Explicit floor, minmax scaling

match ← All logistic growth benchmarks

The same bounds, scaled between the lowest floor and the highest ceiling.

Input

benchmark/inputs/v2/logistic-floor-96.json: 96 history rows and 120 forecast rows.

Output

Largest difference: 3.2e-11 (< 0.0001% of the input's range).

Every forecast row from both libraries.

Date	Effect Prophet	Python Prophet	Difference
2020-01-01	24.4855	24.4855	8.6e-12
2020-01-02	24.9722	24.9722	9.1e-12
2020-01-03	25.4734	25.4734	9.5e-12
2020-01-04	25.9892	25.9892	9.9e-12
2020-01-05	26.5199	26.5199	1.0e-11
2020-01-06	27.0657	27.0657	1.1e-11
2020-01-07	27.6267	27.6267	1.1e-11
2020-01-08	28.2032	28.2032	1.2e-11
2020-01-09	28.7953	28.7953	1.2e-11
2020-01-10	29.4030	29.4030	1.3e-11
2020-01-11	30.0266	30.0266	1.3e-11
2020-01-12	30.6661	30.6661	1.4e-11
2020-01-13	31.3215	31.3215	1.4e-11
2020-01-14	31.9929	31.9929	1.5e-11
2020-01-15	32.6803	32.6803	1.5e-11
2020-01-16	33.3836	33.3836	1.6e-11
2020-01-17	34.1028	34.1028	1.7e-11
2020-01-18	34.8378	34.8378	1.7e-11
2020-01-19	35.5885	35.5885	1.8e-11
2020-01-20	36.3548	36.3548	1.8e-11
2020-01-21	37.1364	37.1364	1.9e-11
2020-01-22	37.9331	37.9331	1.9e-11
2020-01-23	38.7446	38.7446	2.0e-11
2020-01-24	39.5707	39.5707	2.1e-11
2020-01-25	40.4111	40.4111	2.1e-11
2020-01-26	41.2653	41.2653	2.2e-11
2020-01-27	42.1329	42.1329	2.2e-11
2020-01-28	43.0136	43.0136	2.3e-11
2020-01-29	43.9068	43.9068	2.4e-11
2020-01-30	44.8121	44.8121	2.4e-11
2020-01-31	45.7288	45.7288	2.5e-11
2020-02-01	46.6563	46.6563	2.5e-11
2020-02-02	47.5942	47.5942	2.6e-11
2020-02-03	48.5417	48.5417	2.6e-11
2020-02-04	49.4981	49.4981	2.7e-11
2020-02-05	50.4628	50.4628	2.7e-11
2020-02-06	51.4350	51.4350	2.8e-11
2020-02-07	52.4139	52.4139	2.8e-11
2020-02-08	53.3989	53.3989	2.9e-11
2020-02-09	54.3894	54.3894	2.9e-11
2020-02-10	55.3842	55.3842	2.9e-11
2020-02-11	56.3827	56.3827	3.0e-11
2020-02-12	57.3839	57.3839	3.0e-11
2020-02-13	58.3870	58.3870	3.0e-11
2020-02-14	59.3911	59.3911	3.1e-11
2020-02-15	60.3955	60.3955	3.1e-11
2020-02-16	61.3992	61.3992	3.1e-11
2020-02-17	62.4014	62.4014	3.1e-11
2020-02-18	63.4013	63.4013	3.1e-11
2020-02-19	64.3981	64.3981	3.2e-11
2020-02-20	65.3910	65.3910	3.2e-11
2020-02-21	66.3791	66.3791	3.2e-11
2020-02-22	67.3617	67.3617	3.2e-11
2020-02-23	68.3381	68.3381	3.2e-11
2020-02-24	69.3074	69.3074	3.2e-11
2020-02-25	70.2691	70.2691	3.2e-11
2020-02-26	71.2223	71.2223	3.2e-11
2020-02-27	72.1666	72.1666	3.2e-11
2020-02-28	73.1012	73.1012	3.1e-11
2020-02-29	74.0256	74.0256	3.1e-11
2020-03-01	74.9391	74.9391	3.1e-11
2020-03-02	75.8414	75.8414	3.1e-11
2020-03-03	76.7319	76.7319	3.1e-11
2020-03-04	77.6101	77.6101	3.0e-11
2020-03-05	78.4757	78.4757	3.0e-11
2020-03-06	79.3282	79.3282	3.0e-11
2020-03-07	80.1674	80.1674	3.0e-11
2020-03-08	80.9928	80.9928	2.9e-11
2020-03-09	81.8044	81.8044	2.9e-11
