### Install development version from GitHub

Source: https://github.com/arviz-devs/preliz/blob/main/docs/installation.rst

Installs the latest development version of PreliZ directly from the main branch on GitHub.

```bash
pip install git+https://github.com/arviz-devs/preliz.git
```

--------------------------------

### Install with Jupyter Notebook support

Source: https://github.com/arviz-devs/preliz/blob/main/README.md

Installs PreliZ with optional dependencies for full functionality in Jupyter Notebook.

```bash
pip install "preliz[full,notebook]"
```

--------------------------------

### Install with JupyterLab support

Source: https://github.com/arviz-devs/preliz/blob/main/README.md

Installs PreliZ with optional dependencies for full functionality in JupyterLab.

```bash
pip install "preliz[full,lab]"
```

--------------------------------

### Install development version

Source: https://github.com/arviz-devs/preliz/blob/main/README.md

Installs the latest development version of PreliZ from the main branch using pip.

```bash
pip install git+git://github.com/arviz-devs/preliz.git
```

--------------------------------

### Install latest release (base dependencies)

Source: https://github.com/arviz-devs/preliz/blob/main/README.md

Installs the latest version of PreliZ with its base dependencies from PyPI.

```bash
pip install preliz
```

--------------------------------

### Install from conda-forge

Source: https://github.com/arviz-devs/preliz/blob/main/README.md

Installs PreliZ from the conda-forge channel.

```bash
conda install -c conda-forge preliz
```

--------------------------------

### From samples to maximum likelihood distributions (second example)

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Another example of fitting data samples to distributions using `pz.mle`, demonstrating the function's behavior with different sample parameters.

```python
# In a real scenario this will be some data and not a sample from a PreliZ distribution
sample = pz.StudentT(5000, 0, 1).rvs(1000)

dist0 = pz.StudentT()
dist1 = pz.Normal()
dist2 = pz.Laplace()
pz.mle([dist0, dist1, dist2], sample, plot=3);  # we ask to plot all 3 distribution
```

--------------------------------

### Censored Distribution Example

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Demonstrates creating a censored normal distribution.

```python
dist = pz.Normal(mu=0, sigma=1)
censored_dist = pz.Censored(pz.Normal(mu=0, sigma=1), lower=1)
```

--------------------------------

### Get distribution support

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Retrieves the support of the distribution.

```python
dist.support
```

--------------------------------

### Summary of Distributions

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Shows how to get a summary for a normal distribution and a censored normal distribution.

```python
dist.summary()
censored_dist.summary()
```

--------------------------------

### From samples to maximum likelihood distributions

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Example of fitting data samples to a list of distributions using PreliZ's `mle` function to find the maximum likelihood estimate. Includes plotting the best fit.

```python
# In a real scenario this will be some data and not a sample from a PreliZ distribution
sample = pz.StudentT(4, 0, 1).rvs(1000)

dist0 = pz.StudentT()
dist1 = pz.Normal()
dist2 = pz.Laplace()
pz.mle([dist0, dist1, dist2], sample, plot=3);  # we ask to plot all 3 distributions
```

--------------------------------

### From quartiles to distributions

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Example of defining a distribution using PreliZ's `quartile` function, specifying the distribution by its quartiles.

```python
pz.quartile(pz.Gamma(), 2.6, 4.3, 6.6);
```

--------------------------------

### Dirichlet mode example 2

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_ND.md

Creates another Dirichlet distribution with different marginals and bound.

```python
ax, dist = pz.dirichlet_mode([0.2, 0.3, 0.5], bound=0.1)
```

--------------------------------

### Dirichlet mode example 1

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_ND.md

Creates a Dirichlet distribution using dirichlet_mode with specified marginals and bound.

```python
ax, dist = pz.dirichlet_mode([1/3, 1/3, 1/3], bound=0.2)
```

--------------------------------

### Creating PreliZ Distribution from PyMC Distribution

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Shows how to create a PreliZ distribution from an existing PyMC distribution and display its summary.

```python
import pymc as pm

pz.from_pymc(pm.Normal.dist(mu=0, sigma=1)).summary()
```

--------------------------------

### Sample from distribution

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Generates random variates from a Beta distribution.

```python
dist.rvs(10)
```

--------------------------------

### Interactive Exploration with Default Initialization

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Shows interactive plotting of a Gamma distribution using default parameter initialization when none are provided.

```python
import preliz as pz
pz.Gamma().plot_interactive()
```

