Python library for conducting design of experiments.
pip install doex
Try out doex
in an online Jupyter Notebook -
import doex
exp = doex.OneWayANOVA(
[24, 28, 37, 30], # Treatment 1
[37, 44, 31, 35], # Treatment 2
[42, 47, 52, 38], # Treatment 3
)
+---------------------+-----+----------------+---------------------+-------------+---------+
| Source of Variation | DOF | Sum of Squares | Mean Sum of Squares | F statistic | p value |
+---------------------+-----+----------------+---------------------+-------------+---------+
| Treatments | 2 | 450.6667 | 225.3333 | 7.0356 | 0.0145 |
| Error | 9 | 288.2500 | 32.0278 | | |
| Total | 11 | 738.9167 | | | |
+---------------------+-----+----------------+---------------------+-------------+---------+
Visit the doex documentation.
- Completely Randomized Design / One-Way ANOVA
- Randomized Complete Block Design / Two-Way ANOVA
- Latin Square Design
- Graeco-Latin Square Design
- Randomized Complete Block Design with missing values
- Balanced Incomplete Block Design (TODO)
- Factorial Designs (TODO)
- Adapted from https://github.com/pawanaichra/doe.