---
url: >-
  https://jer-k.github.io/effect-prophet/benchmarks/logistic/uncertainty-logistic-implicit-floor-minmax-full-samples-512.md
---

# Samples, weekly pattern

 [← All logistic growth benchmarks](/benchmarks/logistic)

512 raw simulated values per date over the history and forecast.

## Input

[`benchmark/inputs/v1/logistic-implicit-floor.json`](https://github.com/jer-k/effect-prophet/blob/main/benchmark/inputs/v1/logistic-implicit-floor.json): 96 history rows and 24 forecast rows.

## Output

Largest difference: **1.9e-11** (< 0.0001% of the input's range).

Every forecast row from both libraries.

| Date | Effect Prophet | Python Prophet | Difference |
| --- | ---: | ---: | ---: |
| 2020-04-06 | 64.8977 | 64.8977 | 1.9e-11 |
| 2020-04-07 | 65.3452 | 65.3452 | 1.9e-11 |
| 2020-04-08 | 65.9695 | 65.9695 | 1.9e-11 |
| 2020-04-09 | 66.6587 | 66.6587 | 1.9e-11 |
| 2020-04-10 | 67.1452 | 67.1452 | 1.8e-11 |
| 2020-04-11 | 67.2451 | 67.2451 | 1.8e-11 |
| 2020-04-12 | 67.1872 | 67.1872 | 1.7e-11 |
| 2020-04-13 | 67.3187 | 67.3187 | 1.7e-11 |
| 2020-04-14 | 67.7078 | 67.7078 | 1.6e-11 |
| 2020-04-15 | 68.2754 | 68.2754 | 1.6e-11 |
| 2020-04-16 | 68.9095 | 68.9095 | 1.6e-11 |
| 2020-04-17 | 69.3426 | 69.3426 | 1.6e-11 |
| 2020-04-18 | 69.3907 | 69.3907 | 1.5e-11 |
| 2020-04-19 | 69.2826 | 69.2826 | 1.5e-11 |
| 2020-04-20 | 69.3655 | 69.3655 | 1.4e-11 |
| 2020-04-21 | 69.7075 | 69.7075 | 1.4e-11 |
| 2020-04-22 | 70.2296 | 70.2296 | 1.3e-11 |
| 2020-04-23 | 70.8197 | 70.8197 | 1.3e-11 |
| 2020-04-24 | 71.2103 | 71.2103 | 1.3e-11 |
| 2020-04-25 | 71.2173 | 71.2173 | 1.3e-11 |
| 2020-04-26 | 71.0696 | 71.0696 | 1.2e-11 |
| 2020-04-27 | 71.1143 | 71.1143 | 1.1e-11 |
| 2020-04-28 | 71.4195 | 71.4195 | 1.1e-11 |
| 2020-04-29 | 71.9060 | 71.9060 | 1.1e-11 |

## Timing

Median of all measured repetitions.

| Step | Effect Prophet | Python Prophet | Effect Prophet is |
| --- | ---: | ---: | --- |
| Convert input rows | 22 µs | 482 µs | 22× faster |
| Save model to JSON | 29 µs | 468 µs | 16× faster |
| Load model from JSON | 143 µs | 2.33 ms | 16× faster |
| Simulate uncertainty | 1.22 ms | 80.4 ms | Not compared: different work |
| First simulation in a new process | 200 ms | 1.54 s | Not compared: different work |
| Load and simulate in a new process | 165 ms | 1.50 s | Not compared: different work |

Some steps aren't compared, because the two libraries' public calls produce different outputs. Their times show what each call costs, not which library is faster.

* **Simulate uncertainty, First simulation in a new process, Load and simulate in a new process:** Python's `predictive_samples(vectorized=False)` and Effect Prophet's `predictUncertainty` are different public calls: Python's timing includes resetting NumPy's random generator and returns NumPy matrices, and the two don't draw the same random numbers.

## Settings

The case declaration both libraries ran.

```json
{
  "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": []
  },
  "pythonOptimizer": {
    "algorithm": "Auto",
    "maxIterations": 10000,
    "newtonFallback": true,
    "sigFigs": 12
  },
  "uncertainty": {
    "seed": 19,
    "samples": 512,
    "intervalWidth": 0.8,
    "output": "samples",
    "algorithm": "scalar-continuous-time",
    "evidence": "docs/validation/uncertainty.md#evidence-classes",
    "samplerMemoryLimitBytes": 67108864
  }
}
```