some doc changes, requirements in setup.py changes, closing #679

docs/lifelines.fitters.rst

@@ -8,87 +8,76 @@ lifelines.fitters.aalen\_additive\_fitter module
 .. automodule:: lifelines.fitters.aalen_additive_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.aalen\_johansen\_fitter module
 ------------------------------------------------
 
 .. automodule:: lifelines.fitters.aalen_johansen_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.breslow\_fleming\_harrington\_fitter module
 -------------------------------------------------------------
 
 .. automodule:: lifelines.fitters.breslow_fleming_harrington_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.cox\_time\_varying\_fitter module
 ---------------------------------------------------
 
 .. automodule:: lifelines.fitters.cox_time_varying_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.coxph\_fitter module
 --------------------------------------
 
 .. automodule:: lifelines.fitters.coxph_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.exponential\_fitter module
 --------------------------------------------
 
 .. automodule:: lifelines.fitters.exponential_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.kaplan\_meier\_fitter module
 ----------------------------------------------
 
 .. automodule:: lifelines.fitters.kaplan_meier_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.log\_logistic\_fitter module
 ----------------------------------------------
 
 .. automodule:: lifelines.fitters.log_logistic_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.log\_normal\_fitter module
 --------------------------------------------
 
 .. automodule:: lifelines.fitters.log_normal_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.nelson\_aalen\_fitter module
 ----------------------------------------------
 
 .. automodule:: lifelines.fitters.nelson_aalen_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.piecewise\_exponential\_fitter module
 -------------------------------------------------------
 
 .. automodule:: lifelines.fitters.piecewise_exponential_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 
 
@@ -98,7 +87,6 @@ lifelines.fitters.weibull\_fitter module
 .. automodule:: lifelines.fitters.weibull_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 
 lifelines.fitters.weibull\_aft\_fitter module
@@ -107,28 +95,24 @@ lifelines.fitters.weibull\_aft\_fitter module
 .. automodule:: lifelines.fitters.weibull_aft_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.log\_normal\_aft\_fitter module
 ---------------------------------------------------
 
 .. automodule:: lifelines.fitters.log_normal_aft_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 lifelines.fitters.log\_logistic\_aft\_fitter module
 -----------------------------------------------------
 
 .. automodule:: lifelines.fitters.log_logistic_aft_fitter
     :members:
     :undoc-members:
-    :show-inheritance:
 
 
 
 
 .. automodule:: lifelines.fitters
     :members:
     :undoc-members:
-    :show-inheritance:

lifelines/__init__.py

@@ -36,4 +36,5 @@
     "WeibullAFTFitter",
     "LogLogisticAFTFitter",
     "LogNormalAFTFitter",
+    "PiecewiseExponentialFitter",
 ]

lifelines/fitters/__init__.py

@@ -34,6 +34,7 @@
     string_justify,
     format_floats,
     format_p_value,
+    format_exp_floats,
     coalesce,
     check_nans_or_infs,
     pass_for_numeric_dtypes_or_raise_array,
@@ -303,7 +304,7 @@ def __init__(self, *args, **kwargs):
 
         if len(self._bounds) != len(self._fitted_parameter_names) != self._initial_values.shape[0]:
             raise ValueError(
-                "_bounds must be the same shape as _fitted_parameters_names must be the same shape as _initial_values"
+                "_bounds must be the same shape as _fitted_parameter_names must be the same shape as _initial_values"
             )
 
     def _check_cumulative_hazard_is_monotone_and_positive(self, durations, values):
@@ -598,7 +599,12 @@ def print_summary(self, decimals=2, **kwargs):
         print("---")
 
         df = self.summary
-        print(df.to_string(float_format=format_floats(decimals), formatters={"p": format_p_value(decimals)}))
+        print(
+            df.to_string(
+                float_format=format_floats(decimals),
+                formatters={"p": format_p_value(decimals), "exp(coef)": format_exp_floats()},
+            )
+        )
 
     def fit(
         self,
@@ -1336,7 +1342,12 @@ def print_summary(self, decimals=2, **kwargs):
 
         df = self.summary
         # Significance codes as last column
-        print(df.to_string(float_format=format_floats(decimals), formatters={"p": format_p_value(decimals)}))
+        print(
+            df.to_string(
+                float_format=format_floats(decimals),
+                formatters={"p": format_p_value(decimals), "exp(coef)": format_exp_floats()},
+            )
+        )
 
