Book a Demo!
CoCalc Logo Icon
StoreFeaturesDocsShareSupportNewsAboutPoliciesSign UpSign In
probml
GitHub Repository: probml/pyprobml
 Path: blob/master/notebooks/misc/hilbert_space_gp_birthdays_numpyro.ipynb
1192 views
Kernel: Python 3

Open In Colab

%matplotlib inline
!pip install numpyro
Collecting numpyro Downloading numpyro-0.8.0-py3-none-any.whl (264 kB) |████████████████████████████████| 264 kB 5.2 MB/s eta 0:00:01 Requirement already satisfied: jax>=0.2.13 in /usr/local/lib/python3.7/dist-packages (from numpyro) (0.2.25) Requirement already satisfied: tqdm in /usr/local/lib/python3.7/dist-packages (from numpyro) (4.62.3) Requirement already satisfied: jaxlib>=0.1.65 in /usr/local/lib/python3.7/dist-packages (from numpyro) (0.1.71+cuda111) Requirement already satisfied: numpy>=1.18 in /usr/local/lib/python3.7/dist-packages (from jax>=0.2.13->numpyro) (1.19.5) Requirement already satisfied: typing-extensions in /usr/local/lib/python3.7/dist-packages (from jax>=0.2.13->numpyro) (3.10.0.2) Requirement already satisfied: absl-py in /usr/local/lib/python3.7/dist-packages (from jax>=0.2.13->numpyro) (0.12.0) Requirement already satisfied: scipy>=1.2.1 in /usr/local/lib/python3.7/dist-packages (from jax>=0.2.13->numpyro) (1.4.1) Requirement already satisfied: opt-einsum in /usr/local/lib/python3.7/dist-packages (from jax>=0.2.13->numpyro) (3.3.0) Requirement already satisfied: flatbuffers<3.0,>=1.12 in /usr/local/lib/python3.7/dist-packages (from jaxlib>=0.1.65->numpyro) (2.0) Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from absl-py->jax>=0.2.13->numpyro) (1.15.0) Installing collected packages: numpyro Successfully installed numpyro-0.8.0

Example: Hilbert space approximation for Gaussian processes.

This example replicates a few of the models in the excellent case study by Aki Vehtari [1] (originally written using R and Stan). The case study uses approximate Gaussian processes [2] to model the relative number of births per day in the US from 1969 to 1988. The Hilbert space approximation is way faster than the exact Gaussian processes because it circumvents the need for inverting the covariance matrix.

The original case study presented by Aki also emphasizes the iterative process of building a Bayesian model, which is excellent as a pedagogical resource. Here, however, we replicate only 4 out of all the models available in [1]. There are a few minor differences in the mathematical details of our models, which we had to make in order for the chains to mix properly. We have clearly commented on the places where our models are different.

References:

1. Gelman, Vehtari, Simpson, et al (2020), `"Bayesian workflow book - Birthdays"
   <https://avehtari.github.io/casestudies/Birthdays/birthdays.html>`.

2. Riutort-Mayol G, Bürkner PC, Andersen MR, et al (2020),
   "Practical hilbert space approximate bayesian gaussian processes for probabilistic programming".

https://arxiv.org/pdf/2004.11408.pdf

import argparse
import functools
import operator
import os

import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

import jax
import jax.numpy as jnp
from tensorflow_probability.substrates import jax as tfp

import numpyro
from numpyro import deterministic, plate, sample
import numpyro.distributions as dist
from numpyro.infer import MCMC, NUTS


# --- utility functions
def load_data():
    URL = "https://raw.githubusercontent.com/avehtari/casestudies/master/Birthdays/data/births_usa_1969.csv"
    data = pd.read_csv(URL, sep=",")
    day0 = pd.to_datetime("31-Dec-1968")
    dates = [day0 + pd.Timedelta(f"{i}d") for i in data["id"]]
    data["date"] = dates
    data["births_relative"] = data["births"] / data["births"].mean()
    return data


def save_samples(out_path, samples):
    """
    Save dictionary of arrays using numpys compressed binary format
    Fast reading and writing and efficient storage
    """
    np.savez_compressed(out_path, **samples)


class UnivariateScaler:
    """
    Standardizes the data to have mean 0 and unit standard deviation.
    """

    def __init__(self):
        self._mean = None
        self._std = None

    def fit(self, x):
        self._mean = np.mean(x)
        self._std = np.std(x)
        return self

    def transform(self, x):
        return (x - self._mean) / self._std

    def inverse_transform(self, x):
        return x * self._std + self._mean


def _agg(*args, scaler=None):
    """
    Custom function for aggregating the samples
    and transforming back to the desired scale.
    """
    total = functools.reduce(operator.add, args)
    return (100 * scaler.inverse_transform(total)).mean(axis=0)


# --- modelling functions
def modified_bessel_first_kind(v, z):
    v = jnp.asarray(v, dtype=float)
    return jnp.exp(jnp.abs(z)) * tfp.math.bessel_ive(v, z)


def spectral_density(w, alpha, length):
    c = alpha * jnp.sqrt(2 * jnp.pi) * length
    e = jnp.exp(-0.5 * (length**2) * (w**2))
    return c * e


def diag_spectral_density(alpha, length, L, M):
    """spd for squared exponential kernel"""
    sqrt_eigenvalues = jnp.arange(1, 1 + M) * jnp.pi / 2 / L
    return spectral_density(sqrt_eigenvalues, alpha, length)


def phi(x, L, M):
    """
    The first `M` eigenfunctions of the laplacian operator in `[-L, L]`
    evaluated at `x`. These are used for the approximation of the
    squared exponential kernel.
    """
    m1 = (jnp.pi / (2 * L)) * jnp.tile(L + x[:, None], M)
    m2 = jnp.diag(jnp.linspace(1, M, num=M))
    num = jnp.sin(m1 @ m2)
    den = jnp.sqrt(L)
    return num / den


def diag_spectral_density_periodic(alpha, length, M):
    """
    Not actually a spectral density but these are used in the same
    way. These are simply the first `M` coefficients of the Taylor
    expansion approximation for the periodic kernel.
    """
    a = length ** (-2)
    J = jnp.arange(1, M + 1)
    q2 = (2 * alpha**2 / jnp.exp(a)) * modified_bessel_first_kind(J, a)
    return q2


def phi_periodic(x, w0, M):
    """
    Basis functions for the approximation of the periodic kernel.
    """
    m1 = jnp.tile(w0 * x[:, None], M)
    m2 = jnp.diag(jnp.linspace(1, M, num=M))
    mw0x = m1 @ m2
    return jnp.cos(mw0x), jnp.sin(mw0x)


# --- models
class GP1:
    """
    Long term trend Gaussian process
    """

    def __init__(self):
        self.x_scaler = UnivariateScaler()
        self.y_scaler = UnivariateScaler()

    def model(self, x, L, M, y=None):
        # intercept
        intercept = sample("intercept", dist.Normal(0, 1))

        # long term trend
        ρ = sample("ρ", dist.LogNormal(-1.0, 1.0))
        α = sample("α", dist.HalfNormal(1.0))
        eigenfunctions = phi(x, L, M)
        spd = jnp.sqrt(diag_spectral_density(α, ρ, L, M))

        with plate("basis1", M):
            β1 = sample("β1", dist.Normal(0, 1))

        f1 = deterministic("f1", eigenfunctions @ (spd * β1))
        μ = deterministic("μ", intercept + f1)
        σ = sample("σ", dist.HalfNormal(0.5))
        with plate("n_obs", x.shape[0]):
            sample("y", dist.Normal(μ, σ), obs=y)

    def get_data(self):
        data = load_data()
        x = data["id"].values
        y = data["births_relative"].values
        self.x_scaler.fit(x)
        self.y_scaler.fit(y)
        xsd = jnp.array(self.x_scaler.transform(x))
        ysd = jnp.array(self.y_scaler.transform(y))
        return dict(
            x=xsd,
            y=ysd,
            L=1.5 * max(xsd),
            M=10,
        )

    def make_figure(self, samples):
        data = load_data()
        dates = data["date"]
        y = 100 * data["births_relative"]
        μ = 100 * self.y_scaler.inverse_transform(samples["μ"]).mean(axis=0)

        f = plt.figure(figsize=(15, 5))
        plt.axhline(100, color="k", lw=1, alpha=0.8)
        plt.plot(dates, y, marker=".", lw=0, alpha=0.3)
        plt.plot(dates, μ, color="r", lw=2)
        plt.ylabel("Relative number of births")
        plt.xlabel("")
        return f


class GP2:
    """
    Long term trend with year seasonality component.
    """

    def __init__(self):
        self.x_scaler = UnivariateScaler()
        self.y_scaler = UnivariateScaler()

    def model(self, x, w0, J, L, M, y=None):
        intercept = sample("intercept", dist.Normal(0, 1))

        # long term trend
        ρ1 = sample("ρ1", dist.LogNormal(-1.0, 1.0))
        α1 = sample("α1", dist.HalfNormal(1.0))
        eigenfunctions = phi(x, L, M)
        spd = jnp.sqrt(diag_spectral_density(α1, ρ1, L, M))
        with plate("basis", M):
            β1 = sample("β1", dist.Normal(0, 1))

        # year-periodic component
        ρ2 = sample("ρ2", dist.HalfNormal(0.1))
        α2 = sample("α2", dist.HalfNormal(1.0))
        cosines, sines = phi_periodic(x, w0, J)
        spd_periodic = jnp.sqrt(diag_spectral_density_periodic(α2, ρ2, J))
        with plate("periodic_basis", J):
            β2_cos = sample("β2_cos", dist.Normal(0, 1))
            β2_sin = sample("β2_sin", dist.Normal(0, 1))

        f1 = deterministic("f1", eigenfunctions @ (spd * β1))
        f2 = deterministic("f2", cosines @ (spd_periodic * β2_cos) + sines @ (spd_periodic * β2_sin))
        μ = deterministic("μ", intercept + f1 + f2)
        σ = sample("σ", dist.HalfNormal(0.5))
        with plate("n_obs", x.shape[0]):
            sample("y", dist.Normal(μ, σ), obs=y)

    def get_data(self):
        data = load_data()
        x = data["id"].values
        y = data["births_relative"].values
        self.x_scaler.fit(x)
        self.y_scaler.fit(y)
        xsd = jnp.array(self.x_scaler.transform(x))
        ysd = jnp.array(self.y_scaler.transform(y))
        w0 = 2 * jnp.pi / (365.25 / self.x_scaler._std)
        return dict(
            x=xsd,
            y=ysd,
            w0=w0,
            J=20,
            L=1.5 * max(xsd),
            M=10,
        )

    def make_figure(self, samples):
        data = load_data()
        dates = data["date"]
        y = 100 * data["births_relative"]
        y_by_day_of_year = 100 * data.groupby("day_of_year2")["births_relative"].mean()
        μ = 100 * self.y_scaler.inverse_transform(samples["μ"]).mean(axis=0)
        f1 = 100 * self.y_scaler.inverse_transform(samples["f1"]).mean(axis=0)
        f2 = 100 * self.y_scaler.inverse_transform(samples["f2"]).mean(axis=0)

        fig, axes = plt.subplots(1, 2, figsize=(15, 5))
        axes[0].plot(dates, y, marker=".", lw=0, alpha=0.3)
        axes[0].plot(dates, μ, color="r", lw=2, alpha=1, label="Total")
        axes[0].plot(dates, f1, color="C2", lw=3, alpha=1, label="Trend")

        axes[0].set_ylabel("Relative number of births")
        axes[0].set_title("All time")

        axes[1].plot(y_by_day_of_year.index, y_by_day_of_year, marker=".", lw=0, alpha=0.5)
        axes[1].plot(y_by_day_of_year.index, f2[:366], color="r", lw=2, label="Year seaonality")
        axes[1].set_ylabel("Relative number of births")
        axes[1].set_xlabel("Day of year")
        for ax in axes:
            ax.axhline(100, color="k", lw=1, alpha=0.8)
            ax.legend()

        return fig


class GP3:
    """
    Long term trend with yearly seasonaly and slowly varying day-of-week effect.
    """

    def __init__(self):
        self.x_scaler = UnivariateScaler()
        self.y_scaler = UnivariateScaler()

    def model(self, x, day_of_week, w0, J, L, M, L3, M3, y=None):
        intercept = sample("intercept", dist.Normal(0, 1))

        # long term trend
        ρ1 = sample("ρ1", dist.LogNormal(-1.0, 1.0))
        α1 = sample("α1", dist.HalfNormal(1.0))
        eigenfunctions = phi(x, L, M)
        spd = jnp.sqrt(diag_spectral_density(α1, ρ1, L, M))
        with plate("basis", M):
            β1 = sample("β1", dist.Normal(0, 1))

        # year-periodic component
        ρ2 = sample("ρ2", dist.HalfNormal(0.1))
        α2 = sample("α2", dist.HalfNormal(1.0))
        cosines, sines = phi_periodic(x, w0, J)
        spd_periodic = jnp.sqrt(diag_spectral_density_periodic(α2, ρ2, J))
        with plate("periodic_basis", J):
            β2_cos = sample("β2_cos", dist.Normal(0, 1))
            β2_sin = sample("β2_sin", dist.Normal(0, 1))

        # day of week effect
        with plate("plate_day_of_week", 6):
            β_week = sample("β_week", dist.Normal(0, 1))
        # next enforce sum-to-zero -- this is slightly different from Aki's model,
        # which instead imposes Monday's effect to be zero.
        β_week = jnp.concatenate([jnp.array([-jnp.sum(β_week)]), β_week])

        # long term variation of week effect
        α3 = sample("α3", dist.HalfNormal(0.1))
        ρ3 = sample("ρ3", dist.LogNormal(1.0, 1.0))  # prior: very long-term effect
        eigenfunctions_3 = phi(x, L3, M3)
        spd_3 = jnp.sqrt(diag_spectral_density(α3, ρ3, L3, M3))
        with plate("week_trend", M3):
            β3 = sample("β3", dist.Normal(0, 1))

        # combine
        f1 = deterministic("f1", eigenfunctions @ (spd * β1))
        f2 = deterministic("f2", cosines @ (spd_periodic * β2_cos) + sines @ (spd_periodic * β2_sin))
        g3 = deterministic("g3", eigenfunctions_3 @ (spd_3 * β3))
        μ = deterministic("μ", intercept + f1 + f2 + jnp.exp(g3) * β_week[day_of_week])
        σ = sample("σ", dist.HalfNormal(0.5))
        with plate("n_obs", x.shape[0]):
            sample("y", dist.Normal(μ, σ), obs=y)

    def get_data(self):
        data = load_data()
        x = data["id"].values
        y = data["births_relative"].values
        self.x_scaler.fit(x)
        self.y_scaler.fit(y)
        xsd = jnp.array(self.x_scaler.transform(x))
        ysd = jnp.array(self.y_scaler.transform(y))
        w0 = 2 * jnp.pi / (365.25 / self.x_scaler._std)
        dow = jnp.array(data["day_of_week"].values) - 1
        return dict(
            x=xsd,
            day_of_week=dow,
            w0=w0,
            J=20,
            L=1.5 * max(xsd),
            M=10,
            L3=1.5 * max(xsd),
            M3=5,
            y=ysd,
        )

    def make_figure(self, samples):
        data = load_data()
        dates = data["date"]
        y = 100 * data["births_relative"]
        y_by_day_of_year = 100 * (data.groupby("day_of_year2")["births_relative"].mean())
        year_days = y_by_day_of_year.index.values

        μ = samples["μ"]
        intercept = samples["intercept"][:, None]
        f1 = samples["f1"]
        f2 = samples["f2"]
        g3 = samples["g3"]
        β_week = samples["β_week"]
        β_week = np.concatenate([-β_week.sum(axis=1)[:, None], β_week], axis=1)

        fig, axes = plt.subplots(2, 2, figsize=(15, 8), sharey=False, sharex=False)
        axes[0, 0].plot(dates, y, marker=".", lw=0, alpha=0.3)
        axes[0, 0].plot(
            dates,
            _agg(μ, scaler=self.y_scaler),
            color="r",
            lw=0,
            label="Total",
            marker=".",
            alpha=0.5,
        )
        axes[0, 1].plot(dates, y, marker=".", lw=0, alpha=0.3)
        axes[0, 1].plot(dates, _agg(f1, scaler=self.y_scaler), color="r", lw=2, label="Trend")
        axes[1, 0].plot(year_days, y_by_day_of_year, marker=".", lw=0, alpha=0.3)
        axes[1, 0].plot(
            year_days,
            _agg(f2[:, :366], scaler=self.y_scaler),
            color="r",
            lw=2,
            label="Year seasonality",
        )
        axes[1, 1].plot(dates, y, marker=".", lw=0, alpha=0.3)
        for day in range(7):
            dow_trend = (jnp.exp(g3).T * β_week[:, day]).T
            fit = _agg(intercept, f1, dow_trend, scaler=self.y_scaler)
            axes[1, 1].plot(dates, fit, lw=2, color="r")

        axes[0, 0].set_title("Total")
        axes[0, 1].set_title("Long term trend")
        axes[1, 0].set_title("Year seasonality")
        axes[1, 1].set_title("Weekly effects with long term trend")
        for ax in axes.flatten():
            ax.axhline(100, color="k", lw=1, alpha=0.8)
            ax.legend()

        return fig


class GP4:
    """
    Long term trend with yearly seasonaly, slowly varying day-of-week effect,
    and special day effect including floating special days.
    """

    def __init__(self):
        self.x_scaler = UnivariateScaler()
        self.y_scaler = UnivariateScaler()

    def model(
        self,
        x,
        day_of_week,
        day_of_year,
        memorial_days_indicator,
        labour_days_indicator,
        thanksgiving_days_indicator,
        w0,
        J,
        L,
        M,
        L3,
        M3,
        y=None,
    ):
        intercept = sample("intercept", dist.Normal(0, 1))

        # long term trend
        ρ1 = sample("ρ1", dist.LogNormal(-1.0, 1.0))
        α1 = sample("α1", dist.HalfNormal(1.0))
        eigenfunctions = phi(x, L, M)
        spd = jnp.sqrt(diag_spectral_density(α1, ρ1, L, M))
        with plate("basis", M):
            β1 = sample("β1", dist.Normal(0, 1))

        # year-periodic component
        ρ2 = sample("ρ2", dist.HalfNormal(0.1))
        α2 = sample("α2", dist.HalfNormal(1.0))
        cosines, sines = phi_periodic(x, w0, J)
        spd_periodic = jnp.sqrt(diag_spectral_density_periodic(α2, ρ2, J))
        with plate("periodic_basis", J):
            β2_cos = sample("β2_cos", dist.Normal(0, 1))
            β2_sin = sample("β2_sin", dist.Normal(0, 1))

        # day of week effect
        with plate("plate_day_of_week", 6):
            β_week = sample("β_week", dist.Normal(0, 1))
        # next enforce sum-to-zero -- this is slightly different from Aki's model,
        # which instead imposes Monday's effect to be zero.
        β_week = jnp.concatenate([jnp.array([-jnp.sum(β_week)]), β_week])

        # long term separation of week effects
        ρ3 = sample("ρ3", dist.LogNormal(1.0, 1.0))
        α3 = sample("α3", dist.HalfNormal(0.1))
        eigenfunctions_3 = phi(x, L3, M3)
        spd_3 = jnp.sqrt(diag_spectral_density(α3, ρ3, L3, M3))
        with plate("week_trend", M3):
            β3 = sample("β3", dist.Normal(0, 1))

        # Finnish horseshoe prior on day of year effect
        # Aki uses slab_df=100 instead, but chains didn't mix
        # in our case for some reason, so we lowered it to 50.
        slab_scale = 2
        slab_df = 50
        scale_global = 0.1
        τ = sample("τ", dist.HalfCauchy(scale=scale_global * 2))
        c_aux = sample("c_aux", dist.InverseGamma(0.5 * slab_df, 0.5 * slab_df))
        c = slab_scale * jnp.sqrt(c_aux)
        with plate("plate_day_of_year", 366):
            λ = sample("λ", dist.HalfCauchy(scale=1))
            λ_tilde = jnp.sqrt(c) * λ / jnp.sqrt(c + (τ * λ) ** 2)
            β4 = sample("β4", dist.Normal(loc=0, scale=τ * λ_tilde))

        # floating special days
        β5_labour = sample("β5_labour", dist.Normal(0, 1))
        β5_memorial = sample("β5_memorial", dist.Normal(0, 1))
        β5_thanksgiving = sample("β5_thanksgiving", dist.Normal(0, 1))

