API reference

Read models & reliability, create datasets and fit, read fleet forecasts, run strategy calculators, and push operational data — through the reliafy-client Python package or the raw HTTP API. Create a free account, then generate a token under Settings › API access.

reliafy-client is a thin Python wrapper over the HTTP API — each call maps to one endpoint (shown on the right). Pure standard library; you bring SurPyval.

Install & authenticate

pip install reliafy-client        # installed as reliafy-client, imported as `reliafy`

import reliafy
reliafy.configure(token="rlf_...")   # or set RELIAFY_TOKEN; base_url= for self-hosted

Create a token under Settings → API access (Pro on Reliafy Cloud).

reliafy.models

models.push(model, name, *, data=True, unit=None)POST /api/import/models

Push a fitted SurPyval model and return its URL.

Parameters

`model`	SurPyval model	required	A fitted SurPyval distribution.
`name`	str	required	Model name.
`data`	bool	optional	Upload the fitted observations too (default True → full plot, refittable). False = params only.
`unit`	str	optional	Unit of the time axis.

Returns

"https://reliafy.com/modelling/m/a1b2…"   # str — open in the app

models.push_params(distribution, params, name, *, unit=None, extras=None)POST /api/import/models

Push a model by distribution + parameter values (no SurPyval object).

Parameters

`distribution`	str	required	e.g. "weibull".
`params`	number[] \| {name,value}[]	required	Parameter values.
`name`	str	required	Model name.
`unit`	str	optional	Time unit.
`extras`	dict	optional	gamma / p / f0 for offset / LFP / zero-inflation.

Returns

"https://reliafy.com/modelling/m/…"   # str

models.list()GET /api/v1/models

Your saved models.

Returns

[ { "id": "a1b2…", "name": "Bearing life", "distribution": "Weibull",
    "kind": "distribution", "n": 30, "unit": "hours", "url": "/modelling/m/…" }, … ]

models.get(model_id)GET /api/v1/models/{id}

One model's fit.

Parameters

model_id str required Model id.

Returns

{ "id": …, "distribution": "Weibull", "params": [{name, value, se, ci}],
  "coefficients": [], "metrics": null, "gof": [{id, label, value}] }

models.reliability(model_id, t=None, covariates=None)POST /api/v1/models/{id}/reliability

Evaluate the reliability functions.

Parameters

`model_id`	str	required	Model id.
`t`	number	optional	Time to evaluate at. Omit for the whole curve.
`covariates`	dict	optional	Covariate values for a proportional-hazards model.

Returns

{ "at": { "reliability": 0.666, "failure": 0.334, "hazard": …,
          "cumulative_hazard": …, "density": … } }   # or {"curves": {...}}

models.fit(dataset_id, distribution, name, *, mapping=None, unit=None, covariates=None, formula=None)POST /api/v1/fit

Fit and save a model from one of your datasets.

Parameters

`dataset_id`	str	required	Dataset to fit.
`distribution`	str	required	e.g. weibull, weibull_ph, best.
`name`	str	required	Model name.
`mapping`	dict	required	Role → column, e.g. {"x": "hours", "c": "failed"}.
`unit`	str	optional	Time unit.
`covariates`	str[]	optional	Covariate columns (PH).
`formula`	str	optional	Formulaic covariate formula.

Returns

{ "id": …, "name": "Bearing life", "distribution": "Weibull", "url": "/modelling/m/…" }

reliafy.data

data.upload(name, *, csv=None, data=None)POST /api/v1/datasets

Create a dataset from CSV text or column arrays.

Parameters

`name`	str	required	Dataset name.
`csv`	str	optional	Raw CSV text (with a header). Provide this or data.
`data`	dict	optional	Column arrays, e.g. {"hours": [...], "failed": [...]}.

Returns

{ "id": "d3…", "name": "Bearings", "n_rows": 8, "columns": ["hours","failed"], "url": "/datasets/d/…" }

reliafy.strategy

strategy.optimal_replacement(distribution, params, planned_cost, unplanned_cost, *, unit=None)POST /api/v1/strategy/optimal-replacement

Cost-optimal preventive-replacement interval.

Parameters

`distribution`	str	required	e.g. weibull.
`params`	number[]	required	Distribution parameters.
`planned_cost`	number	required	Cost of a planned replacement.
`unplanned_cost`	number	required	Cost of a failure.
`unit`	str	optional	Time unit.

Returns

{ "optimal_time": 580.5, "optimal_cost_rate": 0.51, "beneficial": true, "mttf": 1272.4 }

strategy.failure_finding(distribution, params, target_availability, *, unit=None)POST /api/v1/strategy/failure-finding

Inspection interval for a hidden (protective) function.

Parameters

`distribution`	str	required	e.g. exponential.
`params`	number[]	required	Distribution parameters.
`target_availability`	number	required	Target availability in (0,1), e.g. 0.99.
`unit`	str	optional	Time unit.

Returns

{ "interval": 176.4, "method": "…", "mttf": 8760, "target_availability": 0.99 }

reliafy.fleet

fleet.forecast(fleet_id)GET /api/v1/fleets/{id}/forecast

The live failure forecast for one of your fleets.

Parameters

fleet_id str required Fleet id (from /fleet/forecasts/<id>).

Returns

{ "fleet": {id, name, model_id}, "forecast": {status, method, periods, expected_failures, …} }

Errors raise reliafy.ReliafyError; reads/writes are scoped to your own data.
Full request/response field tables are on the HTTP API tab.