         # Significance code explanation
         print("---")

lifelines/fitters/aalen_additive_fitter.py

@@ -26,6 +26,7 @@
     _to_list,
     format_floats,
     format_p_value,
+    format_exp_floats,
     survival_table_from_events,
     StatisticalWarning,
 )
@@ -563,7 +564,12 @@ def print_summary(self, decimals=2, **kwargs):
         print("---")
 
         df = self.summary
-        print(df.to_string(float_format=format_floats(decimals), formatters={"p": format_p_value(decimals)}))
+        print(
+            df.to_string(
+                float_format=format_floats(decimals),
+                formatters={"p": format_p_value(decimals), "exp(coef)": format_exp_floats()},
+            )
+        )
 
         # Significance code explanation
         print("---")

lifelines/fitters/cox_time_varying_fitter.py

@@ -36,6 +36,7 @@
     check_nans_or_infs,
     string_justify,
     format_p_value,
+    format_exp_floats,
     format_floats,
     coalesce,
 )
@@ -644,7 +645,12 @@ def print_summary(self, decimals=2, **kwargs):
 
         df = self.summary
         # Significance codes last
-        print(df.to_string(float_format=format_floats(decimals), formatters={"p": format_p_value(decimals)}))
+        print(
+            df.to_string(
+                float_format=format_floats(decimals),
+                formatters={"p": format_p_value(decimals), "exp(coef)": format_exp_floats()},
+            )
+        )
 
         # Significance code explanation
         print("---")

lifelines/fitters/coxph_fitter.py

@@ -38,6 +38,7 @@
     string_justify,
     format_p_value,
     format_floats,
+    format_exp_floats,
     dataframe_interpolate_at_times,
 )
 
@@ -1257,7 +1258,12 @@ def print_summary(self, decimals=2, **kwargs):
 
         df = self.summary
         # Significance codes as last column
-        print(df.to_string(float_format=format_floats(decimals), formatters={"p": format_p_value(decimals)}))
+        print(
+            df.to_string(
+                float_format=format_floats(decimals),
+                formatters={"p": format_p_value(decimals), "exp(coef)": format_exp_floats()},
+            )
+        )
 
         # Significance code explanation
         print("---")

lifelines/fitters/exponential_fitter.py

@@ -23,16 +23,16 @@ class ExponentialFitter(KnownModelParametericUnivariateFitter):
     After calling the `.fit` method, you have access to properties like: ``survival_function_``, ``lambda_``, ``cumulative_hazard_``
     A summary of the fit is available with the method ``print_summary()``
 
+    Parameters
+    -----------
+    alpha: float, optional (default=0.05)
+        the level in the confidence intervals.
+
     Important
     ----------
     The parameterization of this model changed in lifelines 0.19.0. Previously, the cumulative hazard looked like
     :math:`\lambda t`. The parameterization is now the reciprocal of :math:`\lambda`.
 
-
-    Notes
-    -----
-    Reference: https://www4.stat.ncsu.edu/~dzhang2/st745/chap3.pdf
-
     Attributes
     ----------
     cumulative_hazard_ : DataFrame

lifelines/fitters/log_logistic_fitter.py

@@ -23,6 +23,11 @@ class LogLogisticFitter(KnownModelParametericUnivariateFitter):
     After calling the `.fit` method, you have access to properties like: ``cumulative_hazard_``, ``plot``, ``survival_function_``, ``alpha_`` and ``beta_``.
     A summary of the fit is available with the method 'print_summary()'
 
+    Parameters
+    -----------
+    alpha: float, optional (default=0.05)
+        the level in the confidence intervals.
+
     Examples
     --------
 

lifelines/fitters/log_normal_fitter.py

@@ -22,6 +22,12 @@ class LogNormalFitter(KnownModelParametericUnivariateFitter):
     After calling the `.fit` method, you have access to properties like: ``survival_function_``, ``mu_``, ``sigma_``.
     A summary of the fit is available with the method ``print_summary()``
 