2020-03-10	82.6017	82.6017	2.9e-11
2020-03-11	83.3846	83.3846	2.8e-11
2020-03-12	84.1529	84.1529	2.8e-11
2020-03-13	84.9066	84.9066	2.7e-11
2020-03-14	85.6454	85.6454	2.7e-11
2020-03-15	86.3694	86.3694	2.7e-11
2020-03-16	87.0784	87.0784	2.6e-11
2020-03-17	87.7725	87.7725	2.6e-11
2020-03-18	88.4516	88.4516	2.5e-11
2020-03-19	89.1159	89.1159	2.5e-11
2020-03-20	89.7653	89.7653	2.4e-11
2020-03-21	90.3999	90.3999	2.4e-11
2020-03-22	91.0199	91.0199	2.3e-11
2020-03-23	91.6253	91.6253	2.3e-11
2020-03-24	92.2163	92.2163	2.2e-11
2020-03-25	92.7931	92.7931	2.2e-11
2020-03-26	93.3558	93.3558	2.1e-11
2020-03-27	93.9047	93.9047	2.1e-11
2020-03-28	94.4398	94.4398	2.0e-11
2020-03-29	94.9615	94.9615	2.0e-11
2020-03-30	95.4699	95.4699	1.9e-11
2020-03-31	95.9653	95.9653	1.9e-11
2020-04-01	96.4480	96.4480	1.8e-11
2020-04-02	96.9181	96.9181	1.8e-11
2020-04-03	97.3759	97.3759	1.7e-11
2020-04-04	97.8217	97.8217	1.7e-11
2020-04-05	98.2558	98.2558	1.7e-11
2020-04-06	98.6784	98.6784	1.6e-11
2020-04-07	99.0897	99.0897	1.6e-11
2020-04-08	99.4901	99.4901	1.5e-11
2020-04-09	99.8799	99.8799	1.5e-11
2020-04-10	100.2592	100.2592	1.4e-11
2020-04-11	100.6283	100.6283	1.4e-11
2020-04-12	100.9876	100.9876	1.4e-11
2020-04-13	101.3373	101.3373	1.3e-11
2020-04-14	101.6777	101.6777	1.3e-11
2020-04-15	102.0090	102.0090	1.2e-11
2020-04-16	102.3315	102.3315	1.2e-11
2020-04-17	102.6454	102.6454	1.2e-11
2020-04-18	102.9510	102.9510	1.1e-11
2020-04-19	103.2485	103.2485	1.1e-11
2020-04-20	103.5383	103.5383	1.1e-11
2020-04-21	103.8204	103.8204	1.0e-11
2020-04-22	104.0953	104.0953	1.0e-11
2020-04-23	104.3630	104.3630	9.8e-12
2020-04-24	104.6239	104.6239	9.5e-12
2020-04-25	104.8781	104.8781	9.2e-12
2020-04-26	105.1259	105.1259	8.9e-12
2020-04-27	105.3675	105.3675	8.6e-12
2020-04-28	105.6031	105.6031	8.3e-12
2020-04-29	105.8329	105.8329	8.0e-12

Timing

Median of all measured repetitions.

Step	Effect Prophet	Python Prophet	Effect Prophet is
Fit	11.6 ms	19.2 ms	1.7× faster
Predict	793 µs	3.08 ms	3.9× faster
Save model to JSON	29 µs	460 µs	16× faster
Load model from JSON	174 µs	2.41 ms	14× faster
Load and predict in a new process	155 ms	1.57 s	10× faster

Settings

The case declaration both libraries ran.

{
  "kind": "map",
  "comparison": {
    "kind": "equivalent-objective",
    "evidenceId": "logistic-python-reconciliation-v1"
  },
  "configuration": {
    "growth": "logistic",
    "scaling": "minmax",
    "seasonalityMode": "additive",
    "holidaysMode": "additive",
    "changepoints": {
      "mode": "explicit",
      "timestamps": [
        "2020-02-08T00:00:00.000Z"
      ]
    },
    "changepointPriorScale": 0.05,
    "seasonalities": [],
    "events": [],
    "regressors": []
  },
  "pythonOptimizer": {
    "algorithm": "Auto",
    "maxIterations": 10000,
    "newtonFallback": true,
    "sigFigs": 12
  }
}