        # combine
        f1 = deterministic("f1", eigenfunctions @ (spd * β1))
        f2 = deterministic("f2", cosines @ (spd_periodic * β2_cos) + sines @ (spd_periodic * β2_sin))
        g3 = deterministic("g3", eigenfunctions_3 @ (spd_3 * β3))
        μ = deterministic(
            "μ",
            intercept
            + f1
            + f2
            + jnp.exp(g3) * β_week[day_of_week]
            + β4[day_of_year]
            + β5_labour * labour_days_indicator
            + β5_memorial * memorial_days_indicator
            + β5_thanksgiving * thanksgiving_days_indicator,
        )
        σ = sample("σ", dist.HalfNormal(0.5))
        with plate("n_obs", x.shape[0]):
            sample("y", dist.Normal(μ, σ), obs=y)

    def _get_floating_days(self, data):
        x = data["id"].values
        memorial_days = data.loc[
            data["date"].dt.month.eq(5) & data["date"].dt.weekday.eq(0) & data["date"].dt.day.ge(25),
            "id",
        ].values

        labour_days = data.loc[
            data["date"].dt.month.eq(9) & data["date"].dt.weekday.eq(0) & data["date"].dt.day.le(7),
            "id",
        ].values
        labour_days = np.concatenate((labour_days, labour_days + 1))

        thanksgiving_days = data.loc[
            data["date"].dt.month.eq(11)
            & data["date"].dt.weekday.eq(3)
            & data["date"].dt.day.ge(22)
            & data["date"].dt.day.le(28),
            "id",
        ].values
        thanksgiving_days = np.concatenate((thanksgiving_days, thanksgiving_days + 1))

        md_indicators = np.zeros_like(x)
        md_indicators[memorial_days - 1] = 1
        ld_indicators = np.zeros_like(x)
        ld_indicators[labour_days - 1] = 1
        td_indicators = np.zeros_like(x)
        td_indicators[thanksgiving_days - 1] = 1
        return {
            "memorial_days_indicator": md_indicators,
            "labour_days_indicator": ld_indicators,
            "thanksgiving_days_indicator": td_indicators,
        }

    def get_data(self):
        data = load_data()
        x = data["id"].values
        y = data["births_relative"].values
        self.x_scaler.fit(x)
        self.y_scaler.fit(y)
        xsd = jnp.array(self.x_scaler.transform(x))
        ysd = jnp.array(self.y_scaler.transform(y))
        w0 = 2 * jnp.pi / (365.25 / self.x_scaler._std)
        dow = jnp.array(data["day_of_week"].values) - 1
        doy = jnp.array((data["day_of_year2"] - 1).values)
        return dict(
            x=xsd,
            day_of_week=dow,
            day_of_year=doy,
            w0=w0,
            J=20,
            L=1.5 * max(xsd),
            M=10,
            L3=1.5 * max(xsd),
            M3=5,
            y=ysd,
            **self._get_floating_days(data),
        )

    def make_figure(self, samples):
        special_days = {
            "Valentine's": pd.to_datetime("1988-02-14"),
            "Leap day": pd.to_datetime("1988-02-29"),
            "Halloween": pd.to_datetime("1988-10-31"),
            "Christmas eve": pd.to_datetime("1988-12-24"),
            "Christmas day": pd.to_datetime("1988-12-25"),
            "New year": pd.to_datetime("1988-01-01"),
            "New year's eve": pd.to_datetime("1988-12-31"),
            "April 1st": pd.to_datetime("1988-04-01"),
            "Independence day": pd.to_datetime("1988-07-04"),
            "Labour day": pd.to_datetime("1988-09-05"),
            "Memorial day": pd.to_datetime("1988-05-30"),
            "Thanksgiving": pd.to_datetime("1988-11-24"),
        }
        β4 = samples["β4"]
        β5_labour = samples["β5_labour"]
        β5_memorial = samples["β5_memorial"]
        β5_thanksgiving = samples["β5_thanksgiving"]

        day_effect = np.array(β4)
        md_idx = special_days["Memorial day"].day_of_year - 1
        day_effect[:, md_idx] = day_effect[:, md_idx] + β5_memorial
        ld_idx = special_days["Labour day"].day_of_year - 1
        day_effect[:, ld_idx] = day_effect[:, ld_idx] + β5_labour
        td_idx = special_days["Thanksgiving"].day_of_year - 1
        day_effect[:, td_idx] = day_effect[:, td_idx] + β5_thanksgiving
        day_effect = 100 * day_effect.mean(axis=0)

        fig = plt.figure(figsize=(12, 5))
        plt.plot(np.arange(1, 367), day_effect)
        for name, day in special_days.items():
            xs = day.day_of_year
            ys = day_effect[day.day_of_year - 1]
            plt.plot(xs, ys, marker="o", mec="k", c="none", ms=10)
            plt.text(xs - 3, ys, name, horizontalalignment="right")
        plt.title("Special day effect")
        plt.ylabel("Relative number of births")
        plt.xlabel("Day of year")
        plt.xlim([-40, None])
        return fig

NAME_TO_MODEL = {
    "t": GP1,
    "ty": GP2,
    "tyw": GP3,
    "tywd": GP4,
}


def main(args):
    is_sphinxbuild = "NUMPYRO_SPHINXBUILD" in os.environ
    model = NAME_TO_MODEL[args.model]()
    data = model.get_data()
    mcmc = MCMC(
        NUTS(model.model),
        num_warmup=args.num_warmup,
        num_samples=args.num_samples,
        num_chains=args.num_chains,
        progress_bar=False if is_sphinxbuild else True,
    )
    mcmc.run(jax.random.PRNGKey(0), **data)
    if not is_sphinxbuild:
        mcmc.print_summary()
    posterior_samples = mcmc.get_samples()
    if args.save_samples:
        print(f"Saving samples at {args.save_samples}")
        save_samples(args.save_samples, posterior_samples)
    if args.save_figure:
        print(f"Saving figure at {args.save_figure}")
        fig = model.make_figure(posterior_samples)
        fig.savefig(args.save_figure)
        plt.close()

    return model, data, mcmc, posterior_samples

def parse_arguments():
    parser = argparse.ArgumentParser(description="Hilbert space approx for GPs")
    parser.add_argument("--num-samples", nargs="?", default=1000, type=int)
    parser.add_argument("--num-warmup", nargs="?", default=1000, type=int)
    parser.add_argument("--num-chains", nargs="?", default=1, type=int)
    parser.add_argument(
        "--model",
        nargs="?",
        default="tywd",
        help="one of"
        '"t" (Long term trend),'
        '"ty" (t + year seasonality),'
        '"tyw" (t + y + slowly varying weekday effect),'
        '"tywd" (t + y + w + special days effect)',
    )
    parser.add_argument("--device", default="cpu", type=str, help='use "cpu" or "gpu".')
    parser.add_argument("--x64", action="store_true", help="Enable float64 precision")
    parser.add_argument(
        "--save-samples",
        default="",
        type=str,
        help="Path where to store the samples. Must be '.npz' file.",
    )
    parser.add_argument(
        "--save-figure",
        default="",
        type=str,
        help="Path where to save the plot with matplotlib.",
    )
    # args = parser.parse_args() # error in colab
    args, unused = parser.parse_known_args()
    return args
Namespace(device='cpu', model='tywd', num_chains=1, num_samples=1000, num_warmup=1000, save_figure='', save_samples='', x64=False) 
!pwd
/content 
args = parse_arguments()
args.device = 'gpu'
args.num-warmup = 200
args.num-samples = 200
args.save-figure="/content"
args.save-samples="/content"
print(args)

if args.x64:
    numpyro.enable_x64()

numpyro.set_platform(args.device)
numpyro.set_host_device_count(args.num_chains)

