---
url: https://jer-k.github.io/effect-prophet/benchmarks/flat/flat-level-absmax.md
---

# Noisy level, absmax scaling

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

96 daily rows wobbling slightly around 35.

## Input

[`benchmark/inputs/v1/flat-level.json`](https://github.com/jer-k/effect-prophet/blob/main/benchmark/inputs/v1/flat-level.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 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-07 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-08 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-09 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-10 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-11 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-12 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-13 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-14 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-15 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-16 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-17 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-18 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-19 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-20 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-21 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-22 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-23 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-24 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-25 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-26 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-27 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-28 | 35.0019 | 35.0019 | 1.1e-11 |
| 2020-04-29 | 35.0019 | 35.0019 | 1.1e-11 |

## Timing

Median of all measured repetitions.

| Step | Effect Prophet | Python Prophet | Effect Prophet is |
| --- | ---: | ---: | --- |
| Fit | 1.92 ms | 7.48 ms | 3.9× faster |
| Predict | 265 µs | 2.72 ms | 10× faster |
| Save model to JSON | 20 µs | 465 µs | 24× faster |
| Load model from JSON | 140 µs | 2.10 ms | 15× faster |
| Load and predict in a new process | 159 ms | 1.43 s | 9.0× faster |

## Settings

The case declaration both libraries ran.

```json
{
  "kind": "map",
  "comparison": {
    "kind": "equivalent-objective",
    "evidenceId": "flat-growth-reduced-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
  }
}
```