No-calendar empty control

match ← All flat growth benchmarks

The same history and forecast rows as the empty-country case, with no events or embedded country data.

Input

benchmark/inputs/v5/country-basic-96.json: 96 history rows and 4 forecast rows.

Output

Largest difference: 1.1e-11 (< 0.0001% of the input's range).

Every forecast row from both libraries.

Date	Effect Prophet	Python Prophet	Difference
2024-12-25	20.0473	20.0473	1.1e-11
2025-01-01	20.0473	20.0473	1.1e-11
2025-01-02	20.0473	20.0473	1.1e-11
2025-01-05	20.0473	20.0473	1.1e-11

Timing

Median of all measured repetitions.

Step	Effect Prophet	Python Prophet	Effect Prophet is
Convert input rows	12 µs	458 µs	39× faster
Fit	1.78 ms	8.19 ms	4.6× faster
Predict	123 µs	2.59 ms	21× faster
Fit and predict	1.21 ms	10.8 ms	8.9× faster
Convert, fit and predict	1.09 ms	11.2 ms	10× faster
First forecast in a new process	170 ms	1.42 s	8.3× faster
Save model to JSON	21 µs	427 µs	20× faster
Load model from JSON	85 µs	2.06 ms	24× faster
Load and predict in a new process	166 ms	1.40 s	8.4× faster

Settings

The case declaration both libraries ran.

{
  "kind": "map",
  "comparison": {
    "kind": "equivalent-objective",
    "evidenceId": "country-calendar-public-v1"
  },
  "configuration": {
    "growth": "flat",
    "scaling": "absmax",
    "seasonalityMode": "additive",
    "holidaysMode": "additive",
    "changepoints": {
      "mode": "auto",
      "count": 25,
      "range": 0.8
    },
    "changepointPriorScale": 0.05,
    "seasonalities": [],
    "events": [],
    "regressors": []
  },
  "pythonOptimizer": {
    "algorithm": "Auto",
    "maxIterations": 10000,
    "newtonFallback": true,
    "sigFigs": 12
  }
}