model, data, mcmc, posterior_samples = main(args)
Namespace(device='gpu', model='tywd', num_chains=1, num_samples=200, num_warmup=200, save_figure='', save_samples='', x64=False)
sample: 100%|██████████| 400/400 [23:10<00:00, 3.48s/it, 1023 steps of size 3.46e-03. acc. prob=0.71]
mean std median 5.0% 95.0% n_eff r_hat c_aux 1.07 0.22 1.06 0.74 1.46 69.10 1.00 intercept 0.31 0.40 0.30 -0.31 0.98 14.52 1.04 α1 1.83 0.41 1.80 1.31 2.47 20.30 1.00 α2 0.24 0.05 0.22 0.19 0.32 3.14 1.79 α3 0.10 0.05 0.08 0.03 0.17 17.07 1.02 β1[0] -0.62 0.70 -0.53 -1.84 0.36 15.77 1.04 β1[1] -1.52 0.23 -1.55 -1.86 -1.17 12.19 1.00 β1[2] 0.04 0.22 0.04 -0.38 0.33 12.94 1.05 β1[3] -1.03 0.12 -1.04 -1.20 -0.82 17.13 1.00 β1[4] -0.49 0.10 -0.50 -0.65 -0.34 22.70 1.04 β1[5] -1.92 0.16 -1.91 -2.14 -1.64 20.57 1.00 β1[6] -1.09 0.11 -1.10 -1.23 -0.87 7.36 1.00 β1[7] -1.57 0.22 -1.59 -1.87 -1.23 5.79 1.00 β1[8] -0.71 0.14 -0.72 -0.92 -0.50 5.61 1.00 β1[9] -0.90 0.25 -0.91 -1.21 -0.54 6.19 1.00 β2_cos[0] -1.77 0.32 -1.88 -2.24 -1.30 3.10 1.75 β2_cos[1] -0.12 0.06 -0.12 -0.22 -0.04 21.21 1.01 β2_cos[2] 0.34 0.08 0.34 0.21 0.47 7.00 1.34 β2_cos[3] 0.63 0.14 0.69 0.41 0.83 3.54 1.73 β2_cos[4] -0.39 0.10 -0.37 -0.55 -0.23 14.87 1.13 β2_cos[5] -0.32 0.09 -0.30 -0.50 -0.20 8.79 1.21 β2_cos[6] 0.58 0.12 0.58 0.36 0.76 10.68 1.15 β2_cos[7] 0.63 0.15 0.63 0.39 0.85 5.99 1.47 β2_cos[8] -0.77 0.15 -0.76 -1.01 -0.54 6.54 1.33 β2_cos[9] -0.03 0.13 -0.02 -0.24 0.16 42.11 1.06 β2_cos[10] -0.37 0.16 -0.36 -0.61 -0.12 58.51 1.06 β2_cos[11] -0.63 0.22 -0.61 -0.99 -0.33 58.47 1.00 β2_cos[12] -0.80 0.28 -0.78 -1.19 -0.33 58.09 1.00 β2_cos[13] -0.98 0.35 -0.96 -1.56 -0.42 104.59 1.01 β2_cos[14] -0.95 0.38 -0.98 -1.52 -0.29 89.25 1.03 β2_cos[15] -1.24 0.53 -1.25 -2.06 -0.36 93.85 1.00 β2_cos[16] -1.14 0.63 -1.06 -2.10 -0.21 21.26 1.00 β2_cos[17] 0.02 0.66 -0.10 -0.93 0.98 98.24 1.01 β2_cos[18] -0.10 0.85 -0.21 -1.40 1.18 33.35 1.00 β2_cos[19] 0.08 0.95 0.08 -1.63 1.36 53.54 1.00 β2_sin[0] -3.36 0.63 -3.59 -4.10 -2.34 3.00 1.87 β2_sin[1] 2.22 0.41 2.35 1.57 2.80 3.10 1.82 β2_sin[2] -0.05 0.06 -0.04 -0.15 0.02 14.35 1.00 β2_sin[3] -1.00 0.19 -1.01 -1.26 -0.69 3.58 1.67 β2_sin[4] -1.08 0.19 -1.12 -1.33 -0.73 3.72 1.75 β2_sin[5] 0.63 0.12 0.64 0.44 0.80 4.83 1.46 β2_sin[6] -0.26 0.09 -0.25 -0.40 -0.14 31.33 1.05 β2_sin[7] -0.69 0.13 -0.70 -0.96 -0.52 5.64 1.39 β2_sin[8] -0.88 0.17 -0.89 -1.09 -0.56 5.46 1.35 β2_sin[9] -0.21 0.13 -0.21 -0.41 -0.03 38.11 1.02 β2_sin[10] -1.04 0.21 -1.03 -1.33 -0.71 12.52 1.23 β2_sin[11] -1.41 0.27 -1.39 -1.83 -1.04 13.84 1.12 β2_sin[12] -0.91 0.30 -0.89 -1.30 -0.35 67.51 1.01 β2_sin[13] -0.78 0.40 -0.80 -1.41 -0.18 25.50 1.02 β2_sin[14] -1.15 0.49 -1.13 -1.81 -0.42 16.01 1.03 β2_sin[15] -1.12 0.59 -1.09 -2.12 -0.27 58.41 1.00 β2_sin[16] -1.26 0.64 -1.21 -2.43 -0.35 18.72 1.02 β2_sin[17] 0.78 0.75 0.77 -0.51 1.81 35.85 1.03 β2_sin[18] -0.56 0.90 -0.60 -1.64 1.13 48.95 1.00 β2_sin[19] 0.17 0.81 0.20 -1.29 1.29 55.95 1.00 β3[0] -0.07 0.73 -0.19 -1.10 1.12 12.59 1.01 β3[1] -1.50 0.41 -1.48 -2.12 -0.90 5.70 1.25 β3[2] 0.77 0.26 0.77 0.36 1.14 28.78 1.01 β3[3] -0.72 0.41 -0.56 -1.52 -0.25 5.24 1.44 β3[4] 0.37 0.32 0.29 -0.03 0.90 16.86 1.20 β4[0] -1.32 0.06 -1.31 -1.41 -1.24 68.47 1.01 β4[1] -0.75 0.07 -0.75 -0.84 -0.62 30.26 1.00 β4[2] -0.08 0.08 -0.06 -0.22 0.01 18.33 1.06 β4[3] -0.00 0.02 -0.00 -0.03 0.02 78.05 0.99 β4[4] -0.00 0.02 -0.00 -0.04 0.02 39.94 1.00 β4[5] 0.00 0.02 -0.00 -0.03 0.03 82.21 1.00 β4[6] -0.00 0.02 -0.00 -0.03 0.03 9.35 1.23 β4[7] -0.01 0.03 -0.00 -0.06 0.02 75.91 1.00 β4[8] -0.05 0.06 -0.02 -0.15 0.01 22.44 1.00 β4[9] -0.01 0.03 0.00 -0.07 0.02 11.83 1.12 β4[10] 0.05 0.07 0.03 -0.01 0.17 22.07 1.00 β4[11] 0.01 0.02 0.00 -0.03 0.04 17.45 1.00 β4[12] 0.00 0.02 0.00 -0.02 0.02 22.67 1.02 β4[13] 0.02 0.04 0.00 -0.02 0.09 36.89 1.03 β4[14] -0.00 0.02 -0.00 -0.05 0.03 29.77 1.02 β4[15] -0.00 0.02 -0.00 -0.01 0.02 59.24 1.01 β4[16] 0.00 0.01 0.00 -0.02 0.01 125.32 0.99 β4[17] 0.01 0.02 0.00 -0.03 0.04 43.54 1.01 β4[18] 0.01 0.04 0.00 -0.05 0.06 14.14 1.20 β4[19] -0.00 0.02 0.00 -0.03 0.02 45.07 1.00 β4[20] -0.00 0.02 -0.00 -0.04 0.03 41.54 1.01 β4[21] -0.01 0.02 -0.00 -0.04 0.01 23.66 1.00 β4[22] -0.02 0.04 -0.00 -0.09 0.02 6.74 1.29 β4[23] -0.00 0.01 -0.00 -0.02 0.01 72.68 1.01 β4[24] 0.02 0.03 0.00 -0.01 0.08 17.45 1.00 β4[25] 0.01 0.01 0.00 -0.01 0.03 10.75 1.24 β4[26] 0.00 0.02 0.00 -0.03 0.03 20.45 1.09 β4[27] -0.01 0.04 0.00 -0.06 0.04 12.15 1.07 β4[28] -0.01 0.02 -0.00 -0.05 0.03 21.58 1.11 β4[29] -0.02 0.04 -0.00 -0.09 0.02 26.34 1.11 β4[30] -0.01 0.02 -0.00 -0.04 0.01 24.96 1.04 β4[31] 0.01 0.03 0.00 -0.04 0.04 21.02 1.02 β4[32] 0.03 0.05 0.01 -0.01 0.11 6.25 1.40 β4[33] 0.00 0.02 0.00 -0.01 0.04 70.83 1.00 β4[34] -0.01 0.02 -0.00 -0.03 0.02 24.87 1.08 β4[35] -0.00 0.01 -0.00 -0.03 0.02 81.60 1.00 β4[36] -0.00 0.01 -0.00 -0.02 0.02 114.22 0.99 β4[37] -0.00 0.04 0.00 -0.02 0.04 21.86 1.03 β4[38] -0.00 0.02 -0.00 -0.02 0.03 49.97 1.02 β4[39] 0.00 0.02 0.00 -0.01 0.04 29.19 1.04 β4[40] 0.01 0.02 0.00 -0.02 0.03 28.59 1.01 β4[41] 0.00 0.01 0.00 -0.02 0.03 59.10 1.00 β4[42] 0.02 0.04 0.00 -0.03 0.09 21.83 1.00 β4[43] -0.07 0.07 -0.06 -0.19 0.00 27.89 1.00 β4[44] 0.30 0.06 0.30 0.21 0.40 53.20 1.00 β4[45] 0.03 0.05 0.01 -0.01 0.13 5.53 1.52 β4[46] 0.00 0.02 0.00 -0.01 0.06 6.18 1.38 β4[47] -0.00 0.02 0.00 -0.04 0.03 23.18 1.03 β4[48] -0.00 0.02 0.00 -0.05 0.03 145.06 1.00 β4[49] -0.01 0.03 -0.00 -0.04 0.02 35.34 1.11 β4[50] 0.00 0.01 0.00 -0.01 0.02 70.36 1.02 β4[51] -0.00 0.01 -0.00 -0.02 0.02 99.58 1.00 β4[52] 0.02 0.04 0.00 -0.02 0.08 10.95 1.09 β4[53] 0.01 0.02 0.00 -0.03 0.03 26.58 1.06 β4[54] 0.01 0.02 0.00 -0.02 0.04 26.79 1.09 β4[55] -0.00 0.01 -0.00 -0.03 0.01 35.35 1.00 β4[56] -0.01 0.02 -0.00 -0.05 0.03 27.96 1.04 β4[57] -0.01 0.04 -0.00 -0.05 0.02 15.16 1.04 β4[58] -0.00 0.02 -0.00 -0.03 0.03 61.36 1.00 β4[59] -0.33 0.17 -0.37 -0.53 0.01 9.70 1.11 β4[60] 0.01 0.02 0.00 -0.02 0.04 31.31 1.00 β4[61] 0.00 0.02 0.00 -0.03 0.02 27.95 1.02 β4[62] 0.09 0.08 0.10 -0.01 0.21 18.35 1.00 β4[63] 0.01 0.02 0.00 -0.02 0.05 13.95 1.26 β4[64] 0.00 0.01 0.00 -0.02 0.01 75.35 1.01 β4[65] -0.01 0.02 -0.00 -0.03 0.02 39.66 1.02 β4[66] -0.01 0.02 -0.00 -0.04 0.01 23.60 1.04 β4[67] 0.01 0.02 0.00 -0.02 0.05 47.14 1.00 β4[68] -0.00 0.02 -0.00 -0.04 0.02 69.83 1.01 β4[69] 0.01 0.02 0.00 -0.02 0.05 10.40 1.17 β4[70] 0.00 0.02 -0.00 -0.03 0.03 36.04 1.02 β4[71] 0.00 0.03 -0.00 -0.03 0.08 17.76 1.07 β4[72] -0.02 0.03 -0.00 -0.08 0.01 36.85 1.00 β4[73] 0.00 0.02 0.00 -0.02 0.03 46.55 1.00 β4[74] 0.01 0.02 0.00 -0.02 0.03 23.48 1.00 β4[75] -0.00 0.01 0.00 -0.02 0.01 47.93 1.00 β4[76] 0.04 0.06 0.02 -0.01 0.15 12.97 1.02 β4[77] 0.00 0.02 -0.00 -0.02 0.03 48.50 1.05 β4[78] -0.00 0.01 -0.00 -0.02 0.02 68.15 1.00 β4[79] 0.00 0.01 0.00 -0.01 0.03 47.75 1.01 β4[80] 0.00 0.01 0.00 -0.01 0.02 59.33 1.00 β4[81] -0.02 0.03 -0.01 -0.06 0.04 16.90 1.01 β4[82] -0.00 0.02 -0.00 -0.05 0.03 38.19 1.09 β4[83] -0.01 0.03 -0.00 -0.05 0.02 51.16 1.00 β4[84] -0.00 0.01 0.00 -0.03 0.02 97.86 1.01 β4[85] -0.00 0.01 -0.00 -0.03 0.02 70.60 1.04 β4[86] -0.01 0.02 -0.00 -0.04 0.03 46.77 1.01 β4[87] -0.00 0.01 -0.00 -0.02 0.02 32.13 1.00 β4[88] 0.00 0.01 0.00 -0.02 0.01 68.99 1.02 β4[89] -0.00 0.01 -0.00 -0.01 0.02 72.40 1.01 β4[90] -0.00 0.02 -0.00 -0.03 0.03 94.51 1.00 β4[91] -0.16 0.06 -0.16 -0.26 -0.08 93.28 1.00 β4[92] 0.01 0.02 0.00 -0.02 0.05 24.06 1.06 β4[93] 0.00 0.01 0.00 -0.01 0.01 54.12 1.01 β4[94] 0.00 0.01 0.00 -0.01 0.03 90.29 1.00 β4[95] -0.01 0.02 -0.00 -0.03 0.04 37.40 1.00 β4[96] -0.00 0.03 0.00 -0.03 0.07 28.28 1.09 β4[97] -0.00 0.01 0.00 -0.01 0.01 175.85 1.00 β4[98] -0.00 0.02 -0.00 -0.02 0.03 41.22 1.00 β4[99] -0.00 0.02 -0.00 -0.01 0.02 25.51 1.08 β4[100] 0.00 0.02 -0.00 -0.02 0.03 103.16 1.01 β4[101] 0.00 0.02 0.00 -0.03 0.05 47.79 1.03 β4[102] 0.00 0.01 0.00 -0.01 0.02 43.98 1.00 β4[103] -0.04 0.06 -0.01 -0.16 0.02 13.56 1.00 β4[104] -0.00 0.02 0.00 -0.02 0.02 52.54 1.00 β4[105] 0.01 0.02 0.00 -0.01 0.04 53.66 1.07 β4[106] -0.01 0.02 -0.00 -0.04 0.01 42.04 1.02 β4[107] -0.00 0.03 -0.00 -0.05 0.03 23.47 1.02 β4[108] 0.01 0.03 0.00 -0.05 0.04 32.37 1.02 β4[109] 0.00 0.02 0.00 -0.02 0.06 29.06 1.00 β4[110] -0.00 0.01 -0.00 -0.03 0.01 39.95 1.01 β4[111] -0.00 0.02 -0.00 -0.03 0.01 78.44 1.00 β4[112] 0.01 0.04 0.00 -0.06 0.07 5.96 1.05 β4[113] 0.01 0.02 0.00 -0.01 0.04 32.27 1.06 β4[114] -0.00 0.01 -0.00 -0.02 0.01 25.01 1.06 β4[115] 0.00 0.02 0.00 -0.02 0.05 18.58 1.12 β4[116] 0.00 0.01 0.00 -0.01 0.02 53.40 1.02 β4[117] -0.01 0.02 -0.00 -0.03 0.02 79.98 1.01 β4[118] 0.00 0.02 0.00 -0.01 0.04 51.12 1.00 β4[119] -0.01 0.03 -0.00 -0.07 0.02 26.33 1.00 β4[120] -0.00 0.01 -0.00 -0.02 0.01 79.52 1.00 β4[121] 0.01 0.03 0.00 -0.03 0.04 12.83 1.08 β4[122] -0.00 0.02 0.00 -0.03 0.02 43.08 1.01 β4[123] -0.00 0.01 -0.00 -0.02 0.02 79.84 1.00 β4[124] -0.00 0.02 -0.00 -0.03 0.02 45.16 1.02 β4[125] 0.00 0.02 0.00 -0.01 0.03 26.29 1.00 β4[126] -0.00 0.01 -0.00 -0.02 0.02 98.48 1.00 β4[127] 0.00 0.01 -0.00 -0.02 0.02 49.18 1.00 β4[128] 0.00 0.01 0.00 -0.02 0.02 68.23 1.00 β4[129] -0.00 0.01 0.00 -0.02 0.02 77.39 1.00 β4[130] 0.01 0.03 0.00 -0.02 0.05 16.81 1.00 β4[131] 0.00 0.02 0.00 -0.02 0.04 66.20 1.00 β4[132] 0.01 0.02 0.00 -0.03 0.03 47.31 1.04 β4[133] -0.01 0.02 -0.00 -0.05 0.02 64.13 1.01 β4[134] -0.00 0.02 -0.00 -0.03 0.03 31.69 1.09 β4[135] 0.00 0.01 0.00 -0.01 0.01 79.31 1.01 β4[136] -0.00 0.02 -0.00 -0.04 0.05 30.01 1.02 β4[137] -0.00 0.02 -0.00 -0.02 0.04 21.91 1.02 β4[138] -0.01 0.02 -0.00 -0.05 0.03 42.61 1.01 β4[139] -0.00 0.01 -0.00 -0.02 0.02 38.28 1.02 β4[140] 0.01 0.03 0.00 -0.01 0.04 24.31 1.02 β4[141] -0.01 0.02 -0.00 -0.05 0.02 29.49 1.05 β4[142] 0.00 0.03 0.00 -0.02 0.05 43.56 1.00 β4[143] -0.00 0.02 0.00 -0.04 0.02 45.53 1.03 β4[144] 0.00 0.02 0.00 -0.02 0.04 55.54 1.00 β4[145] 0.01 0.02 0.00 -0.02 0.04 40.68 1.00 β4[146] 0.00 0.01 0.00 -0.01 0.03 102.83 1.00 β4[147] 0.00 0.01 0.00 -0.01 0.01 90.94 1.00 β4[148] -0.00 0.02 -0.00 -0.02 0.02 32.86 1.01 β4[149] 0.00 0.02 0.00 -0.01 0.02 71.78 1.00 β4[150] -0.03 0.04 -0.01 -0.12 0.01 21.28 1.00 β4[151] 0.00 0.01 0.00 -0.02 0.01 94.41 1.00 β4[152] 0.00 0.02 0.00 -0.02 0.04 26.84 1.00 β4[153] 0.00 0.02 0.00 -0.02 0.03 65.95 1.01 β4[154] -0.00 0.02 -0.00 -0.03 0.02 74.11 1.00 β4[155] -0.01 0.02 -0.00 -0.03 0.02 62.95 1.01 β4[156] -0.00 0.02 -0.00 -0.03 0.03 46.16 1.06 β4[157] 0.01 0.02 0.01 -0.01 0.03 20.86 1.08 β4[158] -0.00 0.02 -0.00 -0.03 0.03 31.34 1.07 β4[159] -0.00 0.01 -0.00 -0.03 0.01 58.32 1.02 β4[160] 0.00 0.02 0.00 -0.02 0.06 79.82 1.00 β4[161] 0.03 0.04 0.01 -0.02 0.11 18.25 1.04 β4[162] -0.00 0.02 -0.00 -0.03 0.03 36.22 1.03 β4[163] -0.00 0.02 0.00 -0.03 0.03 70.90 1.00 β4[164] -0.02 0.04 -0.00 -0.05 0.02 21.44 1.00 β4[165] 0.00 0.02 0.00 -0.02 0.04 66.39 1.00 β4[166] 0.00 0.01 0.00 -0.02 0.02 101.55 1.00 β4[167] 0.01 0.02 0.00 -0.02 0.05 38.57 1.03 β4[168] 0.00 0.02 0.00 -0.02 0.03 70.71 1.00 β4[169] -0.00 0.02 -0.00 -0.02 0.02 26.75 1.11 β4[170] -0.01 0.02 -0.00 -0.05 0.02 11.70 1.02 β4[171] 0.01 0.02 0.00 -0.02 0.03 59.71 1.03 β4[172] -0.00 0.02 -0.00 -0.04 0.03 17.97 1.02 β4[173] -0.00 0.02 -0.00 -0.05 0.02 28.60 1.01 β4[174] -0.01 0.02 -0.00 -0.06 0.02 13.64 1.00 β4[175] -0.00 0.02 -0.00 -0.03 0.02 59.05 1.00 β4[176] 0.00 0.01 0.00 -0.01 0.03 38.02 1.04 β4[177] -0.00 0.02 -0.00 -0.04 0.03 32.94 1.03 β4[178] -0.01 0.02 -0.00 -0.04 0.02 23.27 1.00 β4[179] 0.00 0.02 -0.00 -0.03 0.03 58.12 1.00 β4[180] -0.00 0.01 -0.00 -0.02 0.02 89.44 1.00 β4[181] 0.00 0.02 0.00 -0.02 0.03 87.16 1.00 β4[182] 0.00 0.01 0.00 -0.02 0.02 93.06 1.00 β4[183] 0.01 0.03 0.00 -0.02 0.04 18.05 1.03 β4[184] -0.09 0.07 -0.08 -0.21 0.01 31.72 1.01 β4[185] -1.14 0.07 -1.13 -1.26 -1.04 29.89 1.02 β4[186] -0.42 0.06 -0.43 -0.52 -0.34 51.77 1.00 β4[187] 0.00 0.02 0.00 -0.03 0.03 27.68 1.01 β4[188] 0.10 0.08 0.10 -0.02 0.21 23.94 1.00 β4[189] 0.07 0.07 0.05 -0.01 0.19 10.52 1.09 β4[190] 0.00 0.02 0.00 -0.02 0.04 75.37 1.00 β4[191] 0.01 0.03 0.00 -0.02 0.06 15.50 1.03 β4[192] -0.01 0.01 -0.00 -0.04 0.01 19.34 1.06 β4[193] -0.02 0.03 -0.00 -0.07 0.02 33.07 1.01 β4[194] -0.08 0.08 -0.07 -0.21 0.00 22.15 1.02 β4[195] 0.01 0.02 0.00 -0.03 0.03 16.45 1.08 β4[196] 0.00 0.02 0.00 -0.02 0.03 40.32 1.00 β4[197] 0.01 0.02 0.00 -0.02 0.05 17.36 1.03 β4[198] 0.00 0.03 0.00 -0.05 0.03 30.35 1.05 β4[199] -0.01 0.03 -0.00 -0.05 0.02 40.75 1.00 β4[200] -0.01 0.02 -0.00 -0.05 0.02 33.43 1.03 β4[201] 0.01 0.03 0.00 -0.02 0.05 11.27 1.07 β4[202] 0.01 0.02 0.00 -0.02 0.05 28.55 1.02 β4[203] 0.00 0.01 -0.00 -0.02 0.03 57.91 1.07 β4[204] 0.00 0.02 0.00 -0.03 0.03 42.38 1.00 β4[205] 0.00 0.02 0.00 -0.02 0.03 17.02 1.04 β4[206] -0.01 0.02 0.00 -0.04 0.03 18.88 1.00 β4[207] -0.00 0.02 -0.00 -0.03 0.02 17.04 1.10 β4[208] -0.00 0.02 0.00 -0.04 0.04 23.89 1.04 β4[209] 0.00 0.01 0.00 -0.02 0.02 36.05 1.00 β4[210] 0.01 0.02 0.00 -0.02 0.04 27.07 1.07 β4[211] 0.00 0.01 0.00 -0.02 0.02 32.55 1.02 β4[212] -0.01 0.02 -0.00 -0.04 0.01 33.47 1.02 β4[213] -0.00 0.02 0.00 -0.03 0.02 171.98 1.00 β4[214] 0.00 0.01 -0.00 -0.02 0.01 57.81 1.00 β4[215] 0.00 0.01 0.00 -0.02 0.02 105.64 1.00 β4[216] -0.00 0.02 -0.00 -0.02 0.03 32.01 1.01 β4[217] -0.01 0.02 -0.00 -0.05 0.01 37.16 1.08 β4[218] 0.00 0.02 0.00 -0.02 0.03 82.56 1.00 β4[219] 0.00 0.02 0.00 -0.02 0.03 77.24 1.00 β4[220] 0.02 0.05 0.00 -0.02 0.11 17.84 1.00 β4[221] -0.02 0.05 -0.00 -0.15 0.01 6.82 1.25 β4[222] -0.00 0.02 -0.00 -0.02 0.03 21.16 1.06 β4[223] -0.01 0.03 -0.00 -0.03 0.02 23.88 1.06 β4[224] 0.00 0.02 0.00 -0.03 0.04 86.96 1.00 β4[225] -0.03 0.05 -0.01 -0.09 0.02 38.74 1.03 β4[226] 0.00 0.01 0.00 -0.02 0.02 54.26 1.03 β4[227] 0.00 0.02 0.00 -0.02 0.04 44.34 1.00 β4[228] -0.00 0.01 -0.00 -0.01 0.02 97.05 1.00 β4[229] -0.00 0.02 0.00 -0.03 0.02 53.49 1.00 β4[230] -0.00 0.01 0.00 -0.01 0.02 98.82 1.02 β4[231] 0.00 0.02 -0.00 -0.03 0.03 147.08 1.00 β4[232] 0.00 0.01 0.00 -0.02 0.01 110.23 1.00 β4[233] -0.00 0.01 -0.00 -0.02 0.02 46.74 1.00 β4[234] -0.00 0.02 0.00 -0.02 0.03 68.05 1.00 β4[235] -0.00 0.01 -0.00 -0.03 0.01 61.14 1.01 β4[236] -0.00 0.01 0.00 -0.01 0.02 86.03 1.00 β4[237] 0.01 0.02 -0.00 -0.02 0.03 24.68 1.06 β4[238] 0.01 0.02 0.00 -0.02 0.05 47.98 1.01 β4[239] 0.01 0.02 0.00 -0.04 0.03 47.12 1.03 β4[240] 0.01 0.02 0.00 -0.02 0.05 23.53 1.01 β4[241] -0.00 0.01 -0.00 -0.02 0.02 77.31 1.00 β4[242] -0.01 0.02 -0.00 -0.05 0.02 12.00 1.01 β4[243] -0.00 0.01 -0.00 -0.03 0.01 35.94 1.02 β4[244] -0.14 0.09 -0.16 -0.24 0.00 17.17 1.11 β4[245] -0.00 0.02 -0.00 -0.02 0.03 188.81 1.00 β4[246] 0.00 0.01 0.00 -0.02 0.03 40.98 1.00 β4[247] 0.01 0.02 0.00 -0.01 0.04 90.48 1.01 β4[248] -0.01 0.02 -0.00 -0.03 0.02 56.40 1.00 β4[249] 0.00 0.02 0.00 -0.03 0.03 50.31 1.00 β4[250] 0.00 0.02 0.00 -0.02 0.03 16.52 1.03 β4[251] 0.01 0.03 0.00 -0.02 0.06 22.02 1.07 β4[252] 0.02 0.04 0.01 -0.03 0.08 13.43 1.05 β4[253] 0.01 0.03 0.00 -0.03 0.06 32.20 1.00 β4[254] -0.01 0.02 -0.00 -0.04 0.03 47.09 1.01 β4[255] -0.01 0.02 -0.00 -0.04 0.01 20.39 1.05 β4[256] -0.11 0.07 -0.11 -0.21 0.00 24.54 1.00 β4[257] -0.00 0.01 -0.00 -0.02 0.02 42.59 1.08 β4[258] -0.00 0.01 -0.00 -0.02 0.01 79.65 1.00 β4[259] 0.01 0.03 0.00 -0.03 0.06 6.88 1.33 β4[260] 0.00 0.02 0.00 -0.02 0.02 157.99 1.00 β4[261] -0.00 0.01 -0.00 -0.02 0.01 79.18 1.02 β4[262] 0.00 0.01 -0.00 -0.02 0.02 46.64 1.01 β4[263] 0.00 0.01 0.00 -0.02 0.03 56.37 