Sorted rows, hand-picked changepoint

match ← All linear growth benchmarks

96 daily rows with a slope change. Effect and Python fit the same rows; prediction covers 8 days.

Input

benchmark/inputs/v1/training-ordered.json: 96 history rows and 8 forecast rows.

Output

Largest difference: 3.3e-12 (< 0.0001% of the input's range).

Every forecast row from both libraries.

Date	Effect Prophet	Python Prophet	Difference
2020-04-06	32.8040	32.8040	3.3e-12
2020-04-07	33.0341	33.0341	3.2e-12
2020-04-08	33.2642	33.2642	3.2e-12
2020-04-09	33.4943	33.4943	3.2e-12
2020-04-10	33.7243	33.7243	3.1e-12
2020-04-11	33.9544	33.9544	3.1e-12
2020-04-12	34.1845	34.1845	3.0e-12
2020-04-13	34.4145	34.4145	3.0e-12

Timing

Median of all measured repetitions.

Step	Effect Prophet	Python Prophet	Effect Prophet is
Fit	4.01 ms	20.9 ms	5.2× faster
Predict	169 µs	2.69 ms	16× faster
Save model to JSON	22 µs	403 µs	18× faster
Load model from JSON	135 µs	2.06 ms	15× faster
Load and predict in a new process	154 ms	1.48 s	9.6× faster

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": "explicit",
      "timestamps": [
        "2020-01-17T00:00:00.000Z"
      ]
    },
    "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
  }
}