules          list selectable modules and exit
  --help                  show this help and exit

# Examples
  djehuti2 data.npy
  djehuti2 data.npy --modules heston,evt,copula
  djehuti2 data.npy --modules 0x1400
  djehuti2 data.npy --sims 5000000
  djehuti2 data.npy --module-config '{"random_forest":{"n_trees":300}}'
  djehuti2 --config job.json

Modules

Selectable modules

Run --list-modules to see the names on your build. The Monte Carlo core always runs and is not a --modules name.

Name	Domain	What it computes
`correlation`	cross	Pearson / Spearman / Kendall, partial correlation, significance
`bayesian`	cross	Normal-Normal belief update + change-point
`heston`	quant	Heston stochastic-volatility calibration + simulation
`evt`	quant/risk	Extreme Value Theory (POT / GPD) + return levels
`copula`	quant	Gaussian and C-vine copula dependence
`timeseries`	quant/ML	SARIMA + GJR-GARCH volatility
`portfolio`	quant	Markowitz portfolio + VaR / CVaR
`cointegration`	quant	Engle-Granger cointegration / pairs
`hmm`	ML	Gaussian hidden Markov regime detection
`fixedincome`	quant	Nelson-Siegel yield curve
`randomforest`	ML	Random Forest regression / classification
`qmc`	numeric	Quantum Monte Carlo (harmonic oscillator)
`variants`	genomics	VCF variant calling / cohort summary
`popgen`	genomics	Population genetics (Fst, Tajima's D, diversity)

Tuning

`--module-config`

Pass a JSON string with one block per module you want to tune. Unspecified modules use their defaults.

module-config

# Example: tune Random Forest and EVT
$ djehuti2 data.npy --module-config '{"random_forest":{"n_trees":300,"max_depth":12},"evt":{"threshold":0.97}}'

Block	Parameters (default)
`stochastic_vol`	`kappa` (2.0), `theta` (0.04), `xi` (0.3), `rho` (-0.7), `v0` (0.04)
`timeseries`	`arima_p` (1), `arima_d` (1), `arima_q` (1), `forecast_steps` (10)
`evt`	`threshold` (0.95, a percentile in 0..1)
`random_forest`	`n_trees` (100), `max_depth` (10), `min_samples_split` (2)
`copula`	`type` (`gaussian` or `vine`), `family` (`gaussian`/`clayton`/`gumbel`), `n_variables`, `simulations`
`quantum_mc`	`particles` (100), `steps` (10000), `time_step` (0.01), `omega` (1.0)
`hmm`	`max_iter` (50), `tolerance` (1e-4), `n_restarts` (5)
`cointegration`	`significance` (0.05), `lag_order` (1)
`correlation`	`method` (`pearson` or `spearman`)
`bayesian`	`credible_interval` (0.95), `prior_type` (`noninformative`)
`fixed_income`	`dv01_bp_shift` (1)

Recipes

Cross-industry module combos

Pick the modules that match your question. The Monte Carlo core runs automatically.

Industry	`--modules` recipe
Finance / quant	heston,evt,copula,cointegration,portfolio
Insurance	evt,copula,bayesian
Energy / utilities	timeseries,evt,hmm
Telecom	timeseries,randomforest,bayesian,correlation,hmm
Agriculture	timeseries,randomforest,evt,popgen,correlation
Climate / environment	timeseries,evt,hmm,correlation
Healthcare / genomics	popgen,variants,bayesian,randomforest
Manufacturing / quality	evt,randomforest,correlation
Retail / e-commerce	timeseries,randomforest,bayesian
Mining / commodities	timeseries,evt,copula,cointegration
Research / academia	all

Output

What you get

Two streams, both to standard output:

Terminal log: per-module progress and per-formula timings.
JSON block, after a line containing __DJEHUTI_JSON__, machine readable.

JSON output

{
  "engine": "djehuti2 v3.0",
  "file": {
    "bytes_mapped": 1048576,
    "read_ms": 12.3,
    "mmap_evidence": {
      "access": "O_RDONLY / PROT_READ / MAP_PRIVATE",
      "copy_on_read": false,
      "full_decompress": false,
      "disk_writes": 0,
      "swap_friendly": true,
      "corruption_guarded": true
    }
  },
  "modules_run": ["montecarlo", "heston", "evt"],
  "results": {
    "montecarlo": { "mean": 0.00042, "stddev": 0.0111, "var_95": 0.0179 },
    "heston": { "kappa": 2.0, "theta": 0.04, "mean_vol": 0.199 },
    "evt": { "xi": 0.312, "beta": 0.0078, "return_level_100": 0.032 }
  },
  "timings_ms": [
    { "module": "montecarlo", "ms": 234.0 },
    { "module": "heston", "ms": 15.2 },
    { "module": "evt", "ms": 8.7 }
  ],
  "global": { "total_ms": 270.1, "sims_per_second": 213675213 }
}