1.00 β4[264] 0.00 0.02 0.00 -0.01 0.03 99.33 1.01 β4[265] 0.00 0.02 0.00 -0.02 0.04 111.74 1.00 β4[266] 0.00 0.02 0.00 -0.02 0.02 65.00 1.00 β4[267] 0.00 0.02 -0.00 -0.03 0.04 22.90 1.01 β4[268] 0.00 0.02 0.00 -0.01 0.04 34.94 1.01 β4[269] 0.00 0.02 -0.00 -0.02 0.03 38.51 1.02 β4[270] 0.01 0.02 0.00 -0.03 0.04 94.89 1.00 β4[271] -0.02 0.04 -0.01 -0.10 0.02 13.40 1.07 β4[272] -0.00 0.01 -0.00 -0.03 0.01 43.37 1.02 β4[273] 0.00 0.01 0.00 -0.02 0.03 30.64 1.05 β4[274] -0.00 0.02 -0.00 -0.03 0.02 32.41 1.04 β4[275] -0.00 0.02 -0.00 -0.03 0.02 80.96 1.00 β4[276] -0.01 0.02 -0.00 -0.04 0.02 34.92 1.00 β4[277] -0.00 0.02 -0.00 -0.04 0.01 36.13 1.00 β4[278] 0.00 0.02 -0.00 -0.02 0.03 31.66 1.02 β4[279] 0.00 0.02 -0.00 -0.02 0.04 67.09 1.07 β4[280] -0.01 0.02 -0.00 -0.03 0.02 68.98 1.00 β4[281] -0.01 0.01 -0.00 -0.03 0.01 19.73 1.01 β4[282] -0.00 0.01 -0.00 -0.02 0.01 198.61 1.01 β4[283] 0.02 0.03 0.01 -0.01 0.08 32.81 1.02 β4[284] -0.00 0.02 -0.00 -0.03 0.03 32.79 1.00 β4[285] 0.01 0.04 0.00 -0.02 0.06 15.87 1.06 β4[286] -0.02 0.03 -0.01 -0.07 0.01 37.74 1.00 β4[287] -0.00 0.01 -0.00 -0.01 0.02 45.82 1.00 β4[288] 0.00 0.02 0.00 -0.01 0.04 98.53 1.00 β4[289] -0.00 0.03 0.00 -0.05 0.04 38.22 1.09 β4[290] -0.00 0.02 -0.00 -0.03 0.03 51.46 1.00 β4[291] -0.01 0.02 -0.00 -0.04 0.01 36.95 1.02 β4[292] -0.01 0.02 -0.00 -0.07 0.02 37.91 1.04 β4[293] -0.01 0.02 -0.00 -0.04 0.01 11.81 1.15 β4[294] 0.00 0.02 -0.00 -0.02 0.03 55.34 1.02 β4[295] 0.00 0.01 0.00 -0.02 0.02 88.71 1.00 β4[296] -0.00 0.02 -0.00 -0.03 0.03 50.19 1.00 β4[297] -0.00 0.01 -0.00 -0.02 0.01 144.61 1.00 β4[298] 0.01 0.03 0.00 -0.06 0.04 34.19 1.04 β4[299] 0.02 0.04 0.01 -0.01 0.11 7.87 1.30 β4[300] 0.01 0.03 -0.00 -0.02 0.04 30.42 1.00 β4[301] -0.00 0.02 0.00 -0.04 0.03 39.33 1.02 β4[302] 0.00 0.02 0.00 -0.02 0.03 38.17 1.00 β4[303] 0.00 0.02 0.00 -0.02 0.03 39.07 1.03 β4[304] -0.05 0.07 -0.01 -0.19 0.01 6.88 1.27 β4[305] -0.01 0.02 -0.00 -0.05 0.03 10.19 1.00 β4[306] -0.00 0.01 -0.00 -0.02 0.02 55.61 1.00 β4[307] 0.01 0.03 0.00 -0.03 0.05 32.10 1.08 β4[308] 0.01 0.02 0.00 -0.02 0.03 41.32 1.03 β4[309] -0.00 0.01 -0.00 -0.02 0.01 136.46 1.01 β4[310] -0.00 0.02 -0.00 -0.03 0.04 31.99 1.00 β4[311] -0.00 0.03 -0.00 -0.05 0.02 36.33 1.01 β4[312] 0.01 0.03 0.01 -0.01 0.07 20.83 1.04 β4[313] 0.00 0.02 0.00 -0.03 0.03 54.97 1.05 β4[314] 0.00 0.01 0.00 -0.02 0.03 20.89 1.03 β4[315] 0.00 0.02 0.00 -0.02 0.04 52.74 1.03 β4[316] -0.01 0.02 -0.00 -0.04 0.02 40.15 1.00 β4[317] -0.02 0.03 -0.01 -0.08 0.03 29.63 1.01 β4[318] 0.01 0.02 0.00 -0.02 0.06 12.37 1.12 β4[319] 0.00 0.02 0.00 -0.03 0.04 38.96 1.00 β4[320] 0.00 0.01 0.00 -0.02 0.03 28.92 1.09 β4[321] -0.00 0.02 -0.00 -0.03 0.03 76.35 1.00 β4[322] -0.00 0.01 -0.00 -0.02 0.01 120.68 1.00 β4[323] -0.00 0.03 0.00 -0.03 0.03 19.37 1.05 β4[324] 0.00 0.01 0.00 -0.01 0.03 49.73 1.10 β4[325] 0.00 0.02 0.00 -0.02 0.03 73.04 1.01 β4[326] -0.02 0.03 -0.00 -0.07 0.03 60.02 1.00 β4[327] 0.00 0.01 0.00 -0.02 0.02 103.51 1.02 β4[328] 0.00 0.01 0.00 -0.01 0.02 49.26 1.01 β4[329] 0.01 0.02 0.00 -0.02 0.05 17.93 1.01 β4[330] -0.01 0.03 -0.00 -0.07 0.02 29.07 1.00 β4[331] -0.00 0.02 -0.00 -0.04 0.02 41.68 1.04 β4[332] -0.00 0.02 -0.00 -0.05 0.03 23.89 1.06 β4[333] -0.00 0.02 0.00 -0.03 0.02 15.84 1.04 β4[334] 0.00 0.03 0.00 -0.07 0.03 25.44 1.04 β4[335] 0.02 0.03 0.01 -0.02 0.07 8.90 1.20 β4[336] -0.00 0.02 -0.00 -0.03 0.01 31.24 1.00 β4[337] -0.01 0.02 -0.00 -0.05 0.01 15.82 1.03 β4[338] 0.00 0.02 0.00 -0.03 0.02 30.25 1.05 β4[339] 0.00 0.02 0.00 -0.03 0.03 30.31 1.02 β4[340] -0.01 0.02 -0.00 -0.04 0.02 31.81 1.02 β4[341] -0.00 0.02 0.00 -0.02 0.03 30.34 1.06 β4[342] 0.01 0.02 0.01 -0.02 0.04 24.16 1.06 β4[343] -0.01 0.02 -0.00 -0.05 0.03 15.46 1.13 β4[344] 0.00 0.01 0.00 -0.02 0.03 45.57 1.01 β4[345] -0.02 0.02 -0.01 -0.06 0.01 7.03 1.42 β4[346] 0.00 0.02 0.00 -0.02 0.04 77.00 1.00 β4[347] -0.09 0.07 -0.10 -0.18 0.01 14.05 1.04 β4[348] -0.01 0.02 -0.00 -0.03 0.02 25.19 1.04 β4[349] 0.01 0.03 0.00 -0.02 0.05 94.13 1.00 β4[350] 0.00 0.01 0.00 -0.02 0.02 110.57 1.00 β4[351] 0.00 0.02 0.00 -0.02 0.04 111.13 1.00 β4[352] 0.01 0.03 0.00 -0.02 0.07 45.63 1.01 β4[353] 0.02 0.04 0.00 -0.02 0.06 23.78 1.08 β4[354] -0.01 0.02 -0.00 -0.05 0.01 60.53 1.12 β4[355] -0.20 0.06 -0.21 -0.28 -0.10 10.89 1.00 β4[356] -0.57 0.06 -0.57 -0.66 -0.46 33.25 1.02 β4[357] -0.95 0.06 -0.95 -1.06 -0.86 49.64 1.00 β4[358] -1.51 0.07 -1.50 -1.63 -1.40 31.92 1.02 β4[359] -1.89 0.06 -1.89 -1.99 -1.79 33.05 1.03 β4[360] -0.91 0.08 -0.92 -1.03 -0.78 28.21 1.00 β4[361] 0.00 0.02 0.00 -0.03 0.04 89.17 1.00 β4[362] 0.34 0.08 0.34 0.18 0.44 48.29 1.00 β4[363] 0.44 0.06 0.44 0.36 0.54 35.00 1.02 β4[364] 0.54 0.07 0.54 0.41 0.63 25.78 1.03 β4[365] 0.03 0.04 0.01 -0.02 0.10 23.55 1.00 β5_labour -0.90 0.04 -0.90 -0.97 -0.84 72.12 1.02 β5_memorial -1.20 0.06 -1.20 -1.29 -1.10 86.17 0.99 β5_thanksgiving -1.28 0.04 -1.28 -1.35 -1.23 43.47 1.08 β_week[0] 0.63 0.09 0.63 0.48 0.77 8.28 1.04 β_week[1] 0.38 0.05 0.38 0.29 0.47 8.42 1.05 β_week[2] 0.33 0.05 0.33 0.26 0.41 7.86 1.06 β_week[3] 0.46 0.06 0.46 0.35 0.56 8.70 1.04 β_week[4] -0.86 0.12 -0.87 -1.06 -0.66 8.19 1.05 β_week[5] -1.22 0.17 -1.22 -1.50 -0.95 8.27 1.05 λ[0] 224.83 268.36 138.76 29.46 525.88 15.70 1.06 λ[1] 84.63 62.76 67.27 19.57 176.80 52.25 1.01 λ[2] 6.96 8.59 4.31 0.16 15.40 36.53 1.01 λ[3] 1.97 2.86 1.14 0.07 4.61 30.63 1.00 λ[4] 1.84 2.30 1.00 0.06 4.76 32.32 1.12 λ[5] 1.52 1.93 1.01 0.07 3.10 55.86 1.03 λ[6] 2.08 2.70 1.26 0.02 4.72 60.69 1.01 λ[7] 2.10 5.36 0.79 0.01 4.36 91.57 1.02 λ[8] 4.51 5.48 2.36 0.12 11.20 22.79 1.00 λ[9] 1.73 2.93 0.73 0.01 3.64 17.92 1.00 λ[10] 6.25 7.87 2.77 0.07 16.62 19.73 1.00 λ[11] 1.55 1.76 0.94 0.03 4.04 15.91 1.00 λ[12] 1.58 2.33 0.93 0.00 4.00 23.76 1.02 λ[13] 2.50 3.27 1.06 0.01 6.11 46.35 1.01 λ[14] 1.66 1.86 0.93 0.09 3.61 34.76 1.01 λ[15] 1.21 2.02 0.76 0.01 2.35 42.12 1.02 λ[16] 1.08 1.58 0.53 0.02 2.96 80.64 1.00 λ[17] 1.82 3.68 0.61 0.01 3.40 34.30 1.03 λ[18] 3.93 7.08 1.84 0.02 8.13 51.05 1.00 λ[19] 1.56 2.15 0.80 0.01 3.68 41.20 1.00 λ[20] 1.65 2.00 0.93 0.01 4.33 36.43 1.01 λ[21] 1.49 2.04 0.61 0.01 3.94 22.76 1.00 λ[22] 1.79 2.42 0.87 0.02 4.38 13.56 1.15 λ[23] 0.96 1.32 0.68 0.01 2.00 69.11 1.00 λ[24] 1.71 2.36 0.77 0.01 4.35 27.65 1.00 λ[25] 1.21 1.77 0.59 0.01 3.65 9.86 1.18 λ[26] 1.93 1.89 1.41 0.09 3.64 32.85 1.00 λ[27] 3.00 5.37 1.40 0.00 5.75 17.92 1.07 λ[28] 1.65 1.77 1.11 0.02 3.78 13.41 1.13 λ[29] 2.10 3.49 1.12 0.01 4.85 44.42 1.06 λ[30] 1.64 1.98 1.07 0.09 3.38 97.62 1.00 λ[31] 2.10 3.28 1.04 0.00 4.86 30.27 1.00 λ[32] 4.33 7.06 1.61 0.01 11.82 14.37 1.20 λ[33] 1.45 2.59 0.82 0.01 2.97 43.34 1.08 λ[34] 1.82 2.23 1.06 0.01 4.59 16.47 1.07 λ[35] 1.55 2.56 0.88 0.01 2.90 64.45 1.00 λ[36] 1.07 1.65 0.59 0.01 2.43 19.12 1.05 λ[37] 2.11 4.71 0.52 0.02 5.41 31.34 1.03 λ[38] 1.43 2.08 0.81 0.02 3.93 57.94 1.05 λ[39] 1.30 1.24 1.02 0.02 2.81 47.58 1.01 λ[40] 3.00 7.66 0.65 0.06 8.34 34.49 1.07 λ[41] 1.73 3.88 0.91 0.11 3.22 68.44 1.01 λ[42] 4.59 7.42 1.69 0.30 11.82 11.20 1.05 λ[43] 7.94 10.83 4.75 0.04 18.60 46.05 1.01 λ[44] 136.90 867.49 23.88 6.23 90.66 82.31 1.01 λ[45] 3.23 3.94 1.57 0.08 8.02 11.04 1.18 λ[46] 2.45 4.46 0.87 0.03 5.20 24.27 1.01 λ[47] 1.57 1.77 0.95 0.02 4.31 48.60 1.02 λ[48] 3.25 9.53 1.38 0.13 4.98 83.82 1.00 λ[49] 1.77 2.54 0.91 0.03 5.15 21.63 1.04 λ[50] 1.14 1.64 0.67 0.03 2.50 47.19 1.00 λ[51] 0.94 0.95 0.63 0.01 2.36 42.88 1.11 λ[52] 2.33 3.95 1.18 0.16 4.78 21.77 1.09 λ[53] 1.77 3.76 0.63 0.02 5.36 22.33 1.10 λ[54] 2.84 7.93 0.95 0.02 5.31 30.78 1.04 λ[55] 1.19 1.17 1.00 0.01 2.70 26.56 1.03 λ[56] 1.50 2.16 0.71 0.02 4.04 29.37 1.04 λ[57] 1.33 2.32 0.47 0.00 3.32 19.49 1.01 λ[58] 1.91 6.60 0.57 0.01 4.19 134.86 1.00 λ[59] 29.11 26.78 21.10 0.10 73.91 28.44 1.00 λ[60] 1.74 1.98 1.01 0.01 4.47 50.03 1.02 λ[61] 1.32 1.58 0.81 0.03 2.68 65.56 0.99 λ[62] 11.00 14.84 7.53 0.09 26.58 27.17 1.02 λ[63] 2.29 2.80 1.38 0.06 5.96 47.53 1.03 λ[64] 1.08 1.55 0.63 0.04 2.66 93.00 1.00 λ[65] 1.78 3.04 0.85 0.01 4.52 41.00 1.06 λ[66] 1.55 1.89 1.05 0.01 3.50 46.35 1.00 λ[67] 2.42 6.19 1.09 0.04 3.80 43.64 1.03 λ[68] 1.69 2.48 0.95 0.05 3.70 84.43 1.03 λ[69] 2.00 3.78 0.73 0.06 4.15 15.30 1.05 λ[70] 1.52 2.10 0.85 0.01 3.36 44.21 1.01 λ[71] 3.32 5.35 1.34 0.02 9.37 32.75 1.00 λ[72] 2.28 3.37 0.97 0.04 6.66 39.01 1.00 λ[73] 1.75 2.03 1.11 0.03 3.29 24.68 1.00 λ[74] 1.48 2.58 0.74 0.04 3.07 48.59 1.00 λ[75] 0.84 0.91 0.61 0.00 2.05 26.55 1.06 λ[76] 4.12 5.54 2.09 0.02 10.77 18.86 1.02 λ[77] 1.30 1.70 0.75 0.07 2.59 33.80 1.00 λ[78] 1.16 1.67 0.66 0.00 2.67 18.09 1.04 λ[79] 1.09 0.97 0.89 0.00 2.24 48.62 1.00 λ[80] 0.94 1.38 0.43 0.00 2.31 66.26 1.00 λ[81] 3.38 8.08 1.22 0.09 6.52 48.80 1.05 λ[82] 1.79 2.22 0.91 0.03 4.55 45.28 1.03 λ[83] 2.14 3.42 1.00 0.03 6.16 45.04 1.01 λ[84] 1.19 2.02 0.57 0.01 3.00 16.48 1.07 λ[85] 1.46 1.75 0.88 0.02 3.48 68.86 1.07 λ[86] 1.66 1.62 1.12 0.02 3.68 61.53 1.01 λ[87] 1.13 1.28 0.85 0.00 2.23 80.50 1.00 λ[88] 3.68 24.83 0.64 0.02 2.16 82.45 1.01 λ[89] 0.92 1.10 0.50 0.01 2.42 72.05 1.01 λ[90] 1.21 1.15 0.95 0.01 2.89 10.06 1.25 λ[91] 35.07 62.14 13.14 1.18 99.82 35.42 1.01 λ[92] 1.98 2.14 1.12 0.02 4.96 51.92 1.06 λ[93] 0.65 0.85 0.33 0.00 1.56 61.92 1.00 λ[94] 1.31 1.72 0.87 0.02 2.63 19.86 1.07 λ[95] 2.03 2.94 1.25 0.04 4.24 91.69 1.00 λ[96] 2.25 3.05 1.02 0.02 5.95 35.51 0.99 λ[97] 0.82 1.11 0.46 0.06 1.96 27.75 1.11 λ[98] 7.01 50.47 0.52 0.01 2.82 69.76 1.01 λ[99] 1.06 1.95 0.57 0.01 2.33 38.82 1.02 λ[100] 1.22 1.27 0.78 0.03 3.14 67.67 1.02 λ[101] 1.54 1.73 0.99 0.02 3.81 39.07 1.01 λ[102] 1.07 1.46 0.68 0.01 2.53 37.17 1.04 λ[103] 9.27 16.61 2.68 0.04 31.63 10.42 1.03 λ[104] 1.17 1.76 0.67 0.01 2.31 36.69 1.01 λ[105] 1.97 3.75 0.70 0.01 4.63 34.54 1.12 λ[106] 1.47 4.03 0.56 0.01 3.21 85.54 1.00 λ[107] 1.67 2.09 0.96 0.00 4.92 19.84 1.03 λ[108] 2.06 3.33 0.93 0.04 4.51 23.25 1.13 λ[109] 1.68 2.76 0.89 0.07 3.19 35.23 1.01 λ[110] 1.10 1.41 0.67 0.01 2.34 42.33 1.00 λ[111] 1.01 1.14 0.65 0.00 2.36 14.76 1.00 λ[112] 3.84 7.31 1.81 0.01 8.69 29.92 1.08 λ[113] 2.07 5.91 0.68 0.00 4.29 56.41 1.03 λ[114] 1.19 2.04 0.64 0.02 2.36 36.36 1.01 λ[115] 2.43 3.78 1.04 0.06 6.57 29.97 1.00 λ[116] 0.79 0.90 0.48 0.00 2.12 26.01 1.00 λ[117] 1.62 2.95 0.73 0.00 3.56 51.25 1.00 λ[118] 1.15 1.58 0.74 0.05 2.28 59.70 1.02 λ[119] 1.73 2.01 0.98 0.01 4.34 27.98 1.00 λ[120] 1.62 5.36 0.69 0.01 2.46 62.54 1.01 λ[121] 2.48 5.05 0.95 0.00 5.46 19.80 1.03 λ[122] 1.32 2.31 0.67 0.01 3.21 41.30 1.04 λ[123] 1.07 1.10 0.85 0.01 2.33 19.68 1.03 λ[124] 1.87 3.43 0.93 0.01 3.75 27.39 1.04 λ[125] 1.27 1.32 0.84 0.01 3.04 20.23 1.02 λ[126] 0.94 1.24 0.58 0.00 2.47 29.40 1.06 λ[127] 1.19 1.88 0.53 0.00 2.93 42.69 1.02 λ[128] 1.41 1.76 0.90 0.07 2.93 48.74 1.00 λ[129] 1.26 1.90 0.62 0.03 2.51 55.30 1.00 λ[130] 3.66 8.89 1.21 0.03 5.55 18.40 1.03 λ[131] 1.95 3.59 0.87 0.03 5.24 38.00 1.07 λ[132] 2.06 4.55 1.06 0.00 3.37 33.27 1.05 λ[133] 1.94 1.66 1.49 0.10 3.98 36.44 1.02 λ[134] 1.67 2.22 0.88 0.02 5.35 15.61 1.00 λ[135] 0.38 0.45 0.31 0.00 0.88 14.62 1.05 λ[136] 2.44 3.37 1.02 0.07 6.81 23.56 1.00 λ[137] 1.60 2.00 0.68 0.03 4.79 15.42 1.20 λ[138] 1.98 2.09 1.32 0.04 4.64 48.68 1.09 λ[139] 1.54 2.27 0.74 0.00 4.10 64.34 1.00 λ[140] 1.80 2.22 1.32 0.01 3.93 27.33 1.01 λ[141] 1.90 2.97 0.88 0.01 5.08 40.78 1.05 λ[142] 1.89 2.88 0.64 0.02 5.99 27.81 1.02 λ[143] 2.58 5.48 0.93 0.01 5.55 32.41 1.01 λ[144] 2.43 2.83 1.47 0.18 4.81 43.76 1.01 λ[145] 1.60 1.69 1.11 0.13 3.73 54.37 1.00 λ[146] 1.31 1.23 0.90 0.10 2.81 17.01 1.03 λ[147] 1.10 1.62 0.64 0.01 1.96 19.08 1.19 λ[148] 1.44 1.66 0.90 0.03 3.29 26.49 1.02 λ[149] 0.98 1.69 0.48 0.01 2.17 50.01 1.03 λ[150] 3.03 3.62 1.48 0.10 8.72 19.74 1.01 λ[151] 1.25 3.21 0.51 0.01 2.12 39.79 1.01 λ[152] 2.12 4.83 0.76 0.01 4.49 58.64 1.01 λ[153] 1.59 2.00 1.00 0.01 3.74 53.78 1.01 λ[154] 1.53 1.60 0.99 0.02 3.71 37.85 1.00 λ[155] 1.93 3.70 1.06 0.01 4.04 113.92 1.00 λ[156] 1.58 1.74 1.19 0.03 3.34 20.36 1.00 λ[157] 1.39 1.29 1.06 0.01 3.09 27.45 1.06 λ[158] 1.60 2.06 1.20 0.05 3.33 41.78 1.06 λ[159] 1.47 2.13 0.78 0.01 3.06 107.31 1.00 λ[160] 3.21 6.08 1.16 0.01 6.56 14.89 1.08 λ[161] 3.57 3.35 2.27 0.12 8.69 16.97 1.04 λ[162] 1.61 2.04 0.76 0.01 3.98 18.29 1.05 λ[163] 1.19 1.24 0.87 0.00 2.74 49.82 1.00 λ[164] 2.03 3.12 1.09 0.05 4.26 22.99 1.00 λ[165] 1.83 1.83 1.31 0.03 4.36 55.53 1.02 λ[166] 2.56 10.20 0.99 0.01 3.35 104.67 1.01 λ[167] 1.65 1.86 1.16 0.02 3.63 29.50 1.06 λ[168] 1.38 1.17 1.06 0.15 2.86 32.81 1.00 λ[169] 1.80 2.73 0.83 0.04 4.36 34.92 1.00 λ[170] 2.76 3.61 1.33 0.01 7.36 41.23 1.04 λ[171] 1.46 1.65 0.86 0.08 3.32 34.56 1.00 λ[172] 2.17 2.70 1.16 0.03 5.79 43.47 1.07 λ[173] 1.48 1.97 0.77 0.04 4.25 25.02 1.00 λ[174] 2.25 3.56 0.89 0.06 6.67 9.44 1.02 λ[175] 1.47 2.06 0.79 0.03 3.41 32.73 1.02 λ[176] 8.01 69.36 0.65 0.01 3.61 110.07 1.00 λ[177] 2.13 8.03 0.84 0.01 3.24 82.10 1.01 λ[178] 1.72 3.03 0.69 0.03 4.27 32.85 1.01 λ[179] 2.06 3.54 0.89 0.00 5.45 57.63 1.00 λ[180] 1.28 1.37 0.89 0.03 2.58 53.71 1.01 λ[181] 1.39 1.48 0.91 0.06 3.54 68.44 1.04 λ[182] 1.83 6.87 0.68 0.01 3.03 107.78 1.00 λ[183] 1.89 3.08 1.01 0.01 4.58 32.44 1.01 λ[184] 10.54 14.31 6.27 0.01 21.99 23.99 1.00 λ[185] 270.06 519.27 121.26 20.58 510.91 56.49 1.00 λ[186] 167.55 354.52 32.75 6.67 680.53 9.66 1.03 λ[187] 1.49 1.72 0.91 0.02 3.55 32.04 1.01 λ[188] 9.61 11.10 6.91 0.10 20.04 40.39 1.00 λ[189] 6.99 9.67 3.66 0.06 16.71 19.51 1.13 λ[190] 1.40 2.26 0.64 0.01 3.24 52.60 1.00 λ[191] 1.76 2.33 0.88 0.05 4.43 31.65 1.02 λ[192] 1.29 1.11 0.93 0.01 2.67 31.16 1.04 λ[193] 2.52 3.50 1.24 0.01 6.11 38.32 1.00 λ[194] 9.01 12.80 4.84 0.06 23.28 44.57 1.00 λ[195] 1.67 2.24 0.76 0.04 4.30 15.04 1.05 λ[196] 1.59 2.04 0.78 0.08 3.57 53.23 1.01 λ[197] 1.64 2.22 0.87 0.00 4.29 17.24 1.08 λ[198] 2.38 5.78 1.33 0.00 5.04 45.98 1.02 λ[199] 2.17 3.20 1.22 0.04 4.40 37.42 1.00 λ[200] 1.87 2.52 1.13 0.03 4.27 41.75 1.00 λ[201] 1.95 3.89 0.91 0.02 4.48 27.13 1.02 λ[202] 2.59 6.63 0.86 0.06 5.16 26.37 1.03 λ[203] 1.52 1.57 1.15 0.10 3.52 64.49 1.00 λ[204] 1.92 2.88 1.10 0.02 4.32 51.11 1.00 λ[205] 1.46 1.89 0.79 0.00 3.39 12.96 1.22 λ[206] 1.69 1.74 0.78 0.01 4.40 11.40 1.21 λ[207] 1.50 1.49 1.09 0.07 2.84 29.12 1.11 λ[208] 2.24 3.65 1.07 0.01 5.80 33.51 1.02 λ[209] 1.10 1.78 0.53 0.03 2.48 34.56 1.01 λ[210] 1.84 2.28 0.99 0.01 6.21 33.05 1.05 λ[211] 1.48 1.77 1.03 0.03 3.05 65.09 1.01 λ[212] 1.55 1.50 1.22 0.05 2.99 28.44 1.14 λ[213] 1.05 1.27 0.64 0.02 2.44 32.96 1.03 λ[214] 0.89 0.96 0.57 0.02 2.15 31.33 1.03 λ[215] 2.45 8.78 0.77 0.02 2.32 37.26 1.02 λ[216] 1.41 1.60 0.80 0.03 3.29 49.58 1.04 λ[217] 1.71 2.70 0.84 0.06 4.55 30.43 1.05 λ[218] 1.39 1.31 0.95 0.01 3.05 28.74 1.02 λ[219] 1.42 2.49 0.77 0.04 3.64 59.46 1.02 λ[220] 2.34 3.62 0.90 0.02 6.43 20.12 1.00 λ[221] 6.63 22.29 0.89 0.00 12.98 19.60 1.06 λ[222] 1.55 1.60 1.16 0.05 4.16 24.51 1.00 λ[223] 1.64 2.00 1.02 0.04 3.71 14.80 1.07 λ[224] 2.13 4.27 1.19 0.02 4.28 90.87 1.00 λ[225] 3.47 4.29 1.79 0.03 8.97 43.70 1.00 λ[226] 0.80 1.29 0.19 0.00 2.14 5.06 1.30 λ[227] 1.73 2.37 1.12 0.07 3.20 23.65 1.02 λ[228] 0.87 1.03 0.49 0.04 2.19 29.15 1.07 λ[229] 0.94 1.24 0.55 0.06 2.17 30.98 1.00 λ[230] 0.81 1.11 0.39 0.01 1.71 54.95 1.01 λ[231] 1.57 1.41 1.20 0.12 3.88 68.12 1.00 λ[232] 1.06 2.02 0.52 0.01 2.26 84.34 1.00 λ[233] 1.39 1.57 0.84 0.00 3.16 36.32 1.12 λ[234] 2.02 3.61 0.86 0.06 4.37 24.05 