Goals

A goal is a declared outcome: forecast this. You name a target signal and a goal type; the platform does the rest — it profiles your data, discovers what drives the target, runs a tournament to find the best model, deploys the winner, and keeps the forecast live.

create ──▶ profile ──▶ discover ──▶ tournament ──▶ champion ──▶ live forecast
                                                                    │
                                              alerts ◀─── monitor ──┤
                                              retrain ◀── autopilot ┘

Two fields are all it takes:

curl -X POST "$API_BASE/v1/goals" \
  -H "Authorization: Bearer $TOKEN" \
  -H "X-Workspace-Id: $WORKSPACE_ID" \
  -H "Content-Type: application/json" \
  -d '{ "target_key": "daily_sales", "goal_type": "forecast_value" }'

from predictai import PredictAI

client = PredictAI(token="pa_live_…", workspace_id="ws_…")

data = client.goals.post_goals(json={
    "target_key": "daily_sales",
    "goal_type": "forecast_value",
})

import { PredictAI } from "@predictai/sdk";

const client = new PredictAI({ token: "pa_live_…", workspaceId: "ws_…" });

const data = await client.goals.postGoals({
  json: {
    target_key: "daily_sales",
    goal_type: "forecast_value",
  },
});

Everything you omit — horizon, resolution, lags, model lineup — is resolved from your measured data, and re-resolved on every run. Everything the goal produces hangs off the same object: the discovered relationship graph, the tournament leaderboard, the live forecast, what-if scenarios, a conversational analyst, and alerts.

In this section

Prefer full manual control? Build the loop yourself with Pipelines — a goal's champion is a pipeline under the hood, and the two coexist in the same workspace.