--------------------------------

### Exporting to PyMC

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Demonstrates exporting a PreliZ Normal distribution to a PyMC distribution format.

```python
pz.Normal(0, 1).to_pymc()
```

--------------------------------

### Exporting to Bambi

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Demonstrates exporting a PreliZ Normal distribution to a Bambi distribution format.

```python
pz.Normal(0, 1).to_bambi()
```

--------------------------------

### Interactive Exploration with Higher Parameter Bounds

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Demonstrates interactive plotting of a Gamma distribution initialized with higher values for the 'alpha' parameter to adjust slider bounds.

```python
import preliz as pz
pz.Gamma(alpha=50, beta=1).plot_interactive()
```

--------------------------------

### Get parameters of the updated Gamma distribution

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Retrieves the alpha and beta parameters of the previously updated Gamma distribution.

```python
dist.alpha, dist.beta
```

--------------------------------

### Interactive Exploration of Distribution Parameters

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Demonstrates interactive plotting of a Gamma distribution, allowing users to adjust parameters via sliders.

```python
import preliz as pz
pz.Gamma(mu=2, sigma=1).plot_interactive()
```

--------------------------------

### PyMC interoperability: Estimating parameters

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Demonstrates how to use PyMC distributions with PreliZ's maxent function by passing np.nan for parameters to be estimated.

```python
import pymc as pm

dist = pm.Gamma.dist(np.nan, np.nan)
new_dist, _ = pz.maxent(dist, 1, 10, 0.9);
new_dist
```

--------------------------------

### Launch the prior predictive assistant

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/ppa.md

Initializes and displays the prior predictive assistant with the defined model.

```python
%matplotlib widget
pz.ppa(a_preliz_model)
```

--------------------------------

### Import preliz

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_ND.md

Imports the preliz library.

```python
import preliz as pz
```

--------------------------------

### View current settings

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/global_settings.md

Displays the current preliz configuration settings.

```python
pz.rcParams
```

--------------------------------

### Import necessary libraries

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/ppa.md

Imports the required libraries for the demonstration.

```python
import preliz as pz
import numpy as np
import pandas as pd
```

--------------------------------

### Import necessary libraries

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Imports matplotlib for plotting, numpy for numerical operations, and preliz for probability distributions.

```python
import matplotlib.pyplot as plt
import numpy as np

import preliz as pz
```

--------------------------------

### PyMC interoperability: Fixing parameters by statistic

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Demonstrates fixing parameters using `fixed_stat` for a Gamma distribution, showing equivalence to `fixed_params` for certain statistics like the mean.

```python
dist = pm.Gamma.dist(np.nan, np.nan)
new_dist, _ = pz.maxent(dist, 1, 10, 0.9, fixed_stat=("mean", 4));
new_dist
```

--------------------------------

### Comparing Multiple Distributions

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Plots the PDFs of two Beta distributions on the same graph to compare parameter impacts.

```python
import preliz as pz
pz.Beta(2, 5).plot_pdf()
pz.Beta(5, 2).plot_pdf()
```

--------------------------------

### Plotting PyMC-extras Prior with PreliZ

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Shows how to plot a Prior object from PyMC-extras using PreliZ.

```python
from pymc_extras.prior import Prior

pz.plot(Prior("Gamma", mu=4, sigma=2));
```

--------------------------------

### Instantiate a Normal distribution

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Instantiates a Normal distribution without specifying parameters.

```python
dist = pz.Normal()
dist
```

--------------------------------

### Plotting the Probability Density Function (PDF)

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Plots the PDF of a Beta distribution with an interval visualization.

```python
import preliz as pz
pz.Beta(2, 5).plot_pdf(pointinterval=True)
```

--------------------------------

### Import necessary libraries

Source: https://github.com/arviz-devs/preliz/blob/main/preliz/tests/plot_interactive.ipynb

Imports ipytest, pytest, and preliz for testing interactive plotting.

```python
import ipytest
import pytest

ipytest.autoconfig()

import preliz as pz
```

--------------------------------

### Interactive plot of Gamma distribution

Source: https://github.com/arviz-devs/preliz/blob/main/paper/paper.md

Demonstrates how to create an interactive plot of a Gamma distribution using `plot_interactive()`, allowing users to adjust parameters with sliders.

```python
pz.Gamma(mu=2, sigma=1).plot_interactive()
```

--------------------------------

### Distribution summary with custom mass

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Calculates and returns the summary of a Beta distribution with a custom mass for the ETI.

```python
dist.summary(mass=0.7)
```