1.02 λ[235] 1.60 1.66 1.16 0.11 3.79 61.17 1.00 λ[236] 1.01 1.67 0.62 0.01 2.16 78.56 1.00 λ[237] 1.50 1.73 0.92 0.01 3.86 18.96 1.17 λ[238] 1.54 1.54 1.27 0.05 3.12 47.19 1.00 λ[239] 2.42 3.97 1.40 0.06 4.49 65.85 1.00 λ[240] 2.22 4.89 1.18 0.03 4.70 65.71 1.00 λ[241] 1.17 1.10 0.82 0.00 2.70 44.68 1.03 λ[242] 1.86 3.06 0.57 0.01 6.16 13.60 1.00 λ[243] 1.20 1.53 0.74 0.01 3.02 68.62 1.01 λ[244] 28.51 87.52 10.64 0.16 44.75 78.22 0.99 λ[245] 1.30 1.31 0.81 0.01 3.13 24.19 1.01 λ[246] 1.39 1.49 1.00 0.03 3.50 68.19 1.00 λ[247] 2.07 2.92 0.99 0.03 5.40 49.69 1.00 λ[248] 1.86 3.33 1.13 0.01 3.07 63.72 1.01 λ[249] 1.90 2.80 0.75 0.08 4.47 23.93 1.12 λ[250] 1.22 1.83 0.72 0.01 2.61 40.09 1.00 λ[251] 1.99 2.37 1.00 0.03 4.60 46.34 1.03 λ[252] 3.37 4.42 1.47 0.01 8.46 18.15 1.05 λ[253] 2.68 3.75 1.72 0.06 5.37 31.00 1.00 λ[254] 1.69 2.09 1.13 0.00 4.26 38.47 1.00 λ[255] 1.25 1.54 0.69 0.01 3.25 17.87 1.09 λ[256] 11.63 17.41 8.40 0.26 21.71 46.94 1.00 λ[257] 1.36 1.39 0.98 0.04 2.89 53.34 1.02 λ[258] 1.11 1.22 0.60 0.02 2.74 19.80 1.10 λ[259] 2.22 3.48 0.96 0.04 5.57 12.66 1.09 λ[260] 1.17 1.40 0.71 0.01 2.85 42.85 1.00 λ[261] 1.19 1.70 0.71 0.04 2.08 44.38 1.04 λ[262] 1.36 1.41 0.82 0.01 3.33 60.32 1.01 λ[263] 1.18 1.65 0.65 0.03 2.83 42.66 1.00 λ[264] 0.88 1.04 0.48 0.01 2.44 27.45 1.03 λ[265] 1.33 1.99 0.76 0.03 2.83 58.08 1.00 λ[266] 1.05 1.23 0.69 0.01 2.42 59.52 1.02 λ[267] 4.04 9.85 1.15 0.05 6.44 22.42 1.04 λ[268] 1.48 2.10 0.81 0.01 3.28 62.00 1.00 λ[269] 1.90 2.84 0.95 0.01 4.19 41.11 1.02 λ[270] 2.11 3.64 0.89 0.08 4.97 40.16 1.01 λ[271] 2.73 3.12 1.65 0.08 5.89 39.99 1.00 λ[272] 1.06 1.06 0.72 0.01 2.47 28.90 1.00 λ[273] 1.28 1.57 0.89 0.01 2.83 49.84 1.02 λ[274] 1.39 1.25 1.03 0.04 3.08 41.88 1.01 λ[275] 1.74 2.23 1.02 0.05 3.81 54.15 1.00 λ[276] 2.40 3.50 1.17 0.17 5.82 15.50 1.05 λ[277] 1.42 1.59 1.03 0.04 2.79 75.12 1.00 λ[278] 1.66 2.77 0.61 0.03 4.47 17.59 1.06 λ[279] 1.57 1.96 0.89 0.01 3.86 56.51 1.07 λ[280] 1.39 1.26 0.95 0.03 3.40 55.55 1.00 λ[281] 1.40 1.76 0.73 0.11 3.28 30.39 1.00 λ[282] 1.03 1.35 0.61 0.03 2.61 49.36 1.00 λ[283] 2.70 5.12 1.06 0.05 4.93 25.72 1.06 λ[284] 1.86 2.99 1.00 0.04 3.54 32.68 1.03 λ[285] 4.10 10.60 0.77 0.01 8.60 16.33 1.08 λ[286] 4.09 8.47 1.90 0.02 9.19 58.58 1.00 λ[287] 1.17 1.52 0.63 0.02 2.83 36.68 1.17 λ[288] 1.17 1.42 0.66 0.02 2.67 41.95 1.07 λ[289] 2.97 4.65 1.55 0.03 7.75 35.26 1.06 λ[290] 1.59 3.91 0.84 0.01 3.01 57.32 1.02 λ[291] 2.31 8.70 0.82 0.07 4.04 94.76 1.01 λ[292] 1.71 2.28 0.95 0.03 4.43 24.37 1.01 λ[293] 1.43 1.94 0.72 0.12 3.55 20.62 1.01 λ[294] 1.37 1.47 0.90 0.04 2.64 48.04 1.00 λ[295] 1.33 1.89 0.78 0.01 3.06 31.88 1.06 λ[296] 1.53 2.44 0.63 0.01 3.89 21.00 1.02 λ[297] 0.77 0.89 0.41 0.01 1.81 29.16 1.04 λ[298] 2.49 3.14 1.29 0.02 6.71 15.21 1.02 λ[299] 2.53 3.13 1.30 0.09 6.74 12.08 1.16 λ[300] 1.87 1.68 1.13 0.12 3.86 44.37 1.00 λ[301] 2.24 3.34 1.17 0.01 4.97 83.30 1.02 λ[302] 1.31 2.70 0.45 0.01 3.07 53.11 1.00 λ[303] 2.26 3.72 1.33 0.06 4.14 39.14 1.03 λ[304] 4.67 6.72 1.83 0.10 16.63 9.34 1.23 λ[305] 3.43 5.97 1.28 0.08 8.85 19.34 1.04 λ[306] 0.87 1.06 0.45 0.01 2.78 23.15 1.04 λ[307] 1.96 3.16 0.88 0.00 5.09 27.74 1.16 λ[308] 2.28 4.49 1.08 0.01 4.89 45.31 1.06 λ[309] 0.98 1.31 0.68 0.03 2.10 56.97 1.00 λ[310] 2.11 3.24 1.02 0.04 5.14 42.37 1.00 λ[311] 1.97 3.00 1.06 0.07 4.07 51.62 1.00 λ[312] 2.24 2.88 1.02 0.12 5.73 49.51 1.00 λ[313] 1.83 3.97 0.88 0.04 3.21 48.57 1.03 λ[314] 1.13 1.34 0.72 0.00 2.67 24.51 1.11 λ[315] 1.67 1.67 1.13 0.03 3.61 88.68 1.02 λ[316] 1.76 2.66 0.79 0.01 4.14 32.04 1.01 λ[317] 2.59 2.70 1.39 0.10 7.07 19.15 1.00 λ[318] 1.98 2.57 0.86 0.02 7.24 12.20 1.05 λ[319] 2.44 5.33 1.01 0.04 4.30 65.13 1.00 λ[320] 1.20 1.01 0.95 0.02 2.52 42.48 1.11 λ[321] 1.82 3.02 0.94 0.01 4.00 58.03 1.00 λ[322] 0.88 1.10 0.67 0.00 1.71 44.38 1.03 λ[323] 1.90 2.69 1.04 0.00 4.79 37.09 1.02 λ[324] 1.82 5.68 0.52 0.02 2.38 30.17 1.05 λ[325] 2.03 3.43 1.06 0.04 4.63 44.39 1.02 λ[326] 2.24 2.94 0.97 0.05 5.96 28.43 1.05 λ[327] 1.52 2.03 0.81 0.02 3.86 53.95 1.00 λ[328] 1.23 1.19 0.86 0.02 2.52 47.90 1.02 λ[329] 2.48 3.46 1.27 0.01 6.16 26.53 1.03 λ[330] 1.79 2.04 0.98 0.06 4.23 27.76 1.01 λ[331] 1.53 2.37 0.80 0.00 3.94 47.91 1.01 λ[332] 1.56 2.04 0.97 0.01 3.85 31.46 1.00 λ[333] 2.50 4.92 0.97 0.01 6.07 39.59 1.00 λ[334] 7.24 37.84 0.84 0.06 4.83 45.15 1.02 λ[335] 2.22 2.34 1.36 0.06 6.65 10.79 1.18 λ[336] 2.04 2.71 1.19 0.04 5.20 43.73 1.03 λ[337] 3.61 6.22 0.98 0.02 15.18 12.41 1.00 λ[338] 1.21 1.71 0.64 0.01 2.95 33.48 1.00 λ[339] 1.48 2.00 0.90 0.01 3.59 33.01 1.08 λ[340] 2.58 5.60 1.17 0.00 6.74 80.27 1.03 λ[341] 1.30 1.75 0.76 0.01 2.88 41.07 1.00 λ[342] 2.55 3.27 1.54 0.01 5.92 82.66 1.00 λ[343] 2.22 2.20 1.59 0.11 4.64 15.84 1.20 λ[344] 1.54 1.67 0.98 0.02 4.01 78.30 1.01 λ[345] 2.82 2.41 1.92 0.13 6.79 7.14 1.39 λ[346] 5.34 25.91 1.20 0.07 5.20 56.63 1.02 λ[347] 46.38 252.96 6.18 0.06 26.33 52.65 1.02 λ[348] 1.66 3.36 0.74 0.00 2.97 38.76 1.01 λ[349] 3.29 6.81 0.94 0.07 9.77 25.44 1.02 λ[350] 1.04 1.17 0.65 0.01 2.56 28.34 1.00 λ[351] 1.53 2.36 0.74 0.04 3.29 28.57 1.01 λ[352] 1.69 2.28 0.92 0.02 4.85 21.70 1.00 λ[353] 2.40 5.38 0.78 0.05 4.24 21.18 1.05 λ[354] 1.53 1.60 1.08 0.02 3.37 40.74 1.11 λ[355] 24.94 33.00 16.93 2.18 42.31 56.89 1.00 λ[356] 250.47 1774.95 58.71 11.83 317.89 126.90 1.01 λ[357] 127.29 160.87 73.59 19.37 247.42 21.73 1.12 λ[358] 985.63 1833.77 247.93 36.04 2847.59 12.10 1.06 λ[359] 780.10 1689.04 254.23 47.59 1676.78 20.04 1.02 λ[360] 178.30 359.01 78.64 19.61 457.15 55.16 1.04 λ[361] 1.56 1.60 1.05 0.08 3.41 32.28 1.14 λ[362] 40.22 35.61 25.89 7.95 110.62 21.48 1.05 λ[363] 45.09 38.31 31.93 11.00 89.10 60.26 1.00 λ[364] 67.98 84.07 41.74 11.10 150.94 39.68 1.07 λ[365] 3.60 4.27 2.18 0.07 9.09 25.46 1.09 ρ1 0.30 0.08 0.30 0.17 0.42 6.09 1.00 ρ2 0.20 0.02 0.20 0.17 0.22 11.73 1.17 ρ3 0.69 0.35 0.60 0.13 1.18 2.87 2.18 σ 0.25 0.00 0.24 0.24 0.25 93.77 1.01 τ 0.01 0.00 0.01 0.01 0.02 4.50 1.21 Number of divergences: 93
(<__main__.GP4 at 0x7f58f4d9ba10>, {'J': 20, 'L': 2.597720503807068, 'L3': 2.597720503807068, 'M': 10, 'M3': 5, 'day_of_week': DeviceArray([2, 3, 4, ..., 3, 4, 5], dtype=int32), 'day_of_year': DeviceArray([ 0, 1, 2, ..., 363, 364, 365], dtype=int32), 'labour_days_indicator': array([0, 0, 0, ..., 0, 0, 0]), 'memorial_days_indicator': array([0, 0, 0, ..., 0, 0, 0]), 'thanksgiving_days_indicator': array([0, 0, 0, ..., 0, 0, 0]), 'w0': 36.275986944786204, 'x': DeviceArray([-1.7318137, -1.7313395, -1.7308652, ..., 1.7308652, 1.7313395, 1.7318137], dtype=float32), 'y': DeviceArray([-1.0316722 , -0.5739162 , -0.0948692 , ..., 1.8230929 , 1.9410805 , -0.45770293], dtype=float32)}, <numpyro.infer.mcmc.MCMC at 0x7f58f4e64ed0>, {'c_aux': DeviceArray([1.0035206 , 1.0674242 , 1.096341 , 0.9919214 , 1.0566154 , 1.1029421 , 1.4180164 , 1.069648 , 1.0987023 , 1.1229846 , 1.2166708 , 0.66512513, 0.97954166, 1.0242714 , 1.0297555 , 1.0149187 , 0.8612341 , 1.2083969 , 1.1592759 , 1.298417 , 1.3017476 , 1.0516578 , 0.8360407 , 1.1264281 , 0.87835836, 0.88393086, 0.8620769 , 0.8776349 , 0.8654324 , 0.978372 , 0.997641 , 1.1635505 , 0.7747188 , 1.2213824 , 0.9685421 , 0.8087418 , 1.0505489 , 0.8628135 , 1.0962981 , 0.94462943, 1.2223512 , 1.1883767 , 1.0618249 , 0.8727425 , 1.0900264 , 0.7934674 , 1.1956798 , 0.9195027 , 0.78947216, 0.7955476 , 1.4887316 , 1.0339888 , 0.9811323 , 0.96886826, 0.95314497, 0.98480356, 1.1268278 , 1.3360422 , 1.5007441 , 1.1311313 , 1.2260166 , 1.1754614 , 0.88023347, 0.74161464, 0.6891218 , 0.816009 , 0.8400714 , 0.8959059 , 0.906974 , 1.2304934 , 1.2403107 , 0.9866896 , 0.92455584, 0.92913526, 0.9195807 , 0.9482658 , 1.2795943 , 0.8168847 , 0.8180654 , 0.8226611 , 0.8185304 , 0.82444113, 1.2389952 , 1.2430246 , 0.9024235 , 1.320111 , 1.5501537 , 1.6822745 , 1.3687747 , 0.8074088 , 0.86033905, 1.1465001 , 1.3417796 , 1.1610498 , 1.2761976 , 1.2683651 , 1.2560831 , 1.2323525 , 1.3236456 , 1.3169547 , 1.2183708 , 0.8646758 , 1.4632757 , 1.2344005 , 1.198937 , 1.1944242 , 1.3121778 , 1.3872042 , 1.4093139 , 1.1836346 , 1.1728518 , 1.1526077 , 1.1365019 , 1.1342019 , 1.1936022 , 1.1747388 , 0.9051808 , 0.90766156, 1.1507989 , 0.8858316 , 0.91692114, 0.93562454, 0.9190325 , 1.12983 , 1.0419892 , 1.2758052 , 1.4082571 , 0.63802904, 0.9549757 , 1.0376452 , 1.0291682 , 0.94548124, 1.001138 , 1.4391241 , 1.319605 , 0.9414245 , 0.8502568 , 0.843306 , 0.8384728 , 1.254688 , 1.0784552 , 1.1074822 , 1.1412195 , 1.1918652 , 1.1889344 , 1.2027742 , 1.1910751 , 1.1708274 , 1.1338265 , 1.1269532 , 1.1983341 , 0.7359273 , 0.7378601 , 0.71487606, 0.7613748 , 0.88405657, 0.8576663 , 0.9004777 , 0.90214115, 0.88256633, 0.8957456 , 0.91291654, 1.4553132 , 1.0934666 , 0.7422107 , 1.2505738 , 1.2574342 , 0.83123904, 1.5827439 , 1.5015426 , 1.5771823 , 1.583699 , 1.5921705 , 1.614482 , 1.4927853 , 0.729223 , 0.78664094, 1.1810539 , 0.85858226, 1.0823722 , 1.0457053 , 1.0680901 , 1.1462686 , 1.0312309 , 0.9907951 , 1.1908001 , 1.0682082 , 0.80965066, 0.74296147, 1.4541364 , 1.1872265 , 1.0007716 , 0.8270253 , 0.6570707 , 0.6428224 , 0.816364 , 1.1527146 , 0.95670104, 1.204256 , 0.85738796], dtype=float32), 'f1': DeviceArray([[ 0.377359 , 0.38068178, 0.3839981 , ..., 1.699215 , 1.700231 , 1.7012455 ], [ 0.38644361, 0.38964605, 0.39284217, ..., 1.6683059 , 1.6692538 , 1.6702005 ], [ 0.37316012, 0.376336 , 0.37950572, ..., 1.6627781 , 1.6637313 , 1.6646831 ], ..., [-0.64780456, -0.64473534, -0.6416721 , ..., 0.56986344, 0.5707181 , 0.57157207], [-0.69871 , -0.6953633 , -0.6920233 , ..., 0.61344993, 0.6144256 , 0.61540043], [-0.72821796, -0.72517204, -0.72213227, ..., 0.5071792 , 0.5080856 , 0.50899136]], dtype=float32), 'f2': DeviceArray([[-0.22727081, -0.26810092, -0.30270123, ..., -0.08395241, -0.13332017, -0.1817405 ], [-0.24758771, -0.2911256 , -0.32780707, ..., -0.09172934, -0.1460265 , -0.19863206], [-0.25746155, -0.30386013, -0.34270108, ..., -0.09014036, -0.14857635, -0.20506047], ..., [-0.20340192, -0.2455698 , -0.28173548, ..., -0.05507993, -0.10648457, -0.15651321], [-0.19871669, -0.23915087, -0.27334505, ..., -0.05342664, -0.10421441, -0.15322919], [-0.17879991, -0.22706652, -0.26868236, ..., -0.01266012, -0.06945241, -0.12557851]], dtype=float32), 'g3': DeviceArray([[-0.19302195, -0.19308133, -0.19314067, ..., 0.3619371 , 0.36193025, 0.3619232 ], [-0.11172947, -0.1117872 , -0.11184491, ..., 0.4498641 , 0.44984144, 0.44981855], [-0.0924181 , -0.09248593, -0.09255379, ..., 0.42489433, 0.4248724 , 0.42485023], ..., [-0.15533724, -0.15539303, -0.15544878, ..., 0.4158779 , 0.41586444, 0.4158507 ], [-0.1784026 , -0.17845319, -0.1785037 , ..., 0.3884407 , 0.38841876, 0.38839656], [-0.2666909 , -0.26679796, -0.26690498, ..., 0.29958504, 0.29961693, 0.29964855]], dtype=float32), 'intercept': DeviceArray([-0.59894073, -0.5592387 , -0.5711104 , -0.6194704 , -0.79444045, -0.8233949 , -0.67696303, -0.19243515, 0.02549784, 0.00233592, 0.13723227, 0.13052376, 0.05214081, 0.0187568 , 0.06510649, 0.139789 , 0.1998374 , 0.39346972, 0.32861632, 0.39126626, 0.38828814, 0.32836092, 0.4751022 , -0.0520874 , 0.04407027, -0.01765933, 0.13810064, 0.15839703, 0.15405853, 0.19140747, 0.18732038, 0.08701577, 0.36430347, 0.22721766, 0.33054647, 0.22539657, 0.31899574, 0.28817084, 0.30123937, 0.24469745, 0.20482069, 0.49674153, 0.44378132, 0.5256225 , 0.9438303 , 1.1951636 , 1.193363 , 0.86804163, 0.97138494, 0.9830971 , 0.95592666, 0.6475183 , 1.4222251 , 0.91010636, 0.8763497 , 0.47409663, 0.4580807 , 0.40216085, -0.03155162, 0.02712694, 0.0679799 , 0.07167282, 0.16708696, 0.10017075, 0.09996809, 0.19228773, 0.1368613 , 0.5702524 , 0.5500841 , 0.11216484, 0.00725829, 0.14828314, 0.7757977 , 0.7629436 , 0.7722038 , 0.7552395 , 0.42222086, 0.67400956, 0.66453135, 0.6686824 , 0.67929983, 0.6648182 , 0.13127711, 0.16028254, 0.08713239, 0.05112964, 0.01561272, 0.13198246, 0.11580215, 0.17485929, 0.1383318 , -0.20730798, -0.1856014 , -0.24916013, -0.305092 , -0.35251075, -0.33386314, -0.35665938, -0.3097334 , -0.31280896, -0.27127397, -0.20117901, -0.18208875, -0.15706363, -0.10299411, -0.10165437, -0.05206104, -0.0732981 , -0.08980646, -0.0625944 , -0.02934018, -0.03087701, -0.03370749, -0.03114881, -0.02181915, -0.01680631, 0.30005902, 0.29660895, 0.2199393 , -0.05369288, -0.07653379, -0.0451227 , -0.02205094, 0.0746526 , 0.36273423, 0.29259542, 0.43061903, 0.5541557 , 0.61467934, 0.60098284, 0.61680675, 0.6332072 , 0.3306894 , 0.3973225 , 0.40170714, 0.84383774, 1.0279374 , 0.6905603 , 0.7052811 , 0.79261404, 0.7092741 , 0.6269576 , 0.60505784, 0.6058909 , 0.6139264 , 0.6217106 , 0.62224877, 0.5937625 , 0.64013577, 0.6108975 , 0.60234886, 0.72290224, 0.7072538 , 0.7285793 , 0.8978359 , 0.9175436 , 0.9189768 , 1.2140446 , 1.153974 , 1.159469 , 1.1374316 , 1.0268486 , 0.672204 , 0.4923932 , 0.4303534 , 0.34623587, 0.45747742, 0.00797029, 0.26100847, 0.27803287, 0.18240836, 0.18385525, 0.20171489, 0.11956622, 0.01171808, 0.22036947, 0.2136803 , 0.23401348, -0.04048583, 0.36804792, -0.13499017, 0.23321724, 0.22837077, 0.3015689 , 0.39559492, 0.4192172 , 0.36632118, 0.46105283, 0.32810852, 0.79390943, 0.66354173, 0.690837 , 0.50967056, 0.49540672, 0.4930198 , 0.505943 , 0.568622 , 0.4933618 , 0.47677878, 0.5799701 ], dtype=float32), 'α1': DeviceArray([1.3359749, 1.2482646, 1.2413077, 1.3224478, 1.5632898, 1.5732967, 1.4549211, 1.4243919, 1.3111628, 1.3764184, 1.3687131, 1.1690855, 1.1294719, 1.1190931, 1.1184206, 1.2001898, 1.3384919, 1.450103 , 1.519442 , 1.5908291, 1.4118488, 1.3936036, 1.3364915, 1.1366606, 1.5949749, 1.5238892, 1.4241333, 1.4214945, 1.4126892, 1.3754357, 1.4185014, 1.5849097, 1.8427287, 1.9559699, 2.1795967, 2.1636074, 2.6980803, 2.2884095, 2.0342875, 3.2961798, 3.1438117, 2.75353 , 2.7279053, 2.6983216, 2.3822403, 2.1984415, 1.6833689, 1.5118405, 1.6097907, 1.6053339, 1.3254938, 1.5094352, 2.1191876, 1.6010479, 1.5839583, 1.4942684, 1.5647097, 1.4757909, 1.3774376, 1.6536766, 1.6527101, 1.6757987, 1.5048286, 1.4981296, 1.4348319, 1.4055456, 1.5337709, 1.9447963, 1.969125 , 2.244154 , 2.4117806, 2.407055 , 2.2448964, 2.2550404, 2.262904 , 2.1980734, 1.9496902, 2.1106553, 2.1180956, 2.250595 , 2.0758455, 2.061819 , 1.4063196, 1.4304821, 1.72105 , 1.7211047, 1.792273 , 1.8554609, 2.4551432, 3.041382 , 3.2314916, 2.437834 , 2.4377537, 2.500463 , 2.501313 , 2.4694142, 2.28744 , 2.0481608, 2.0770576, 2.0489676, 2.0684729, 2.1099334, 2.1207938, 2.022807 , 1.9207529, 1.9344333, 2.029993 , 2.035339 , 1.971619 , 2.188078 , 2.1727192, 2.0522916, 2.074209 , 2.08004 , 2.0837412, 2.0652702, 2.0819209, 2.074897 , 1.7503512, 1.5345459, 1.5607837, 1.6062338, 1.6267451, 1.5105932, 1.5277495, 1.8625882, 1.5816203, 1.4619143, 1.4435433, 1.4998032, 1.4918414, 1.4561709, 1.5153692, 1.7348632, 1.7002518, 1.9063381, 1.7462097, 1.5154009, 1.5004205, 1.6849105, 1.6903695, 1.7096765, 1.8051343, 1.8833171, 1.929699 , 1.91061 , 1.908497 , 1.938478 , 1.8423413, 1.8013263, 2.0305185, 2.1656713, 2.1610646, 2.128735 , 2.208515 , 2.328595 , 2.349392 , 2.0672019, 2.1096468, 2.0752716, 2.0387907, 2.049701 , 1.7293447, 1.9051994, 2.4276783, 2.512868 , 1.9668308, 1.3139303, 2.0160475, 1.7880834, 1.8767233, 1.9114075, 1.9772534, 1.9328009, 1.9759803, 1.515483 , 1.5134788, 1.5448198, 1.7854766, 2.4027264, 1.6727587, 2.0831332, 2.0927124, 1.7717481, 