+    Parameters
+    -----------
+    alpha: float, optional (default=0.05)
+        the level in the confidence intervals.
+
+
     Attributes
     ----------
     cumulative_hazard_ : DataFrame

lifelines/fitters/piecewise_exponential_fitter.py

@@ -22,6 +22,13 @@ class PiecewiseExponentialFitter(KnownModelParametericUnivariateFitter):
     After calling the `.fit` method, you have access to properties like: ``survival_function_``, ``plot``, ``cumulative_hazard_``
     A summary of the fit is available with the method ``print_summary()``
 
+    Parameters
+    -----------
+    breakpoints: list
+        a list of times when a new exponential model is constructed.
+    alpha: float, optional (default=0.05)
+        the level in the confidence intervals.
+
     Important
     ----------
     The parameterization of this model changed in lifelines 0.19.1. Previously, the cumulative hazard looked like
@@ -66,10 +73,10 @@ class PiecewiseExponentialFitter(KnownModelParametericUnivariateFitter):
 
     def __init__(self, breakpoints, *args, **kwargs):
         breakpoints = np.sort(breakpoints)
-        if not (breakpoints[-1] < np.inf):
+        if len(breakpoints) > 0 and not (breakpoints[-1] < np.inf):
             raise ValueError("Do not add inf to the breakpoints.")
 
-        if breakpoints[0] < 0:
+        if len(breakpoints) > 0 and breakpoints[0] < 0:
             raise ValueError("First breakpoint must be greater than 0.")
 
         self.breakpoints = np.append(breakpoints, [np.inf])
@@ -84,7 +91,6 @@ def _cumulative_hazard(self, params, times):
 
         n = times.shape[0]
         times = times.reshape((n, 1))
-
         bp = self.breakpoints
         M = np.minimum(np.tile(bp, (n, 1)), times)
         M = np.hstack([M[:, tuple([0])], np.diff(M, axis=1)])

lifelines/fitters/weibull_fitter.py

@@ -24,6 +24,10 @@ class WeibullFitter(KnownModelParametericUnivariateFitter):
     After calling the `.fit` method, you have access to properties like: ``cumulative_hazard_``, ``survival_function_``, ``lambda_`` and ``rho_``.
     A summary of the fit is available with the method ``print_summary()``.
 
+    Parameters
+    -----------
+    alpha: float, optional (default=0.05)
+        the level in the confidence intervals.
 
     Important
     ----------

lifelines/generate_datasets.py

@@ -9,6 +9,46 @@
 from scipy.integrate import cumtrapz
 
 
+def piecewise_exponential_survival_data(n, breakpoints, lambdas):
+    """
+
+    Note
+    --------
+    No censoring is present here.
+
+    Examples
+    --------
+
+    >>> T = piecewise_exponential_survival_data(100000, [1, 3], [0.2, 3, 1.])
+    >>> NelsonAalenFitter().fit(T).plot()
+
+    """
+    assert len(breakpoints) == len(lambdas) - 1
+
+    breakpoints = np.append([0], breakpoints)
+
+    delta_breakpoints = np.diff(breakpoints)
+
+    T = np.empty(n)
+    for i in range(n):
+        U = np.random.random()
+        E = -np.log(U)
+
+        running_sum = 0
+        for delta, lambda_, bp in zip(delta_breakpoints, lambdas, breakpoints):
+            factor = lambda_ * delta
+            if E < running_sum + factor:
+                t = bp + (E - running_sum) / lambda_
+                break
+            running_sum += factor
+        else:
+            t = breakpoints[-1] + (E - running_sum) / lambdas[-1]
+
+        T[i] = t
+
+    return T
+
+
 def exponential_survival_data(n, cr=0.05, scale=1.0):
 
     t = stats.expon.rvs(scale=scale, size=n)
@@ -273,7 +313,7 @@ def cumulative_integral(fx, x):
 def construct_survival_curves(hazard_rates, timelines):
     """
     Given hazard rates, reconstruct the survival curves
-    
+
     Parameters
     ----------
     hazard_rates: (n,t) array