--------------------------------

### PyMC interoperability: Fixing parameters by name

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Illustrates fixing parameters for PyMC distributions by name using the `fixed_params` argument, which is recommended for robustness.

```python
dist = pm.Gamma.dist(np.nan, np.nan)
new_dist, _ = pz.maxent(dist, 1, 10, 0.9, fixed_params={"mu": 4});
new_dist
```

--------------------------------

### All moments

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Calculates all standard moments (mean, variance, skewness, kurtosis) of a Beta distribution in a single call.

```python
dist.moments()
```

--------------------------------

### PyMC-extras interoperability

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Shows how to pass `Prior` objects from PyMC-extras to PreliZ's `maxent` function, allowing partial initialization of Priors.

```python
from pymc_extras.prior import Prior

dist = Prior("Gamma", mu=4)
pz.maxent(dist, 1, 10, 0.9);
```

--------------------------------

### Explore PreliZ Model with predictive_explorer

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/predictive_explorer.md

Demonstrates the use of `predictive_explorer` with a PreliZ model, automatically generating interactive textboxes for parameter exploration.

```python
pz.predictive_explorer(a_preliz_model)
```

--------------------------------

### Import necessary libraries

Source: https://github.com/arviz-devs/preliz/blob/main/preliz/tests/predictive_explorer.ipynb

Imports for ipytest, pytest, numpy, pandas, pymc, and preliz components.

```python
import ipytest
import pytest

ipyt<ctrl62>   
```

```python
predictive_explorer(model, 50, plot_func=lin_reg)
```

--------------------------------

### Distribution summary

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Calculates and returns the mean, median, standard deviation, and equal-tailed interval (ETI) for a Beta distribution.

```python
dist = pz.Beta(2, 5)
dist.summary()
```

--------------------------------

### Roulette Method Initialization

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Initializing the Roulette method for eliciting a prior distribution by drawing on a grid.

```python
%matplotlib widget
result = pz.Roulette()
```

--------------------------------

### Gamma distribution with alpha and beta

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Defines a Gamma distribution using its alpha and beta parameters and displays its summary, demonstrating alternative parametrization.

```python
pz.Gamma(dist.alpha, dist.beta).summary()
```

--------------------------------

### PyMC interoperability: Fixing parameters by value

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Shows how to fix parameters in PyMC distributions by passing their values directly instead of np.nan.

```python
dist = pm.StudentT.dist(nu=7, mu=np.nan, sigma=np.nan)
pz.maxent(dist, 1, 10, 0.9);
```

--------------------------------

### Import necessary libraries and configure ipytest

Source: https://github.com/arviz-devs/preliz/blob/main/preliz/tests/ppa.ipynb

Imports ipytest, pytest, numpy, and distributions from preliz. Also configures ipytest for automatic testing.

```python
%matplotlib widget
import ipytest
import pytest

ipytest.autoconfig()

import numpy as np

from preliz import HalfNormal, Normal, ppa
```

--------------------------------

### Compute maximum entropy priors for Beta and Normal distributions

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Calculates maximum entropy priors for Beta and Normal distributions given lower and upper bounds and a desired mass.

```python
pz.maxent(pz.Beta(), lower=0.3, upper=0.8, mass=0.6)
pz.maxent(pz.Normal(), lower=0.3, upper=0.8, mass=0.6);
```

--------------------------------

### Instantiate and update Gamma distribution

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Instantiates a Gamma distribution and then uses maxent to update it, allowing access to its parameters afterwards.

```python
dist = pz.Gamma(mu=4)
pz.maxent(dist, 1, 10, 0.9);
```

--------------------------------

### Gamma distribution with mu and sigma

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Defines a Gamma distribution using its mean (mu) and standard deviation (sigma) parameters and displays its summary.

```python
dist = pz.Gamma(mu=2, sigma=1)
dist.summary()
```

--------------------------------

### Compute maximum entropy Gamma distribution with fixed mean

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Finds a Gamma distribution with a fixed mean and 90% of the mass between 1 and 10.

```python
pz.maxent(pz.Gamma(mu=4), 1, 10, 0.9);
```

--------------------------------

### Explore PyMC Model with predictive_explorer

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/predictive_explorer.md

Demonstrates the use of `predictive_explorer` with a PyMC model, automatically detecting the PyMC engine.

```python
pz.predictive_explorer(a_pymc_model)
```