1.7698737, 1.9535074, 1.7749281, 1.8376243, 1.8539479, 1.6465328, 1.335743 , 1.3138717, 1.3702506, 1.5280087, 1.5053774, 1.453827 , 1.8013334, 1.4458423, 1.5285999, 1.3908235], dtype=float32), 'α2': DeviceArray([0.34261918, 0.35577485, 0.3593351 , 0.3520383 , 0.35963637, 0.30557743, 0.29768956, 0.30236617, 0.31619674, 0.3249453 , 0.31666175, 0.31631967, 0.30288258, 0.31629294, 0.31821206, 0.3325294 , 0.3151526 , 0.3637639 , 0.369634 , 0.36260888, 0.3589953 , 0.33967084, 0.32484597, 0.27861887, 0.28138432, 0.27442914, 0.270515 , 0.2718083 , 0.27329043, 0.27545643, 0.2755983 , 0.27231058, 0.28802007, 0.2721932 , 0.27418905, 0.2684641 , 0.2611554 , 0.26142776, 0.26461104, 0.27138352, 0.26179776, 0.2595555 , 0.25964645, 0.26320535, 0.26557052, 0.28903663, 0.29772466, 0.30544725, 0.31928644, 0.32251963, 0.30607978, 0.2988059 , 0.30073607, 0.28227758, 0.28693056, 0.2976276 , 0.30374527, 0.28869754, 0.29774192, 0.2899946 , 0.28921622, 0.28858018, 0.2878983 , 0.29099277, 0.28535828, 0.27938268, 0.27290455, 0.2931041 , 0.2929452 , 0.27203676, 0.26959395, 0.26268712, 0.24454302, 0.24393985, 0.24504273, 0.24348657, 0.23249555, 0.23121063, 0.23198026, 0.23664458, 0.23586158, 0.2398621 , 0.21833481, 0.21782017, 0.22418496, 0.22205617, 0.22075565, 0.21611165, 0.22493967, 0.21186163, 0.20373693, 0.20710875, 0.20644306, 0.20096114, 0.20590182, 0.20434909, 0.19839773, 0.20858906, 0.20461553, 0.20583153, 0.2047986 , 0.22346568, 0.22041337, 0.22214997, 0.22409701, 0.22402509, 0.21667434, 0.21900769, 0.21980691, 0.21644427, 0.21551934, 0.21744198, 0.21783897, 0.21792626, 0.21409489, 0.21381955, 0.20918329, 0.20961146, 0.20731905, 0.21232395, 0.21446754, 0.21594802, 0.21453793, 0.21782105, 0.23072295, 0.20986193, 0.21585663, 0.20049892, 0.2071506 , 0.21063136, 0.20762645, 0.20897365, 0.18556389, 0.22034608, 0.21903251, 0.22404708, 0.21813592, 0.21310803, 0.21304119, 0.20224743, 0.20795533, 0.20153506, 0.19907747, 0.20370813, 0.20331961, 0.19969918, 0.19896935, 0.20078273, 0.2037607 , 0.21079057, 0.20597059, 0.2208384 , 0.21945338, 0.21498804, 0.2250881 , 0.21822892, 0.21630922, 0.2231871 , 0.22485583, 0.22782367, 0.22792587, 0.2285149 , 0.21491098, 0.22873646, 0.21333098, 0.22398126, 0.20951386, 0.20267323, 0.198971 , 0.19576684, 0.1994264 , 0.19736026, 0.19681999, 0.20029779, 0.20336907, 0.217321 , 0.20973459, 0.22263022, 0.20313917, 0.19235612, 0.2109843 , 0.19432807, 0.19507396, 0.19736414, 0.18551067, 0.1900514 , 0.18659264, 0.19457912, 0.19424577, 0.19805093, 0.20720175, 0.20783974, 0.20816627, 0.20203218, 0.20247555, 0.19678165, 0.19899297, 0.19733094, 0.1949629 , 0.20598885], dtype=float32), 'α3': DeviceArray([0.08942518, 0.07973044, 0.075909 , 0.08128737, 0.05652755, 0.08247125, 0.06615686, 0.0632497 , 0.1340084 , 0.11557455, 0.09924944, 0.09207249, 0.0994093 , 0.09159754, 0.0922543 , 0.07758522, 0.08497488, 0.11254886, 0.1010536 , 0.10529424, 0.09926244, 0.1171812 , 0.12875636, 0.06542897, 0.06027276, 0.05627688, 0.08165479, 0.07760776, 0.07484715, 0.09644184, 0.0971431 , 0.09766815, 0.07826201, 0.12475363, 0.13487203, 0.13427106, 0.08452386, 0.09300109, 0.06872709, 0.06494492, 0.06195194, 0.07376265, 0.08309042, 0.06838817, 0.10198341, 0.07644703, 0.05482794, 0.06409483, 0.04765965, 0.05271098, 0.06522322, 0.05988888, 0.05870394, 0.05066052, 0.05042728, 0.03312424, 0.03510564, 0.0295094 , 0.02692202, 0.03955153, 0.04682909, 0.04590474, 0.0391186 , 0.03976667, 0.03926837, 0.04999361, 0.05358674, 0.06329434, 0.06603529, 0.07023308, 0.08431753, 0.09802187, 0.15801258, 0.15722524, 0.15523922, 0.15451296, 0.15619308, 0.0797955 , 0.08027352, 0.07430212, 0.07379183, 0.06493596, 0.10549113, 0.09612752, 0.18007934, 0.25779498, 0.22860502, 0.13541801, 0.16367733, 0.10700822, 0.11056818, 0.10307644, 0.10571951, 0.11275399, 0.12496949, 0.11477432, 0.11486813, 0.127678 , 0.13256522, 0.13941675, 0.12665068, 0.11645649, 0.131214 , 0.16063847, 0.15424389, 0.15604441, 0.18117142, 0.18537344, 0.18871525, 0.18619728, 0.18292499, 0.16877028, 0.17461503, 0.17298327, 0.17704731, 0.16871041, 0.12745535, 0.1266063 , 0.10892701, 0.0750357 , 0.07221004, 0.06191223, 0.06486583, 0.08121255, 0.09267172, 0.11462413, 0.10654753, 0.2815021 , 0.16506508, 0.16040227, 0.155888 , 0.13600013, 0.10347732, 0.18356018, 0.17517985, 0.07174227, 0.0779641 , 0.09894547, 0.10227562, 0.05272068, 0.05765352, 0.06503303, 0.06320313, 0.06455172, 0.06660246, 0.06648647, 0.06695017, 0.06291368, 0.07024955, 0.07759967, 0.07148168, 0.06858776, 0.06932076, 0.07533816, 0.07103124, 0.07700936, 0.0787188 , 0.09249075, 0.09355436, 0.08910121, 0.08630875, 0.10427912, 0.1401635 , 0.20712179, 0.15374234, 0.1440241 , 0.13938744, 0.09986412, 0.05299292, 0.06788759, 0.06392246, 0.05713381, 0.05564745, 0.05112166, 0.05484453, 0.0563199 , 0.06111703, 0.04633876, 0.03736979, 0.03561047, 0.04913949, 0.07073411, 0.07364811, 0.04203748, 0.04032195, 0.04572336, 0.04735657, 0.04803981, 0.04682632, 0.18300445, 0.11276303, 0.20035324, 0.18472895, 0.15087411, 0.14992408, 0.13338618, 0.13302717, 0.10987154, 0.15696459, 0.0709467 ], dtype=float32), 'β1': DeviceArray([[ 1.0186691 , -1.5967587 , 0.53186256, ..., -1.9416343 , -0.85504377, -1.1704478 ], [ 1.0018059 , -1.6568508 , 0.5658884 , ..., -1.8843524 , -0.8433561 , -1.1180311 ], [ 1.0066857 , -1.6509707 , 0.5631488 , ..., -1.8808551 , -0.83471346, -1.130671 ], ..., [-1.2564312 , -1.8122745 , -0.09859934, ..., -1.4100857 , -0.6458877 , -0.6775133 ], [-1.1370164 , -1.7457957 , -0.12023044, ..., -1.4936265 , -0.67909265, -0.7231631 ], [-1.4266887 , -1.8124841 , -0.12707376, ..., -1.4388802 , -0.65941733, -0.70175254]], dtype=float32), 'β2_cos': DeviceArray([[-1.1439912 , -0.11747075, 0.26400453, ..., -0.16831358, 0.80903494, -0.5449284 ], [-1.1150702 , -0.10100719, 0.18186182, ..., -0.42756158, 1.0244316 , -0.5459449 ], [-1.1080353 , -0.0956095 , 0.19046232, ..., -0.3491494 , 1.1101207 , -0.5778732 ], ..., [-2.1362352 , -0.20966646, 0.43483877, ..., 0.65586287, -0.9721226 , 1.6891339 ], [-2.072359 , -0.16471301, 0.47168234, ..., -0.6104178 , -0.3777624 , 0.28434357], [-1.9601179 , -0.15582149, 0.3775222 , ..., -0.57208 , 0.65434265, 0.29855445]], dtype=float32), 'β2_sin': DeviceArray([[-2.1800478 , 1.5242962 , -0.05817219, ..., 2.0887554 , -3.066425 , 0.310512 ], [-2.0168016 , 1.4090737 , -0.04600942, ..., 2.477024 , -2.5826929 , 0.39085478], [-2.0212257 , 1.4013579 , -0.04644156, ..., 2.1943502 , -2.5644035 , 0.7875392 ], ..., [-3.9120274 , 2.570239 , -0.07230773, ..., 0.8573707 , -0.35681483, 0.34993708], [-3.9947417 , 2.577894 , -0.06142015, ..., 1.1433209 , -0.22744548, -0.5166404 ], [-3.7567298 , 2.6355183 , -0.11767332, ..., 0.5588796 , -0.85279554, -1.7955583 ]], dtype=float32), 'β3': DeviceArray([[-2.25494042e-01, -1.46071517e+00, 6.08891845e-01, -4.64001358e-01, 9.95878279e-02], [ 2.79709637e-01, -1.64807665e+00, 8.93958092e-01, -4.43303972e-01, 1.88137695e-01], [ 2.87243128e-01, -1.59595358e+00, 8.71975958e-01, -3.49336922e-01, 2.30955899e-01], [ 2.34873131e-01, -1.60696983e+00, 8.16743255e-01, -4.39857841e-01, 1.78526357e-01], [ 4.49096680e-01, -1.72168922e+00, 1.00062287e+00, -5.27549326e-01, 3.18604082e-01], [-1.07580282e-01, -1.63254583e+00, 7.68265784e-01, -3.74199808e-01, 1.79116040e-01], [-4.57857817e-01, -1.79507613e+00, 7.53895938e-01, -4.58943546e-01, 2.20478341e-01], [-8.43444645e-01, -1.89823806e+00, 6.11702979e-01, -5.25012136e-01, 1.65870950e-01], [-7.12224960e-01, -1.90883124e+00, 6.84252203e-01, -4.61463988e-01, 5.91973327e-02], [-9.85753477e-01, -2.24334526e+00, 6.34973466e-01, -4.68662590e-01, 7.46297985e-02], [-8.70377362e-01, -2.11231041e+00, 7.94557393e-01, -5.09331763e-01, 7.25615397e-02], [-1.00346577e+00, -2.33452463e+00, 7.81866252e-01, -6.83736682e-01, 2.37716526e-01], [-1.54092789e+00, -2.36985826e+00, 6.49759114e-01, -5.77267885e-01, 2.13664293e-01], [-1.23136640e+00, -2.35090470e+00, 7.48063684e-01, -4.83707190e-01, 2.18829334e-01], [-1.31145537e+00, -2.32063818e+00, 7.42327452e-01, -4.16286469e-01, 1.09693676e-01], [-1.68850577e+00, -2.55097985e+00, 6.65252328e-01, -7.03421056e-01, 4.72567976e-02], [-1.74991632e+00, -2.54185796e+00, 5.88188350e-01, -6.30843163e-01, 1.78753927e-01], [-4.60469484e-01, -2.14867949e+00, 7.65095770e-01, -5.65930426e-01, 1.18230112e-01], [-4.38302577e-01, -2.21099162e+00, 9.14637208e-01, -4.15203720e-01, 1.83080584e-01], [-1.69566959e-01, -2.12449074e+00, 9.91077662e-01, -5.01934171e-01, 3.01179498e-01], [-3.30435097e-01, -2.09933257e+00, 7.99450696e-01, -4.36468840e-01, 9.83947664e-02], [-1.87637955e-01, -1.94997656e+00, 7.50894845e-01, -4.37132418e-01, 1.96877941e-01], [ 2.96254661e-02, -1.91298103e+00, 8.81883264e-01, -4.18196797e-01, 2.32391313e-01], [-9.19643119e-02, -2.02324510e+00, 9.20799494e-01, -5.00029087e-01, 1.51736110e-01], [ 1.14378914e-01, -2.32241130e+00, 1.02734590e+00, -5.97073793e-01, 2.38370135e-01], [-1.39159843e-01, -2.13912940e+00, 7.88018346e-01, -5.60290158e-01, 1.84181869e-01], [-2.09185019e-01, -1.88439953e+00, 8.53836417e-01, -4.17922914e-01, 2.71195412e-01], [-1.89038366e-01, -1.88305378e+00, 8.43142748e-01, -4.40794110e-01, 1.97577983e-01], [-2.07409680e-01, -1.88462663e+00, 8.25415492e-01, -4.24800932e-01, 2.36095265e-01], [-1.46537006e-01, -1.74578929e+00, 7.56973028e-01, -4.22532111e-01, 1.86355695e-01], [-2.02339485e-01, -1.75219989e+00, 7.76456058e-01, -4.20111239e-01, 1.93937898e-01], [ 1.30380869e-01, -1.76388192e+00, 7.77662516e-01, -4.50950712e-01, 1.04762934e-01], [ 6.97997630e-01, -1.85241640e+00, 1.03324473e+00, -4.53591168e-01, 2.61448264e-01], [ 1.05733740e+00, -1.55256355e+00, 1.02932370e+00, -3.74881476e-01, 3.01198423e-01], [ 7.55229771e-01, -1.39928865e+00, 8.60567987e-01, -2.79986888e-01, 2.11456001e-01], [ 9.51294482e-01, -1.40892053e+00, 8.25252652e-01, -4.01627064e-01, 2.42437467e-01], [ 1.02229571e+00, -1.65127707e+00, 9.47772801e-01, -4.69930887e-01, 2.83901423e-01], [ 1.43696558e+00, -1.58326316e+00, 1.07949054e+00, -4.91392583e-01, 3.17990422e-01], [ 1.38598132e+00, -1.71267867e+00, 1.25488174e+00, -4.64300454e-01, 3.78681242e-01], [ 6.69273555e-01, -1.81344843e+00, 9.15592313e-01, -5.39932787e-01, 1.82600468e-01], [ 6.08919442e-01, -1.77470136e+00, 1.06585467e+00, -4.31823373e-01, 3.03190768e-01], [ 3.43890935e-01, -1.67993927e+00, 8.25350106e-01, -5.09028554e-01, 2.31905520e-01], [ 2.76367426e-01, -1.55946720e+00, 8.26879263e-01, -3.72194171e-01, 2.94043273e-01], [ 2.74279147e-01, -1.74547851e+00, 7.85828114e-01, -5.35509348e-01, -1.56865753e-02], [ 1.41933918e-01, -1.48784280e+00, 8.36284876e-01, -3.25039387e-01, 4.21862096e-01], [ 5.51803827e-01, -1.69557214e+00, 8.18087459e-01, -5.12778640e-01, 6.73534721e-02], [-4.95394915e-02, -2.04257727e+00, 7.69567072e-01, -6.36391759e-01, 2.00235590e-01], [-6.97343886e-01, -1.95309639e+00, 5.78191936e-01, -6.01836205e-01, 9.68116522e-02], [-5.26053607e-01, -2.14441037e+00, 9.63778257e-01, -4.62975979e-01, 2.65288144e-01], [-6.56922460e-01, -2.10389805e+00, 7.43471742e-01, -5.00526726e-01, 2.69098938e-01], [-8.79454985e-02, -1.85236013e+00, 7.37395823e-01, -4.44686949e-01, 1.05701312e-01], [ 2.51430839e-01, -1.88913798e+00, 1.03513932e+00, -4.32765692e-01, 1.82110161e-01], [ 1.12964317e-01, -1.88991141e+00, 8.45836818e-01, -5.91421127e-01, 2.84351230e-01], [-2.45337665e-01, -2.08818412e+00, 7.96322584e-01, -5.87974012e-01, 9.20709223e-02], [-2.22614422e-01, -2.02833486e+00, 9.63909447e-01, -5.32965720e-01, 3.41790736e-01], [-2.17517063e-01, -2.41629434e+00, 1.09213197e+00, -6.46200240e-01, 3.39689374e-01], [-1.91491559e-01, -2.32503557e+00, 1.06915104e+00, -6.79378867e-01, 3.26626658e-01], [-4.14124876e-01, -2.42675257e+00, 1.18893397e+00, -5.61446071e-01, 3.69903684e-01], [-1.09037831e-01, -2.45465636e+00, 1.04366589e+00, -9.12030697e-01, 2.40252018e-01], [ 2.18783140e-01, -1.93723416e+00, 1.09009886e+00, -4.99098033e-01, 4.25499856e-01], [ 1.82018906e-01, -1.78137910e+00, 9.70363200e-01, -6.72083497e-01, 4.32674974e-01], [ 2.28622571e-01, -1.77176499e+00, 9.55990672e-01, -5.54767132e-01, 3.60952586e-01], [-6.76337957e-01, -2.11034703e+00, 8.55224490e-01, -6.03544235e-01, 2.53897369e-01], [-9.62054968e-01, -2.06457663e+00, 7.45520055e-01, -7.43948042e-01, 1.99956521e-01], [-9.09929574e-01, -2.05766869e+00, 7.68307030e-01, -5.84118962e-01, 2.46361598e-01], [-1.00714505e+00, -1.69883204e+00, 5.70977092e-01, -7.79227614e-01, 2.21504986e-01], [-1.01425231e+00, -1.62400293e+00, 6.18276596e-01, -5.90439737e-01, 3.18112642e-01], [-6.65010452e-01, -1.51264620e+00, 5.19941270e-01, -6.30123913e-01, 1.50898740e-01], [-6.96269214e-01, -1.51094031e+00, 5.13903081e-01, -6.15088284e-01, 6.84080571e-02], [-8.86263400e-02, -1.60798419e+00, 7.65802681e-01, -4.85964507e-01, 2.31880307e-01], [-1.68349827e-04, -1.45701659e+00, 7.19238460e-01, -4.10075694e-01, 1.89236179e-01], [-2.22459316e-01, -1.43589807e+00, 6.52936339e-01, -3.18239868e-01, 1.75585568e-01], [-3.63694876e-01, -1.15284240e+00, 4.75840986e-01, -2.99298704e-01, 1.43135533e-01], [-3.97501588e-01, -1.15072715e+00, 4.61487174e-01, -3.06207091e-01, 1.47214442e-01], [-3.59178871e-01, -1.14951420e+00, 4.41145688e-01, -3.23164314e-01, 1.37355149e-01], [-3.48467410e-01, -1.15866685e+00, 4.31316644e-01, -3.17953020e-01, 7.30287060e-02], [-4.11006622e-02, -1.17668569e+00, 5.28797328e-01, -3.11099082e-01, 1.17353268e-01], [ 4.72868621e-01, -1.39557016e+00, 8.71886909e-01, -4.48816806e-01, 3.61247003e-01], [ 4.59691703e-01, -1.39189732e+00, 8.75288248e-01, -4.47119355e-01, 3.67031515e-01], [ 4.63798612e-01, -1.36794221e+00, 8.37729573e-01, -4.94993240e-01, 3.44714373e-01], [ 4.25464481e-01, -1.44846523e+00, 9.66437042e-01, -5.36575675e-01, 3.91954780e-01], [ 4.95635390e-01, -1.46460009e+00, 8.99052680e-01, -5.32166421e-01, 2.44073138e-01], [ 9.73180830e-01, -1.31535947e+00, 9.15990651e-01, -3.78152370e-01, 3.85849237e-01], [ 1.09069359e+00, -1.28984308e+00, 8.86083782e-01, -3.93899769e-01, 2.07158759e-01], [ 1.04594100e+00, -9.79770720e-01, 7.69310176e-01, -2.90554494e-01, 2.35855818e-01], [ 9.98630166e-01, -8.39011967e-01, 6.99349880e-01, -3.01527321e-01, 2.42219672e-01], [ 1.01357520e+00, -8.99715304e-01, 6.91976607e-01, -2.45474130e-01, 1.78057343e-01], [ 1.38253784e+00, -1.17981577e+00, 9.27878499e-01, -3.11510712e-01, 2.74155796e-01], [ 1.47079813e+00, -1.01223135e+00, 9.26917911e-01, -3.27065855e-01, 3.51164013e-01], [ 8.20944667e-01, -1.24115479e+00, 8.72149467e-01, -3.69439870e-01, 2.77577311e-01], [ 8.26721430e-01, -1.26946306e+00, 8.17348540e-01, -3.98808986e-01, 2.88489640e-01], [ 3.49103838e-01, -1.24715114e+00, 6.75485313e-01, -4.35911447e-01, 2.57801741e-01], [ 2.69622236e-01, -1.19343114e+00, 7.13927984e-01, -3.96944344e-01, 2.15930462e-01], [ 1.91592917e-01, -1.10138714e+00, 6.75536573e-01, -5.03293991e-01, 2.76794881e-01], [ 1.41129717e-01, -1.10622036e+00, 5.93459427e-01, -5.01655400e-01, 2.07226500e-01], [ 1.59448385e-01, -1.09172380e+00, 6.70570314e-01, -4.97277021e-01, 1.38556793e-01], [ 2.24152580e-01, -1.13045025e+00, 6.50340080e-01, -4.89501894e-01, 4.67745811e-01], [ 3.45434070e-01, -1.08856332e+00, 7.53556371e-01, -5.35800576e-01, 5.54345369e-01], [ 2.99436092e-01, -9.77158904e-01, 7.35457301e-01, -4.35213387e-01, 3.74354333e-01], [ 2.92039156e-01, -1.00590920e+00, 7.14553237e-01, -4.39038157e-01, 3.15563142e-01], [ 