Ranges with an explicit floor

match ← All logistic growth benchmarks

80% ranges from 512 simulated paths for a model bounded by a floor and ceiling.

Input

benchmark/inputs/v1/logistic-explicit-floor.json: 96 history rows and 24 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
2020-04-06	65.8327	65.8327	1.1e-11
2020-04-07	66.2481	66.2481	1.1e-11
2020-04-08	67.2854	67.2854	1.1e-11
2020-04-09	67.5699	67.5699	1.1e-11
2020-04-10	67.9570	67.9570	1.1e-11
2020-04-11	68.0033	68.0033	1.0e-11
2020-04-12	67.8977	67.8977	1.0e-11
2020-04-13	67.9429	67.9429	9.9e-12
2020-04-14	68.6956	68.6956	9.4e-12
2020-04-15	68.9348	68.9348	9.5e-12
2020-04-16	69.5493	69.5493	9.3e-12
2020-04-17	69.8952	69.8952	8.9e-12
2020-04-18	69.9016	69.9016	8.7e-12
2020-04-19	69.7572	69.7572	8.5e-12
2020-04-20	70.1556	70.1556	7.9e-12
2020-04-21	70.0990	70.0990	8.1e-12
2020-04-22	70.6851	70.6851	7.9e-12
2020-04-23	71.2654	71.2654	7.7e-12
2020-04-24	71.5783	71.5783	7.4e-12
2020-04-25	71.5528	71.5528	7.2e-12
2020-04-26	71.7759	71.7759	6.6e-12
2020-04-27	71.3554	71.3554	6.8e-12
2020-04-28	71.6607	71.6607	6.7e-12
2020-04-29	72.2190	72.2190	6.5e-12

Timing

Median of all measured repetitions.

Step	Effect Prophet	Python Prophet	Effect Prophet is
Convert input rows	43 µs	492 µs	11× faster
Save model to JSON	34 µs	499 µs	15× faster
Load model from JSON	200 µs	2.52 ms	13× faster
Simulate uncertainty	945 µs	89.0 ms	94× faster
First simulation in a new process	185 ms	1.59 s	8.6× faster
Load and simulate in a new process	154 ms	1.55 s	10× faster

Settings

The case declaration both libraries ran.

{
  "kind": "map",
  "comparison": {
    "kind": "equivalent-objective",
    "evidenceId": "logistic-map-v1"
  },
  "configuration": {
    "growth": "logistic",
    "scaling": "minmax",
    "seasonalityMode": "additive",
    "holidaysMode": "additive",
    "changepoints": {
      "mode": "explicit",
      "timestamps": [
        "2020-02-10T00:00:00.000Z"
      ]
    },
    "changepointPriorScale": 0.2,
    "seasonalities": [
      {
        "name": "weekly-custom",
        "periodDays": 7,
        "fourierOrder": 2,
        "priorScale": 10
      }
    ],
    "events": [],
    "regressors": [
      {
        "name": "promotion",
        "priorScale": 10,
        "standardization": "never",
        "mode": "multiplicative"
      }
    ]
  },
  "pythonOptimizer": {
    "algorithm": "Auto",
    "maxIterations": 10000,
    "newtonFallback": true,
    "sigFigs": 12
  },
  "uncertainty": {
    "seed": 19,
    "samples": 512,
    "intervalWidth": 0.8,
    "output": "intervals",
    "algorithm": "scalar-continuous-time",
    "evidence": "docs/validation/uncertainty.md#evidence-classes",
    "samplerMemoryLimitBytes": 67108864
  }
}