$$ % Define your custom commands here \newcommand{\bmat}[1]{\begin{bmatrix}#1\end{bmatrix}} \newcommand{\E}{\mathbb{E}} \newcommand{\P}{\mathbb{P}} \newcommand{\S}{\mathbb{S}} \newcommand{\R}{\mathbb{R}} \newcommand{\S}{\mathbb{S}} \newcommand{\norm}[2]{\|{#1}\|_{{}_{#2}}} \newcommand{\pd}[2]{\frac{\partial #1}{\partial #2}} \newcommand{\pdd}[2]{\frac{\partial^2 #1}{\partial #2^2}} \newcommand{\vectornorm}[1]{\left|\left|#1\right|\right|} \newcommand{\abs}[1]{\left|{#1}\right|} \newcommand{\mbf}[1]{\mathbf{#1}} \newcommand{\mc}[1]{\mathcal{#1}} \newcommand{\bm}[1]{\boldsymbol{#1}} \newcommand{\nicefrac}[2]{{}^{#1}\!/_{\!#2}} \newcommand{\argmin}{\operatorname*{arg\,min}} \newcommand{\argmax}{\operatorname*{arg\,max}} \newcommand{\dd}{\operatorname{d}\!} $$

config

config

Subject metadata, goals, and safety limits.

Loaded from a TOML file (see data/templates/subject_template.toml). Nothing here is estimated from data – these are the inputs you supply. Their only roles are to center the priors on the metabolic-rate parameters and to bound what the controller is permitted to prescribe.

Classes

Name	Description
Goal	Where you want to end up, for the controller’s deadline mode.
Maintain	A weight to hold, for the controller’s maintain mode.
Preferences	Diet-style choices with no principled default – only your own.
Safety	Hard limits the controller may not violate.
Subject	Anthropometrics used to build the resting-metabolic-rate prior.

Goal

config.Goal(target_weight_lb, target_date)

Where you want to end up, for the controller’s deadline mode.

Maintain

config.Maintain(weight_lb)

A weight to hold, for the controller’s maintain mode.

Once a goal is reached the question changes from “how fast” to “how much keeps me here”. With this section in the config, maintain mode is the default; it tracks this weight with no deadline, easing back toward it over the planning horizon when the trend drifts off.

Preferences

config.Preferences(target_carb_fraction=None, target_fat_fraction=None)

Diet-style choices with no principled default – only your own.

Unlike Safety, none of these are guardrails; they are taste.

Attributes

Name	Description
target_carb_fraction	Carbohydrate’s share of energy to aim for, e.g. `0.05-0.10` for a
target_fat_fraction	Fat’s share of energy to aim for. Only used as a real preference if

Safety

config.Safety(
    min_kcal=MIN_KCAL_FLOOR,
    max_loss_frac_per_week=MAX_LOSS_FRAC_PER_WEEK,
    min_protein_g_per_kg=MIN_PROTEIN_G_PER_KG,
    max_kcal=6000.0,
)

Hard limits the controller may not violate.

These are your numbers, read from the config file rather than imposed. The package defaults are conservative; if you have set them otherwise, the controller respects your setting.

Subject

config.Subject(
    sex,
    age_years,
    height_cm,
    body_fat_frac=None,
    habitual_carb_g=None,
    habitual_sodium_mg=None,
)

Anthropometrics used to build the resting-metabolic-rate prior.

Methods

Name	Description
expenditure_prior	Prior center for the (intercept, mass-slope) of total expenditure.
habitual_intake	Carbohydrate (g/day) and sodium (mg/day) eaten before a diet begins.
rmr_mifflin	Mifflin-St Jeor resting metabolic rate, kcal/day.

expenditure_prior

config.Subject.expenditure_prior(weight_kg, pal=DEFAULT_PAL)

Prior center for the (intercept, mass-slope) of total expenditure.

Returns (c0, c1) such that non-adaptive, non-TEF expenditure is approximately c0 + c1 * weight_kg in kcal/day.

The decomposition is obtained by scaling Mifflin-St Jeor by the activity level and splitting into the mass-dependent term and everything else. The two coefficients are strongly correlated given the narrow range of body mass in any single dataset; see the identifiability discussion in the report before reading much into either one alone.

habitual_intake

config.Subject.habitual_intake()

Carbohydrate (g/day) and sodium (mg/day) eaten before a diet begins.

The reference the water compartments measure deviations from, so that a diet starting from ordinary eating starts with both at zero and the first days’ drop in glycogen and sodium water is predicted rather than absorbed as noise or credited to tissue. Defaults are roughly US adult averages by sex; set them in subject.toml if you know better. On the record so far, the fit is better with any plausible habitual level than with the alternative, the diet’s own mean.

rmr_mifflin

config.Subject.rmr_mifflin(weight_kg)

Mifflin-St Jeor resting metabolic rate, kcal/day.

Note this predicts resting expenditure only. Total daily expenditure is larger by the physical-activity level and the thermic effect of food.