3.81815761e-01, -1.10217142e+00, 6.85234606e-01, -4.68941569e-01, 3.25370193e-01], [ 6.55947626e-01, -1.05293810e+00, 8.45307052e-01, -4.47039336e-01, 4.05940264e-01], [ 8.40049803e-01, -1.10532832e+00, 8.38797271e-01, -2.66902894e-01, 2.98491359e-01], [ 7.96793640e-01, -9.87409174e-01, 7.69299686e-01, -3.44986588e-01, 3.30901921e-01], [ 7.75956571e-01, -9.76529479e-01, 7.57621944e-01, -2.53023773e-01, 2.96693355e-01], [ 7.75115848e-01, -9.94830132e-01, 7.64276743e-01, -2.54621685e-01, 3.07852626e-01], [ 9.92556214e-01, -9.23570931e-01, 7.68685758e-01, -3.44671339e-01, 3.02431703e-01], [ 9.21900272e-01, -9.03208017e-01, 7.53853381e-01, -2.99781203e-01, 3.12629610e-01], [ 9.14606690e-01, -9.17439520e-01, 7.62656271e-01, -3.30813438e-01, 3.25886726e-01], [ 8.77336204e-01, -9.07337725e-01, 7.57423341e-01, -2.96203941e-01, 3.19393516e-01], [ 8.69183719e-01, -9.30487216e-01, 7.56383359e-01, -3.06134224e-01, 3.36653858e-01], [ 9.08373475e-01, -9.48780179e-01, 7.52439678e-01, -2.88038224e-01, 3.13960224e-01], [ 9.45978165e-01, -9.19269800e-01, 7.90054739e-01, -2.95406371e-01, 3.14700305e-01], [ 9.44216907e-01, -9.21224952e-01, 7.87064016e-01, -2.94974208e-01, 3.07018787e-01], [ 8.66877794e-01, -9.72053170e-01, 7.76424229e-01, -2.85477608e-01, 3.24374586e-01], [ 9.24762487e-01, -9.57461417e-01, 7.66142488e-01, -2.96624601e-01, 3.25003982e-01], [ 1.37529802e+00, -1.01170015e+00, 1.12246430e+00, -4.46265966e-01, 6.68734193e-01], [ 1.38357162e+00, -1.01343894e+00, 1.10487235e+00, -4.44562674e-01, 6.51211023e-01], [ 1.89285481e+00, -1.19707239e+00, 1.35585964e+00, -5.28841972e-01, 9.61302280e-01], [ 1.11770797e+00, -1.42479527e+00, 1.39745855e+00, -7.00237036e-01, 1.21359789e+00], [ 1.13537371e+00, -1.40972650e+00, 1.43246019e+00, -7.15384722e-01, 1.29799986e+00], [ 1.21472704e+00, -1.49125361e+00, 1.42363107e+00, -7.01574266e-01, 8.81546795e-01], [ 1.21233594e+00, -1.47320890e+00, 1.38924468e+00, -7.37288713e-01, 8.90930176e-01], [-1.98350057e-01, -1.37617981e+00, 8.26889455e-01, -1.00938523e+00, 8.92491579e-01], [-1.79454252e-01, -1.29788888e+00, 7.79264271e-01, -9.43269372e-01, 3.38354409e-01], [-2.19297826e-01, -1.11879432e+00, 7.02415526e-01, -7.33361542e-01, 7.30615854e-01], [-1.94584161e-01, -1.12293422e+00, 7.04802334e-01, -7.07507372e-01, 3.94215316e-01], [-1.47757322e-01, -7.99892843e-01, 5.88642061e-01, -1.25611186e+00, 1.36253262e+00], [-3.90230417e-01, -1.00208998e+00, 5.49974382e-01, -1.17859435e+00, 8.09185028e-01], [-3.60986173e-01, -1.04481816e+00, 5.36109269e-01, -1.14901543e+00, 7.61778891e-01], [-3.44202965e-01, -1.04681265e+00, 6.11427784e-01, -1.30300796e+00, 1.11547542e+00], [-3.80502433e-01, -1.12170804e+00, 7.97704101e-01, -1.40245974e+00, 1.44682944e+00], [ 1.29505903e-01, -1.17980945e+00, 7.70615518e-01, -7.89252102e-01, 6.14891410e-01], [-8.39953795e-02, -9.34743226e-01, 5.34008861e-01, -7.79947281e-01, 5.35937369e-01], [-1.03947021e-01, -9.67223227e-01, 6.02730393e-01, -7.23428786e-01, 5.54148614e-01], [-8.37088764e-01, -1.46951461e+00, 4.42605555e-01, -1.12003112e+00, 2.28749499e-01], [-8.99515867e-01, -1.39600122e+00, 5.42936146e-01, -7.68963695e-01, 3.02860379e-01], [-8.70877624e-01, -1.24112499e+00, 3.74492526e-01, -9.04047072e-01, 3.95396471e-01], [-8.85800779e-01, -1.21938944e+00, 3.75089288e-01, -8.91411543e-01, 3.62937093e-01], [-5.75828910e-01, -1.70707417e+00, 9.19318020e-01, -1.53421676e+00, 8.47000182e-01], [-4.43824530e-01, -1.67268634e+00, 8.62688959e-01, -1.62446809e+00, 7.31781363e-01], [-2.67021656e-01, -1.63810706e+00, 8.91521871e-01, -1.60824370e+00, 8.60207379e-01], [-2.85323530e-01, -1.60312831e+00, 1.17499292e+00, -1.58923197e+00, 1.03729832e+00], [-2.88092434e-01, -1.52695906e+00, 1.08449745e+00, -1.60233259e+00, 8.83469760e-01], [-2.74542332e-01, -1.51718843e+00, 1.06998360e+00, -1.58412063e+00, 8.83555770e-01], [-3.24236751e-01, -1.56850040e+00, 1.14700520e+00, -1.58454907e+00, 8.61864090e-01], [-3.21044892e-01, -1.56429958e+00, 1.13868427e+00, -1.58223331e+00, 8.52095008e-01], [-3.27768236e-01, -1.56183696e+00, 1.13155162e+00, -1.53620946e+00, 9.01327372e-01], [-4.34017807e-01, -1.55612254e+00, 1.14223611e+00, -1.52268589e+00, 7.62608945e-01], [-3.88032794e-01, -1.52189898e+00, 8.96825135e-01, -1.54516733e+00, 5.97722232e-01], [-4.93166268e-01, -1.47272491e+00, 8.08962166e-01, -1.61111188e+00, 5.68811357e-01], [-1.08707702e+00, -1.45654857e+00, 4.12581116e-01, -1.30002630e+00, 1.44972622e-01], [-1.06609905e+00, -1.46241748e+00, 4.07561719e-01, -1.30351377e+00, 1.28040597e-01], [-1.08914411e+00, -1.47823966e+00, 3.80639464e-01, -1.21203876e+00, 1.87558353e-01], [-1.20896876e+00, -1.42324507e+00, 3.99990231e-01, -9.68398511e-01, 1.68957353e-01], [-1.08821046e+00, -1.41292572e+00, 3.23496282e-01, -1.16198337e+00, 2.35423055e-02], [-1.08240294e+00, -1.41019511e+00, 3.19467217e-01, -1.13440275e+00, -5.19280024e-02], [-1.08106148e+00, -1.33759105e+00, 3.24167639e-01, -1.26574516e+00, -2.28975937e-01], [-1.09827602e+00, -1.33213198e+00, 3.16603720e-01, -1.23092830e+00, -2.39273027e-01], [-1.22326803e+00, -1.35827446e+00, 3.85944009e-01, -1.25936007e+00, -1.46488309e-01], [-1.23481059e+00, -1.33689356e+00, 3.94955248e-01, -1.29056501e+00, -1.20649979e-01], [-1.05614269e+00, -1.26422048e+00, 1.74362034e-01, -1.48382688e+00, -1.73866689e-01], [-7.61524558e-01, -1.19913375e+00, 3.92717898e-01, -1.66102457e+00, -7.52078146e-02], [-8.43497694e-01, -9.52314496e-01, 1.51366308e-01, -1.66176617e+00, 2.30459049e-01], [-9.12266433e-01, -1.13751364e+00, 1.47703037e-01, -1.65876114e+00, -3.15255485e-02], [-8.46185625e-01, -1.19206989e+00, 3.95358503e-01, -1.65326202e+00, 2.67349482e-01], [-7.62795985e-01, -1.16164184e+00, 3.59085470e-01, -1.64216185e+00, 2.11183518e-01], [-8.61025691e-01, -1.26816595e+00, 3.98124158e-01, -1.20498419e+00, 3.04048628e-01], [-1.03026819e+00, -1.62052608e+00, 5.17306507e-01, -8.27251017e-01, 1.91920966e-01], [-8.65754426e-01, -1.42138982e+00, 4.60226953e-01, -6.93146765e-01, 1.58507675e-01], [-7.75406420e-01, -1.50120258e+00, 6.16970897e-01, -6.86604798e-01, 3.42571884e-01], [-7.40206599e-01, -1.52778435e+00, 6.85642004e-01, -7.08871305e-01, 3.44831705e-01], [-7.56010115e-01, -1.53903115e+00, 6.31874084e-01, -6.78667963e-01, 3.38048100e-01], [-8.08861375e-01, -1.57711911e+00, 6.68016732e-01, -6.78430319e-01, 2.09933087e-01], [-5.79566300e-01, -1.59802246e+00, 7.37039089e-01, -7.45132864e-01, 6.40664876e-01], [ 2.23687887e-01, -1.58146262e+00, 1.01243889e+00, -4.91730481e-01, 6.23510897e-01], [ 5.02520084e-01, -1.55143154e+00, 1.02049780e+00, -4.34801519e-01, 2.60072112e-01], [ 7.57331431e-01, -1.75564241e+00, 1.04335797e+00, -6.58461094e-01, 1.27291679e-01], [ 1.24185002e+00, -2.04313803e+00, 1.40932512e+00, -7.83354998e-01, 7.37119794e-01], [ 1.06601703e+00, -1.98077369e+00, 1.39669144e+00, -9.27249789e-01, 5.76615930e-01], [-3.28042418e-01, -1.79402626e+00, 1.06804824e+00, -1.39586544e+00, 1.32745802e+00], [-5.64565122e-01, -1.52188683e+00, 7.37974048e-01, -1.66559923e+00, 9.91712451e-01], [-5.79900086e-01, -1.51015210e+00, 6.23572767e-01, -1.60454917e+00, 8.21896970e-01], [-6.50357723e-01, -1.89402843e+00, 8.61601055e-01, -1.27432883e+00, 3.81861508e-01], [-3.49902570e-01, -1.88432896e+00, 9.83938456e-01, -1.12694597e+00, 7.46418595e-01], [-3.40156168e-01, -1.72764242e+00, 9.94555831e-01, -1.21823311e+00, 9.18833375e-01], [-1.47062629e-01, -1.73178518e+00, 1.10169625e+00, -1.14578140e+00, 7.22169220e-01], [-8.80597234e-02, -1.69322729e+00, 9.90764737e-01, -1.14298165e+00, 6.87217414e-01], [ 3.21525848e-03, -1.74529958e+00, 9.51791108e-01, -1.17777741e+00, 4.28713143e-01], [-5.53083181e-01, -8.99014235e-01, 3.62973690e-01, -5.60345769e-01, 3.10066104e-01], [-9.68888402e-01, -1.12072861e+00, 2.61815637e-01, -7.17939138e-01, 1.71475902e-01], [-6.07478499e-01, -8.80139768e-01, 3.61030638e-01, -8.33042145e-01, 6.29521191e-01], [-4.79981363e-01, -9.25220430e-01, 3.56436849e-01, -6.48564398e-01, -1.34425089e-01], [-4.61643308e-01, -1.02619052e+00, 4.16774482e-01, -7.98941910e-01, 1.76140428e-01], [-4.38511491e-01, -1.01489913e+00, 4.47448432e-01, -7.87421823e-01, 2.18223765e-01], [-6.30705178e-01, -9.77186918e-01, 3.83747429e-01, -4.42319244e-01, 1.91537157e-01], [-5.63201785e-01, -1.09358394e+00, 5.18520653e-01, -8.60416055e-01, 8.51437211e-01], [ 3.99127901e-02, -1.17596698e+00, 8.75760734e-01, -1.10919118e+00, 7.95817137e-01], [-4.81193401e-02, -1.02370262e+00, 8.13679874e-01, -1.19553113e+00, 4.37769055e-01], [-6.28497899e-01, -1.51298594e+00, 8.68679523e-01, -1.59677875e+00, 1.78451049e+00]], dtype=float32), 'β4': DeviceArray([[-1.29459429e+00, -6.46476805e-01, -1.04380324e-01, ..., 4.66587126e-01, 5.22054136e-01, 8.99345949e-02], [-1.21025634e+00, -6.87893212e-01, -1.12456292e-01, ..., 4.65159357e-01, 5.59796333e-01, 6.01715930e-02], [-1.19163465e+00, -7.05938876e-01, -9.88886431e-02, ..., 4.62876022e-01, 5.62246263e-01, 4.50099595e-02], ..., [-1.29665112e+00, -7.97928333e-01, -1.51913822e-01, ..., 4.52890873e-01, 4.51787919e-01, 2.57362449e-03], [-1.36449683e+00, -6.79005325e-01, -8.63917619e-02, ..., 3.98053259e-01, 4.18700665e-01, -8.76740494e-04], [-1.36791694e+00, -8.89937103e-01, -2.40562275e-01, ..., 4.19276297e-01, 5.62726438e-01, 5.40159456e-03]], dtype=float32), 'β5_labour': DeviceArray([-0.8211012 , -0.8455037 , -0.85414845, -0.8818088 , -0.91380775, -0.8867797 , -0.84825444, -0.89664495, -0.9000473 , -0.9034106 , -0.9301592 , -0.89364064, -0.9562681 , -0.95119405, -0.94431746, -0.9611569 , -0.9207143 , -0.9091836 , -0.8888415 , -0.88593507, -0.87059426, -0.9854968 , -0.9960173 , -0.8087519 , -0.8593087 , -0.89012873, -0.8905929 , -0.89759314, -0.90100306, -0.90009195, -0.90097743, -0.9155342 , -0.8852692 , -0.9167913 , -0.9586716 , -0.94855845, -0.9751304 , -0.93085825, -0.9176425 , -0.9155564 , -0.8888004 , -0.88222855, -0.9038922 , -0.90291613, -0.9429694 , -0.8534102 , -0.8727986 , -0.9040838 , -0.9561598 , -0.97038996, -0.8865989 , -0.90038055, -0.9112894 , -0.85963196, -0.8275642 , -0.9644685 , -0.9541344 , -0.99113625, -0.96167463, -0.9249519 , -0.9394911 , -0.94370735, -0.8779986 , -0.8679626 , -0.8693049 , -0.88739663, -0.8532517 , -0.8839778 , -0.8832549 , -0.90802675, -0.87347734, -0.8799753 , -0.9015929 , -0.90027386, -0.89743274, -0.8957918 , -0.94940215, -0.878533 , -0.8786606 , -0.8682962 , -0.88305515, -0.8918906 , -0.9526935 , -0.95252645, -0.8589482 , -0.9665897 , -0.95774794, -0.91616255, -0.8337833 , -1.0229919 , -1.0324087 , -0.8770212 , -0.8826374 , -0.87960297, -0.86618614, -0.86750025, -0.84657943, -0.9200025 , -0.93132174, -0.9325887 , -0.91969943, -0.94920444, -0.8478174 , -0.83915174, -0.8423631 , -0.84093046, -0.88933957, -0.887299 , -0.88258964, -0.9021181 , -0.90000653, -0.9008188 , -0.8989647 , -0.8986879 , -0.9021104 , -0.9038444 , -0.90090567, -0.9013481 , -0.8815742 , -0.88889325, -0.88856375, -0.9051248 , -0.91187584, -0.93239444, -0.8643482 , -0.830946 , -0.8658417 , -0.89771646, -0.8830005 , -0.8855662 , -0.90893894, -0.9003961 , -0.9337367 , -0.9121187 , -0.909395 , -0.930227 , -0.8870004 , -0.85493207, -0.85663235, -0.9210262 , -0.91711617, -0.9092481 , -0.92222625, -0.9220136 , -0.92640096, -0.9133849 , -0.9117228 , -0.91316247, -0.90390074, -0.8708623 , -0.8658044 , -0.88882035, -0.89126635, -0.89049387, -0.8796412 , -0.8159574 , -0.8199297 , -0.8685883 , -0.86478555, -0.86581624, -0.8702363 , -0.8777868 , -0.8225077 , -0.80064034, -0.9007618 , -0.8379441 , -0.787583 , -0.992449 , -0.8551429 , -0.85392106, -0.86111844, -0.85006046, -0.85455143, -0.8931118 , -0.8860228 , -0.9259205 , -0.9213139 , -0.9276205 , -0.9124331 , -0.9460579 , -0.9215458 , -0.9654869 , -0.96105343, -0.90376806, -0.91599715, -0.9038252 , -0.84679633, -0.8942748 , -0.9174948 , -0.8939622 , -0.9095015 , -0.91820943, -0.9433106 , -0.9333364 , -0.93639815, -0.9020475 , -0.94818366, -0.88327503, -0.8717114 , -0.9163115 ], dtype=float32), 'β5_memorial': DeviceArray([-1.2497492, -1.2558298, -1.2491478, -1.2611171, -1.2290733, -1.2321496, -1.145841 , -1.145111 , -1.1803838, -1.188772 , -1.162937 , -1.174526 , -1.1747129, -1.1571174, -1.1417644, -1.1711872, -1.1636744, -1.2126799, -1.2503695, -1.2510941, -1.2823303, -1.2449069, -1.2793278, -1.1570547, -1.1721781, -1.2646953, -1.2907826, -1.2799145, -1.281673 , -1.3126476, -1.3150351, -1.2813773, -1.116992 , -1.215999 , -1.1982043, -1.2149308, -1.165123 , -1.2029762, -1.2340522, -1.1080865, -1.1430879, -1.1395249, -1.129469 , -1.2271845, -1.1979561, -1.1559327, -1.2294707, -1.1949136, -1.2471505, -1.237333 , -1.1016897, -1.2809657, -1.0865926, -1.3335369, -1.2965559, -1.1107098, -1.1790837, -1.2197542, -1.1730356, -1.1464381, -1.0527769, -1.0628942, -1.2788405, -1.2708672, -1.2433745, -1.2402942, -1.1758095, -1.185512 , -1.1811388, -1.1867927, -1.2531286, -1.1801386, -1.1483921, -1.1589415, -1.15567 , -1.1542312, -1.1537869, -1.2292153, -1.2284573, -1.2050482, -1.231881 , -1.2520047, -1.1474847, -1.1185743, -1.2117682, -1.1982377, -1.2184768, -1.1708355, -1.0831771, -1.290286 , -1.2763194, -1.0520995, -1.0567672, -1.1412413, -1.1687002, -1.1666651, -1.1979454, -1.1605471, -1.1133792, -1.1034732, -1.1087041, -1.21979 , -1.2167898, -1.2445152, -1.2591039, -1.2623708, -1.2572569, -1.2454056, -1.2431785, -1.250246 , -1.2494617, -1.2640896, -1.2621006, -1.2630111, -1.2640401, -1.2615246, -1.11996 , -1.1192317, -1.2908705, -1.234246 , -1.231116 , -1.2271416, -1.2397218, -1.1879256, -1.263524 , -1.1465526, -1.1544713, -1.1349306, -1.1321813, -1.1249273, -1.1077168, -1.1393168, -1.2160715, -1.1724843, -1.1521741, -1.2468389, -1.2539723, -1.1713587, -1.176647 , -1.1458688, -1.1485912, -1.1711756, -1.2152859, -1.23034 , -1.2246088, -1.232088 , -1.2344108, -1.2338476, -1.2436762, -1.2393297, -1.2477502, -1.1547481, -1.159483 , -1.1739151, -1.1460958, -1.1441855, -1.1464311, -1.190856 , -1.1867414, -1.1683557, -1.1651429, -1.1246363, -1.2142823, -1.2223155, -1.2247417, -1.1703734, -1.1373782, -1.2777249, -1.1297836, -1.0702503, -1.1022397, -1.1012455, -1.1009398, -1.0648389, -1.1083899, -1.3179376, -1.317645 , -1.2871469, -1.129949 , -1.2728837, -1.1644554, -1.1538652, -1.1640104, -1.2237648, -1.2297416, -1.2193412, -1.1740324, -1.0948089, -1.1155593, -1.2632071, -1.2719038, -1.1913456, -1.2578521, -1.1952434, -1.207996 , -1.2062765, -1.272438 , -1.1161267, -1.0902779, -1.2756807], dtype=float32), 'β5_thanksgiving': DeviceArray([-1.2912065, -1.2853844, -1.2858827, -1.2805132, -1.1840191, -1.2467893, -1.3332982, -1.3521019, -1.1331464, -1.1269491, -1.1364444, -1.1835064, -1.2915336, -1.2768314, -1.2578129, -1.3348229, -1.3166715, -1.2541313, -1.2340806, -1.2275896, -1.2317922, -1.2693607, -1.2555742, -1.2723687, -1.2523328, -1.2160906, -1.2272304, -1.2371948, -1.239447 , -1.2608297, -1.2562796, -1.2302587, -1.2538621, -1.3013253, -1.3274953, -1.267262 , -1.2755101, -1.3008424, -1.2929671, -1.2619581, -1.2519296, -1.2329335, -1.2474003, -1.2390668, -1.2870073, -1.305569 , -1.2313657, -1.2952102, -1.3350405, -1.3627489, -1.1913983, -1.2559513, -1.3248631, -1.2564191, -1.245345 , -1.2859777, -1.2845305, -1.2882111, -1.2664598, -1.2923157, -1.2888643, -1.2797042, -1.2799814, -1.2754813, -1.2618895, -1.2450371, -1.2260146, -1.2802685, -1.281176 , -1.2660514, -1.2456888, -1.278034 , -1.3400381, -1.334999 , -1.3329939, -1.3272707, -1.2649368, -1.2419959, -1.241668 , -1.2338262, -1.2213867, -1.2191037, -1.2867527, -1.3051507, -1.2940884, -1.2884061, -1.3124903, -1.2891103, -1.2877511, -1.2824945, -1.2886626, -1.2811128, -1.2853903, -1.2738348, -1.2812479, -1.2795949, -1.2971599, -1.294621 , -1.3256202, -1.3212558, -1.3144082, -1.302047 , -1.3165848, -1.319723 , -1.3245635, -1.3211728, -1.3386734, -1.3242638, -1.3307241, -1.3379952, -1.3389513, -1.3337901, -1.3346905, -1.3345402, -1.3349653, -1.34543 , -1.2104373, -1.2121803, -1.3297064, -1.30039 , -1.2932082, -1.2874278, -1.292627 , -1.2514502, -1.2377005, -1.3086833, -1.3208246, -1.3136237, -1.2462364, -1.24498 , -1.27632 , -1.278083 , -1.2514446, -1.3110495, -1.3276858, -1.3257644, -1.293669 , -1.2808068, -1.2798065, -1.255792 , -1.2686998, -1.2639989, -1.2689403, -1.2834069, -1.2797126, -1.2651469, -1.2647194, -1.2664324, -1.2535881, -1.2731448, -1.2671177, -1.294062 , -1.2908038, -1.2857578, -1.2827847, -1.2886126, -1.290324 , -1.3125848, -1.315012 , -1.3319623, -1.3309103, -1.3152875, -1.3470109, -1.3481104, -1.2727739, -1.311444 , -1.327863 , -1.196557 , -1.3246753, -1.3474746, -1.3500147, -1.3441137, -1.3461131, -1.3368045, -1.3658355, -1.2536998, -1.2584862, -1.2882155, -1.2878914, -1.337157 , -1.2833825, -1.2544005, -1.259538 , -1.2621917, -1.2759392, -1.3123102, -1.2911363, -1.3589498, -1.3550626, -1.1938491, -1.2464436, -1.2624991, -1.2644324, -1.2323657, -1.234984 , -1.2325617, -1.3236203, -1.3216622, -1.2831839, -1.2630209], dtype=float32), 'β_week': DeviceArray([[ 0.64273924, 0.37591752, 0.33642012, 0.46567297, -0.8753445 , -1.2303711 ], [ 0.60269666, 0.34509683, 0.30699566, 0.4431114 , -0.8171725 , -1.1459669 ], [ 0.5867868 , 0.3459284 , 0.31026855, 0.44436553, -0.8122369 , -1.153937 ], ..., [ 0.6167569 , 0.3711478 , 0.32794264, 0.43508238, -0.8405107 , -1.1750915 ], [ 0.6312179 , 0.38396534, 0.32433102, 0.45137066, -0.8591447 , -1.2023954 ], [ 0.6783961 , 0.42686486, 0.37456468, 0.51191497, -0.94996876, -1.3550221 ]], dtype=float32), 'λ': DeviceArray([[ 72.502 , 62.626015 , 6.939151 , ..., 16.766533 , 31.56063 , 8.222864 ], [ 68.41814 , 145.91295 , 13.544937 , ..., 25.374376 , 34.29928 , 6.5847363 ], [ 62.213455 , 140.9496 , 14.479413 , ..., 26.236351 , 33.69487 , 6.9440837 ], ..., [137.32362 , 32.311665 , 4.718723 , ..., 31.125721 , 11.104366 , 0.29069778], [120.25457 , 23.686852 , 5.15542 , ..., 36.679302 , 14.680668 , 0.16458689], [ 69.873505 , 102.66788 , 15.986472 , ..., 17.282774 , 85.37081 , 1.0387183 ]], dtype=float32), 'μ': DeviceArray([[-1.4335164 , -0.85548687, -0.23813926, ..., 1.9660449 , 2.1587765 , -0.24657345], [-1.3220237 , -0.8740853 , -0.21043721, ..., 1.963897 , 2.2186117 , -0.30884773], [-1.3316553 , -0.92171335, -0.22811076, ..., 1.938937 , 2.1859014 , -0.3086815 ], ..., [-1.3367463 , -0.914127 , -0.2095156 , ..., 1.9580991 , 2.068829 , -0.36293373], [-1.4639173 , -0.86541724, -0.19740029, ..., 1.9131397 , 2.0713031 , -0.32883397], [-1.3680247 , -0.9753535 , -0.25941002, ..., 1.9991653 , 2.272078 , -0.31308857]], dtype=float32), 'ρ1': DeviceArray([0.34706438, 0.3352192 , 0.33921126, 0.3506392 , 0.31589702, 0.27441388, 0.27775303, 0.2909958 , 0.35447028, 0.33187002, 0.3284175 , 0.29571605, 0.29095864, 0.2938913 , 0.30277976, 0.3098187 , 0.31561032, 0.26168773, 0.2512251 , 0.23890093, 0.25420058, 0.26597643, 0.25912002, 0.2594984 , 0.33000344, 0.3501781 , 0.3404681 , 0.34666133, 0.34514198, 0.34723094, 0.3488388 , 0.36597222, 0.42241603, 0.41575322, 0.41002145, 0.3989416 , 0.40888974, 0.3967265 , 0.38358435, 0.42140236, 0.4107465 , 0.4194709 , 0.4189138 , 0.3865917 , 0.42756453, 0.44017738, 0.38840464, 0.36457294, 0.36071095, 0.35967514, 0.32902908, 0.31816274, 0.34641156, 0.28762755, 0.28745875, 0.31582797, 0.32129714, 0.29775408, 0.27575362, 0.32392326, 0.33565712, 0.32875326, 0.26768357, 0.25290897, 0.2379358 , 0.26218447, 0.24548264, 0.24180187, 0.24033295, 0.24607383, 0.27099463, 0.28684983, 0.29945135, 0.30136204, 0.29790616, 0.29444262, 0.27229905, 0.2644242 , 0.26483417, 0.2622201 , 0.2593848 , 0.25373417, 0.23907194, 0.23885 , 0.21805139, 0.2284887 , 0.22568713, 0.19663577, 0.1961632 , 0.09624879, 0.09273873, 0.12656766, 0.12715758, 0.13599372, 0.14233476, 0.13757141, 0.14884928, 0.14225323, 0.17201209, 0.17145036, 0.16203083, 0.1559504 , 0.15853934, 0.16744578, 0.1649254 , 0.16536623, 0.19615723, 0.19601926, 0.19911653, 0.16951883, 0.17108235, 0.17089142, 0.17297079, 0.17294022, 0.17750803, 0.17521751, 0.2133363 , 0.21388169, 0.24871948, 0.2320983 , 0.23097676, 0.21516058, 0.21941759, 0.2628687 , 0.265031 , 0.25873846, 0.2590696 , 0.2918774 , 0.28368455, 0.28441197, 0.27618736, 0.2936741 , 0.30159828, 0.26974803, 0.23924637, 0.2320375 , 0.24188142, 0.21681377, 0.21519239, 0.3426475 , 0.3377651 , 0.34476787, 0.3436372 , 0.33444497, 0.3349494 , 0.32433534, 0.32623708, 0.3329968 , 0.3321579 , 0.3356542 , 0.33181056, 0.36535463, 0.3672039 , 0.37279883, 0.3934795 , 0.40177706, 0.4027956 , 0.38468942, 0.38527822, 0.38585305, 0.38880354, 0.38003364, 0.38268623, 0.3991906 , 0.41570497, 0.42326552, 0.41494584, 0.31635264, 0.3731703 , 0.3720796 , 0.37948135, 0.3795013 , 0.37898803, 0.3686445 , 0.38131952, 0.30479193, 0.3097082 , 0.3278635 , 0.38757935, 0.42123678, 0.39603153, 0.39800775, 0.39880797, 0.38117486, 0.3900452 , 0.39516965, 0.389528 , 0.38604236, 0.37694797, 0.34789187, 0.28018346, 0.26412845, 0.23718825, 0.22301231, 0.22324 , 0.25096908, 0.20729385, 0.2132658 , 0.23928137, 0.22625771], dtype=float32), 'ρ2': DeviceArray([0.21673423, 0.2303456 , 0.22659536, 0.23791257, 0.22796902, 0.21279725, 0.19952582, 0.19750573, 0.22060333, 0.21540147, 0.23041247, 0.22887215, 0.21683376, 0.21580692, 0.2121938 , 0.20449632, 0.21507186, 0.22269034, 0.22643118, 0.22518186, 0.22043131, 0.21638863, 0.21224621, 0.17775653, 0.19357644, 0.20329128, 0.21117157, 0.21127619, 0.21021529, 0.20363577, 0.20182289, 0.20043148, 0.20912293, 0.17996955, 0.17693397, 0.18436642, 0.17759953, 0.18165515, 0.17711043, 0.20186985, 0.20386286, 0.19636562, 0.20377794, 0.2053461 , 0.2018716 , 0.20445642, 0.17415829, 0.20430703, 0.20167665, 0.20116125, 0.21476974, 0.20053294, 0.2162081 , 0.20969991, 0.21023601, 0.20668666, 0.20651326, 0.20436423, 0.21045901, 0.21618423, 0.21659946, 0.21787003, 0.20932242, 0.19303752, 0.20702103, 0.20757717, 0.2034655 , 0.21470033, 0.21508975, 0.19040799, 0.188826 , 0.19942398, 0.20494007, 0.20486529, 0.20322858, 0.20178294, 0.19353992, 0.2028468 , 0.2029242 , 0.1992178 , 0.20133628, 0.20287777, 0.17408413, 0.17520581, 0.19370647, 0.18881157, 0.19041036, 0.1868224 , 0.18485394, 0.17365056, 0.18538304, 0.2031722 , 0.2063173 , 0.20264794, 0.19153595, 0.19119914, 0.20033301, 0.17314738, 0.18400005, 0.18608202, 0.18762723, 0.19819188, 0.19814034, 0.20195962, 0.19867936, 0.19821212, 0.20078783, 0.19624184, 0.19877227, 0.2001378 , 0.20041099, 0.19696178, 0.19657178, 0.19675726, 0.19910012, 0.20025767, 0.19748347, 0.1972607 , 0.1856773 , 0.17918408, 0.1812619 , 0.18564036, 0.1824787 , 0.18700036, 0.19482608, 0.18033902, 0.17650425, 0.18426506, 0.1534907 , 0.1585094 , 0.15719756, 0.16087884, 0.19044206, 0.1777655 , 0.18445934, 0.1810136 , 0.18278441, 0.17667338, 0.17629316, 0.18666705, 0.1795812 , 0.1746223 , 0.1854041 , 0.1814739 , 0.18185394, 0.18208417, 0.18280664, 0.1840229 , 0.18212892, 0.17447445, 0.17591664, 0.18922746, 0.18975742, 0.18933688, 0.19509444, 0.20117912, 0.20121552, 0.20149025, 0.20153452, 0.19120927, 0.19045824, 0.20047021, 0.18148437, 0.16939208, 0.19147606, 0.18884216, 0.19339673, 0.1963772 , 0.19792055, 0.19081007, 0.19181988, 0.1930454 , 0.19422244, 0.18067402, 0.19305298, 0.20128831, 0.19109611, 0.20750986, 0.22683243, 0.21183577, 0.21244152, 0.20982772, 0.2124917 , 0.19665138, 0.21378031, 0.19708225, 0.21020935, 0.20268641, 0.20757243, 0.19518898, 0.1804057 , 0.19785543, 0.18389206, 0.19536608, 0.19444062, 0.20741954, 0.15098226, 0.19783367, 0.19990698, 0.19994971], dtype=float32), 'ρ3': DeviceArray([0.38542235, 0.3651004 , 0.37595525, 0.36979982, 0.47805992, 0.3601256 , 0.39307117, 0.33943427, 0.14421889, 0.1323007 , 0.15643558, 0.14112605, 0.13019831, 0.13490038, 0.14495069, 0.13779601, 0.13572457, 0.14081661, 0.14186022, 0.14415745, 0.16553535, 0.16910644, 0.15778324, 0.26333028, 0.2239265 , 0.30585575, 0.25434968, 0.2562155 , 0.25565186, 0.24588259, 0.24749316, 0.253579 , 0.28695616, 0.23623496, 0.2887429 , 0.29343903, 0.35279015, 0.33373395, 0.3606613 , 0.3711966 , 0.38839358, 0.39651492, 0.40050253, 0.384892 , 0.32651886, 0.32184035, 0.3944019 , 0.35403955, 0.34058982, 0.33118334, 0.32612008, 0.37119728, 0.4181981 , 0.34384555, 0.36327353, 0.44449544, 0.4712061 , 0.46136236, 0.5546728 , 0.57605726, 0.61387503, 0.6144784 , 0.5359837 , 0.5573027 , 0.580486 , 0.6647447 , 0.7083867 , 0.69499654, 0.70503414, 0.43732 , 0.47195318, 0.39981604, 0.3679118 , 0.36850366, 0.37188888, 0.36823475, 0.34722984, 0.58531034, 0.5866436 , 0.60886985, 0.6175734 , 0.5781819 , 0.52993304, 0.5282708 , 0.42381936, 0.49032274, 0.47433156, 0.45883372, 0.50910854, 0.48775297, 0.47672188, 0.57698375, 0.6037404 , 0.7389846 , 0.75442296, 0.7590955 , 0.72890466, 0.8106725 , 0.7670484 , 0.76626635, 0.7334112 , 0.7247586 , 0.5646848 , 0.56579244, 0.5788322 , 0.5739828 , 0.58154 , 0.6001709 , 0.5845343 , 0.57364976, 0.5765079 , 0.55849886, 0.5516462 , 0.5520846 , 0.5475442 , 0.54625607, 0.80586326, 0.81212 , 0.78620595, 0.94815695, 0.9285243 , 0.8269818 , 0.842043 , 1.028476 , 1.0400169 , 1.0647984 , 0.9986073 , 1.3044196 , 1.1973485 , 1.1689965 , 1.2011597 , 1.2653837 , 0.94481015, 1.0599902 , 1.0985664 , 1.0561833 , 0.99089545, 1.0695802 , 1.0660765 , 1.0810069 , 1.0852343 , 1.1181467 , 1.148317 , 1.1620392 , 1.1637905 , 1.1977587 , 1.1951233 , 1.194494 , 1.2065823 , 1.1823168 , 1.1750066 , 1.0966877 , 1.1028194 , 1.0952212 , 1.033758 , 1.0152864 , 1.0185583 , 1.1589917 , 1.1818727 , 1.2051135 , 1.1835196 , 1.1933377 , 1.307939 , 1.2659702 , 1.2763153 , 1.3458002 , 1.2841748 , 1.1474612 , 0.9175341 , 0.8630463 , 0.85918903, 0.8505507 , 0.8499827 , 0.8243371 , 0.86381173, 0.5865283 , 0.60571367, 0.5982061 , 0.6640022 , 0.7340557 , 1.0453389 , 1.1428595 , 1.1100714 , 0.9636378 , 0.94585925, 0.96718806, 1.0306013 , 0.92317575, 0.9214294 , 0.9571524 , 1.0028392 , 1.1645598 , 1.0463401 , 1.028207 , 1.0162468 , 0.8695074 , 1.1365907 , 1.0731087 , 1.1762427 , 1.14842 ], dtype=float32), 'σ': DeviceArray([0.24712439, 0.24562451, 0.2459602 , 0.24541366, 0.24430592, 0.24607061, 0.24520826, 0.24544126, 0.24441601, 0.24403921, 0.24579567, 0.24471535, 0.24527693, 0.24211918, 0.24394788, 0.24700896, 0.24546361, 0.245562 , 0.24752267, 0.246684 , 0.24574636, 0.24634168, 0.24793194, 0.24243036, 0.24209784, 0.24484564, 0.24428043, 0.24454796, 0.24448647, 0.24531223, 0.24512169, 0.24501365, 0.24460083, 0.24485698, 0.24565652, 0.24815609, 0.24601603, 0.24365714, 0.24440998, 0.24806447, 0.24511673, 0.24534844, 0.24582154, 0.24311526, 0.24613261, 0.24429823, 0.2459068 , 0.24549673, 0.2457168 , 0.24483411, 0.24555822, 0.24327956, 0.24851322, 0.24073176, 0.24057628, 0.25108317, 0.24895371, 0.24697165, 0.24544294, 0.24413836, 0.24584657, 0.24595083, 0.24457 , 0.24570711, 0.24396801, 0.24581525, 0.24468504, 0.24622634, 0.24627535, 0.2426381 , 0.24668738, 0.2458696 , 0.24474686, 0.244556 , 0.24463214, 0.24484058, 0.24641332, 0.2445111 , 0.24450517, 0.24392854, 0.2437225 , 0.24352242, 0.24940369, 0.24864282, 0.24427086, 0.24474765, 0.24588563, 0.24796218, 0.24422808, 0.24429569, 0.24398196, 0.2466814 , 0.24682882, 0.24324638, 0.24496768, 0.24539402, 0.24598601, 0.2475422 , 0.2425912 , 0.24296474, 0.2439889 , 0.24542218, 0.24152444, 0.24181046, 0.24144602, 0.24166314, 0.24500151, 0.24462622, 0.2446781 , 0.24112928, 0.24099995, 0.24115466, 0.24116091, 0.24115962, 0.24114753, 0.2417057 , 0.24762784, 0.24744152, 0.24169248, 0.24670978, 0.24632882, 0.24658848, 0.24667795, 0.2423034 , 0.24473949, 0.24655722, 0.24775565, 0.24218684, 0.24378487, 0.24280387, 0.24397083, 0.24838352, 0.2414646 , 0.24873723, 0.24272566, 0.24827491, 0.24503681, 0.24774632, 0.24781343, 0.24443235, 0.2468569 , 0.24568303, 0.24663523, 0.24626338, 0.24598645, 0.2457397 , 0.24585325, 0.24570063, 0.24522369, 0.24370997, 0.24324892, 0.24777535, 0.24782132, 0.24723491, 0.24790882, 0.24748082, 0.24740839, 0.24330646, 0.2429565 , 0.24468991, 0.24530643, 0.24245645, 0.2435771 , 0.24449508, 0.24629045, 0.24407841, 0.24817146, 0.24456067, 0.24460395, 0.24308944, 0.24300611, 0.24336278, 0.24321936, 0.24214329, 0.24406828, 0.2470091 , 0.24496035, 0.2439964 , 0.24677344, 0.24353781, 0.2452157 , 0.2460108 , 0.24620399, 0.24559799, 0.24635462, 0.2448688 , 0.24741626, 0.24321155, 0.24444261, 0.2433432 , 0.24587324, 0.24383652, 0.24649745, 0.2448723 , 0.24486049, 0.24423496, 0.24409585, 0.24321675, 0.24299085, 0.24472779], dtype=float32), 'τ': DeviceArray([0.01116914, 0.01059679, 0.01064383, 0.0109238 , 0.01182803, 0.01130889, 0.01132126, 0.01265572, 0.0130745 , 0.0125589 , 0.01278759, 0.01444284, 0.01283798, 0.01318507, 0.01371803, 0.01332116, 0.01391815, 0.01460021, 0.01399529, 0.01302744, 0.0135108 , 0.01351917, 0.01316967, 0.0136638 , 0.01526458, 0.01306709, 0.01366668, 0.01351362, 0.01347828, 0.01267754, 0.01268692, 0.0129271 , 0.01213905, 0.01196666, 0.01290006, 0.01219959, 0.01305656, 0.01247589, 0.01310758, 0.01085014, 0.01126369, 0.01164215, 0.01091566, 0.0101073 , 0.00773612, 0.00928911, 0.01042034, 0.00846595, 0.00971272, 0.00974922, 0.0107684 , 0.01009322, 0.00964624, 0.01024828, 0.0099438 , 0.01000259, 0.01027995, 0.00880891, 0.00969005, 0.00834786, 0.00805798, 0.00835952, 0.00865725, 0.00834359, 0.00839735, 0.00849188, 0.00837943, 0.00898642, 0.00897508, 0.00819306, 0.00875456, 0.00818559, 0.00919791, 0.00918698, 0.00905153, 0.00897259, 0.00721512, 0.00754693, 0.00757488, 0.00781822, 0.00811914, 0.0083743 , 0.00925336, 0.00937009, 0.01028179, 0.00991045, 0.00931695, 0.00892676, 0.00855242, 0.01136933, 0.01150645, 0.01048808, 0.01049663, 0.0103258 , 0.01066439, 0.01055769, 0.01053836, 0.00949427, 0.0099148 , 0.00996285, 0.01015613, 0.00946222, 0.00974287, 0.00990673, 0.00976126, 0.00974688, 0.00954588, 0.00952678, 0.00957264, 0.00936561, 0.00934143, 0.00936524, 0.00930687, 0.00930316, 0.00921883, 0.0090611 , 0.00887125, 0.00886176, 0.0089979 , 0.01037043, 0.01029399, 0.01030701, 0.01029132, 0.01346441, 0.01210377, 0.01197209, 0.01181228, 0.01276465, 0.01420293, 0.01391546, 0.01404381, 0.01409464, 0.01180979, 0.01176839, 0.01213444, 0.01234511, 0.00998182, 0.0106769 , 0.01067608, 0.01174101, 0.01149755, 0.0119192 , 0.01159975, 0.01140781, 0.01135812, 0.01152831, 0.01150372, 0.01149688, 0.01133561, 0.01135073, 0.01188905, 0.01469594, 0.01461065, 0.01434913, 0.01536949, 0.01455125, 0.01462778, 0.01371187, 0.01339297, 0.01490703, 0.01477796, 0.01390325, 0.0148015 , 0.01882763, 0.01776791, 0.0156221 , 0.01600355, 0.01957637, 0.01637322, 0.01598678, 0.01528125, 0.01502175, 0.01512914, 0.01488372, 0.01493131, 0.01339893, 0.01393087, 0.01447955, 0.01632412, 0.01354388, 0.01613444, 0.01583788, 0.01639266, 0.01345641, 0.01424601, 0.0141243 , 0.01384322, 0.01586977, 0.01495008, 0.01356761, 0.01418791, 0.01439383, 0.01773744, 0.01796618, 0.01789748, 0.0167682 , 0.01607374, 0.01444568, 0.01438913, 0.01450074], dtype=float32)})