How to use the pastas.decorators.model_tmin_tmax function in pastas

    @model_tmin_tmax
    def many(self, tmin=None, tmax=None, stats=None):
        """This method returns the values for a provided list of statistics.

        Parameters
        ----------
        tmin
        tmax
        stats: list
            list of statistics that need to be calculated.

        Returns
        -------

        """
        if not stats:
            stats = ['evp', 'rmse', 'rmsi', 'rsq']

    @model_tmin_tmax
    def plot(self, tmin=None, tmax=None, oseries=True, simulation=True,
             show=True, **kwargs):
        """Make a plot of the observed and simulated series.

        Parameters
        ----------
        oseries: Boolean
            True to plot the observed time series.
        simulation: Boolean
            True to plot the simulated time series.

        Returns
        -------
        fig: matplotlib.figure
            MPL figure with the simulated and optionally the observed time
            series.

    @model_tmin_tmax
    def rmse(self, tmin=None, tmax=None):
        """Root mean squared error of the residuals.

        Notes
        -----
        .. math:: rmse = sqrt(sum(residuals**2) / N)

        where N is the number of residuals.
        """
        res = self.ml.residuals(tmin=tmin, tmax=tmax).values
        N = res.size
        return np.sqrt((res ** 2).sum() / N)

    @model_tmin_tmax
    def summary(self, tmin=None, tmax=None, stats='basic'):
        """Prints a summary table of the model statistics. The set of
        statistics that are printed are stats by a dictionary of the desired
        statistics.

        Parameters
        ----------
        tmin
        tmax
        stats : str or dict
            dictionary of the desired statistics or a string with one of the
            predefined sets. Supported options are: 'basic', 'all', and 'dutch'

        Returns
        -------
        stats : Pandas.DataFrame

    @model_tmin_tmax
    def results(self, tmin=None, tmax=None, show=True):
        """Plot different results in one window to get a quick overview.

        Parameters
        ----------
        tmin/tmax: str
            start and end time for plotting


        Returns
        -------

        """
        fig = self._get_figure()

        #Number of row to make the figure with

    @model_tmin_tmax
    def rmsn(self, tmin=None, tmax=None):
        """Root mean squared error of the noise.

        Returns
        -------
        float or nan
            Return a float if noisemodel is present, nan if not.

        Notes
        -----
        .. math:: rmsi = \\sqrt{\\frac{\\sum(noise^2)}{N}}

        where N is the number of noise.

        """
        if not self.ml.noisemodel:

    @model_tmin_tmax
    def avg_dev(self, tmin=None, tmax=None):
        """Average deviation of the residuals.

        Notes
        -----
        .. math:: avg_dev = sum(E) / N

        Where N is the number of the residuals.

        """
        res = self.ml.residuals(tmin=tmin, tmax=tmax).values
        return res.mean()

    @model_tmin_tmax
    def rmsi(self, tmin=None, tmax=None):
        """Root mean squared error of the noise.

        Notes
        -----
        .. math:: rmsi = sqrt(sum(noise**2) / N)

        where N is the number of noise.
        """
        res = self.ml.noise(tmin=tmin, tmax=tmax).values
        N = res.size
        return np.sqrt((res ** 2).sum() / N)

    @model_tmin_tmax
    def bic(self, tmin=None, tmax=None):
        """Bayesian Information Criterium. The noise is used if a noisemodel is
         present, otherwise the residuals are used.

        Notes
        -----
        The Bayesian Information Criterium is calculated as follows:

        .. math:: BIC = -2 log(L) + nparam * log(N)

        Where nparam  is the number of free parameters

        """
        if self.ml.noisemodel:
            noise = self.ml.noise(tmin=tmin, tmax=tmax).values
        else:

    @model_tmin_tmax
    def all(self, tmin=None, tmax=None):
        """Returns a dictionary with all the statistics.

        Parameters
        ----------
        tmin: str
        tmax: str

        Returns
        -------
        stats: pd.DataFrame
            Dataframe with all possible statistics

        """
        stats = DataFrame(columns=['Value'])
        for k in self.ops.keys():