Every row on the same date, different values

intentional difference ← All linear growth benchmarks

Three rows sharing one timestamp. A trend over zero time is undefined. Python Prophet's optimizer crashes on a NaN starting value; we return a FittingError before fitting.

Input

benchmark/inputs/v1/training-zero-span-varied.json: 3 history rows and 2 forecast rows.

What happened

Effect Prophet

FittingError: Linear MAP fitting requires a positive training time range [reason=degenerate-observations]

Python Prophet

RuntimeError: Error during optimization! Command '/opt/effect-prophet-benchmark/lib/python3.12/site-packages/prophet/stan_model/prophet_model.bin random seed=88090 data file=/tmp/tmpr6ay_rmv/k6nk7dgg.…
Exception: normal_lpdf: Random variable is nan, but must be not nan! (in 'prophet.stan', line 130, column 2 to column 19)

Output

Neither library produced a usable forecast for this input.

Timing

This case isn't timed, because only matching cases are.

Settings

The case declaration both libraries ran.

{
  "kind": "map",
  "comparison": {
    "kind": "equivalent-objective",
    "evidenceId": "linear-output-first-v4"
  },
  "configuration": {
    "growth": "linear",
    "scaling": "absmax",
    "seasonalityMode": "additive",
    "holidaysMode": "additive",
    "changepoints": {
      "mode": "auto",
      "count": 25,
      "range": 0.8
    },
    "changepointPriorScale": 0.05,
    "seasonalities": [],
    "events": [],
    "regressors": []
  },
  "effectOptimizer": {
    "algorithm": "auto",
    "maxIterations": 10000,
    "fallback": "newton",
    "lbfgs": {
      "historySize": 5,
      "initAlpha": 0.001,
      "tolObj": 1e-12,
      "tolRelObj": 10000,
      "tolGrad": 1e-8,
      "tolRelGrad": 10000000,
      "tolParam": 1e-8
    }
  },
  "pythonOptimizer": {
    "algorithm": "Auto",
    "maxIterations": 10000,
    "newtonFallback": true,
    "sigFigs": 12
  }
}