--------------------------------

### Prior predictive assistant (ppa)

Source: https://github.com/arviz-devs/preliz/blob/main/paper/paper.md

Demonstrates the experimental prior predictive assistant `pz.ppa` which computes priors consistent with user-made choices of prior predictive samples. Requires `%matplotlib widget` for interactive plots.

```python
def another_preliz_model():
    a = pz.Normal(0, 10).rvs()
    b = pz.HalfNormal(10).rvs()
    y = pz.Normal(a, b).rvs(100)
    return a, b, y

%matplotlib widget
pz.ppa(another_preliz_model)
```

--------------------------------

### Predictive sliders for model exploration

Source: https://github.com/arviz-devs/preliz/blob/main/paper/paper.md

Illustrates the use of `pz.predictive_sliders` to explore how the prior predictive distribution of a model changes with the prior, with `kind_plot='kde'` for kernel density estimation plots.

```python
def a_preliz_model(a_mu, a_sigma):
    a = pz.Normal(a_mu, a_sigma).rvs()
    b = pz.Normal(a, 1).rvs(100)
    return b

pz.predictive_sliders(a_preliz_model, samples=50, kind_plot='kde')
```

--------------------------------

### Plotting Truncated PyMC Distribution with PreliZ

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Demonstrates plotting a truncated PyMC NegativeBinomial distribution using PreliZ.

```python
with pm.Model():
    x = pm.Truncated("x", pm.NegativeBinomial.dist(2.5, 3), 0, 7)
pz.plot(x);
```

--------------------------------

### Plotting using the top-level pz.plot() function (PDF)

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Uses the general pz.plot() function to plot the PDF of a Gamma distribution.

```python
import preliz as pz
pz.plot(pz.Gamma(2, 0.5))
```

--------------------------------

### Percentile point function (PPF)

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Computes the quantiles for a Beta distribution at specified probabilities.

```python
dist.ppf([0.1, 0.5, 0.9])
```

--------------------------------

### Import necessary libraries

Source: https://github.com/arviz-devs/preliz/blob/main/preliz/tests/quartile_int.ipynb

Imports ipytest, pytest, and QuartileInt from preliz.

```python
%matplotlib widget
import ipytest
import pytest

ipytest.autoconfig()

from preliz import QuartileInt
```

--------------------------------

### From quartiles to distributions interactively

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Demonstrates the interactive `quartile_int` function for defining distributions based on quartiles and a list of candidate distribution families. Requires matplotlib widget and ipywidgets.

```python
%matplotlib widget
pz.QuartileInt(1, 2, 3, ["StudentT", "TruncatedNormal", "BetaScaled"]);
```

--------------------------------

### Temporary settings change

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/global_settings.md

Demonstrates using a context manager to temporarily change plot settings.

```python
with pz.rc_context({"plots.show_plot": False}):
    pz.maxent(pz.Normal())
```

--------------------------------

### Plotting PyMC Distribution with PreliZ

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Illustrates plotting a PyMC Normal distribution directly using PreliZ's plot function.

```python
pz.plot(pm.Normal.dist(mu=0, sigma=1));
```

--------------------------------

### Import necessary libraries

Source: https://github.com/arviz-devs/preliz/blob/main/preliz/tests/check_inside_notebook.ipynb

Imports ipytest, pytest, sys, and the check_inside_notebook function from preliz.

```python
import ipytest
import pytest

ipytest.autoconfig()
import sys

from preliz.internal.plot_helper import check_inside_notebook
```

--------------------------------

### Individual moments

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Calculates the mean, variance, skewness, and kurtosis of a Beta distribution individually.

```python
dist.mean(), dist.var(), dist.skewness(), dist.kurtosis()
```

--------------------------------

### Moment Matching

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Approximating a distribution with another by matching its moments (mean and variance).

```python
pz.match_moments(pz.Gamma(7.8, 2),
                   pz.Normal(),
                   )
```

--------------------------------

### Plotting the Probability Mass Function (PMF) for Discrete Distributions

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Plots the PMF of a Poisson distribution. PreliZ uses 'plot_pdf' for both continuous and discrete distributions.

```python
import preliz as pz
pz.Poisson(4.3).plot_pdf()
```

--------------------------------

### Specific moments

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/working_with_distributions.md

Calculates specific moments (mean and standard deviation) of a Beta distribution.

```python
dist.moments(types="md")
```

--------------------------------

### Accessing updated distributions after MLE fit

Source: https://github.com/arviz-devs/preliz/blob/main/docs/examples/gallery/direct_elicitation_1D.md

Shows how to access the updated distribution objects after they have been fitted using the `pz.mle` function.

```python
dist0
```