ENH: disp more results to rocketpy.plots & rocketpy.prints #753
Conversation
Gui-FernandesBR
force-pushed
the
enh/more-monte-carlo-results
branch
from
2e99e7f to
2c2641d
Compare
| pt = self.monte_carlo.results[key] | ||
| print (f"{key:>25} {value[0]:>15.3f} {value[1]:>15.3f} {np.quantile(pt,0):>15.3f} {np.quantile(pt,0.025):>15.3f} {np.quantile(pt,0.5):>15.3f} {np.quantile(pt,0.975):>15.3f} {np.quantile(pt,1):>15.3f}") | ||
| except TypeError: | ||
| print(f"{key:>25} {str(value[0]):>15} {str(value[1]):>15}") | ||
| print (f"{key:>25} {str(value[0]):>15} {str(value[1]):>15} {str(np.quantile(pt,0)):>15} {str(np.quantile(pt,0.025)):>15} {str(np.quantile(pt,0.5)):>15} {str(np.quantile(pt,0.975)):>15} {str(np.quantile(pt,1)):>15}") |
There was a problem hiding this comment.
I believe "min" and "max" are way more clear than "0% quantile" and "100% quantile".
| ], | ||
| "files.associations": { | ||
| "plyconfig.json": "jsonc" | ||
| } |
There was a problem hiding this comment.
Unrelated to the issue.
| figure, plt = plt.subplots(3,1,sharex=True,gridspec_kw={'height_ratios':[1,3]}) | ||
|
|
||
| plt[0].boxplot(self.monte_carlo.results[key],vert=False) | ||
| plt[0].ytick([]) | ||
|
|
||
| plt[1].hist(self.monte_carlo.results[key]) | ||
| plt[1].title(f"Histogram of {key}") | ||
| plt[1].ylabel("Number of Occurrences") | ||
|
|
||
| plt[2].hist(self.monte_carlo.results[key], density=True) | ||
| plt[2].title(f" Density {key}") | ||
| plt[2].ylabel("Probability Density") | ||
| kde = kde.gaussian_kde(self.monte_carlo.results[key]) | ||
| x_array = np.linspace(min(self.monte_carlo.results[key]), max(self.monte_carlo.results[key]), 100) | ||
| plt[2].plot(x_array, kde(x_array), label='KDE') | ||
| plt.show() |
There was a problem hiding this comment.
This generates errors when I try to create plot MonteCarlo results. Indeed, you are replacing a global variable, the alias plt for the submodule matplotlib.pyplot for a local variable with the same name. Moreover, where does kde come from? It seems you have not imported it. Finally, I tried to do a quick clean of the errors just to get a gist of how it would look, and it seems that it creates three different plots. As mentioned in the PR, the idea was to create a single plot that blends the three (histogram, boxplot and density plot).
Gui-FernandesBR
force-pushed
the
enh/more-monte-carlo-results
branch
from
2c2641d to
cdbf376
Compare
See issue 730
Pull request type
Checklist
black rocketpy/ tests/) has passed locallypytest tests -m slow --runslow) have passed locallyCHANGELOG.mdhas been updated (if relevant)Current behavior
NA
New behavior
NA
Breaking change