more docs

doc/source/tutorial.py

@@ -132,14 +132,14 @@ def as_table(features, values):
       <h4 class="panel-title">
         <a role="button" data-toggle="collapse" data-parent="#extractors" href="#collapse-feature-{{ name }}" aria-expanded="true" aria-controls="collapse-feature-{{ name }}">
           Extractor <span class="text-info">{{ name }}</span>
-        
+
          <span class="pull-right">
              {% for feature in fresume %}
              <span class="label lb-lg feature-resume label-default">{{ feature }}</span>
              {% endfor %}
          </span>
          </a>
-         
+
       </h4>
     </div>
     <div id="collapse-feature-{{ name }}" class="panel-collapse collapse" role="tabpanel" aria-labelledby="heading-feature-{{ name }}">
@@ -151,14 +151,14 @@ def as_table(features, values):
                  <span class="label label-info">{{ data }}</span>
              {% endfor %}
          </div>
-         
+
          <h5>Full list of features</h5>
          <div>
              {% for feature in ext.get_features() %}
              <span class="label label-default">{{ feature }}</span>
              {% endfor %}
          </div>
-         
+
          <h5>Dependencies</h5>
          <div>
          {% for dep in ext.get_dependencies() %}
@@ -178,11 +178,11 @@ def as_table(features, values):
 
 def features_doc():
     import feets
-    
+
     import rst2html5_
-    
+
     from docutils.core import publish_parts
-    
+
     rows = []
     extractors = sorted({
         e for e in feets.registered_extractors().values()})
@@ -191,15 +191,14 @@ def features_doc():
 
         doc = publish_parts(
             ext.__doc__ or name,
-            writer_name='html5', 
+            writer_name='html5',
             writer=rst2html5_.HTML5Writer())["body"]
-        
+
         features = ext.get_features()
         data = sorted(ext.get_data(), key=feets.extractors.DATAS.index)
-        
+
         if len(features) > 4:
             features = list(features)[:4] + ["..."]
         rows.append((name, doc, ext, features, data))
     rows.sort()
     return HTML(DOC_TEMPLATE.render(rows=rows))
-

feets/extractors/ext_car.py

@@ -112,24 +112,6 @@ def _car_like(parameters, t, x, error_vars):
 # =============================================================================
 
 class CAR(Extractor):
-    r"""In order to model the irregular sampled times series we use CAR(
-    (Brockwell and Davis, 2002), a continious time auto regressive model.
-    CAR process has three parameters, it provides a natural and consistent
-    way of estimating a characteristic time scale and variance of light-curves.
-    CAR process is described by the following stochastic differential
-    equation:
-
-    .. math::
-
-        dX(t) = - \frac{1}{\tau} X(t)dt + \sigma_C \sqrt{dt} \epsilon(t) + bdt,
-
-
-    .. math::
-
-        for \: \tau, \sigma_C, t \geq 0
-
-
-    """
 
     data = ['magnitude', 'time', 'error']
     features = ["CAR_sigma", "CAR_tau", "CAR_mean"]

feets/extractors/ext_gskew.py

@@ -52,7 +52,21 @@
 # =============================================================================
 
 class Gskew(Extractor):
-    """Median-based measure of the skew"""
+    """Median-of-magnitudes based measure of the skew.
+
+    .. math::
+
+        Gskew = m_{q3} + m_{q97} - 2m
+
+    Where:
+
+    - :math:`m_{q3}` is the median of magnitudes lesser or equal than the
+      quantile 3.
+    - :math:`m_{q97}` is the median of magnitudes greater or equal than the
+      quantile 97.
+    - :math:`m` is the median of magnitudes.
+
+    """
 
     data = ['magnitude']
     features = ["Gskew"]

feets/extractors/ext_linear_trend.py

@@ -52,6 +52,19 @@
 # =============================================================================
 
 class LinearTrend(Extractor):
+    r"""
+    **LinearTrend**
+
+    Slope of a linear fit to the light-curve.
+
+    .. code-block:: pycon
+
+        >>> fs = feets.FeatureSpace(only=['LinearTrend'])
+        >>> features, values = fs.extract(**lc_normal)
+        >>> dict(zip(features, values))
+        {'LinearTrend': -3.2084065290292509e-06}
+
+    """
 
     data = ['magnitude', 'time']
     features = ["LinearTrend"]

feets/extractors/ext_max_slope.py

@@ -53,15 +53,30 @@
 
 class MaxSlope(Extractor):
     """
+    **MaxSlope**
+
+    Maximum absolute magnitude slope between two consecutive observations.
+
     Examining successive (time-sorted) magnitudes, the maximal first difference
     (value of delta magnitude over delta time)
+
+    .. code-block:: pycon
+
+        >>> fs = feets.FeatureSpace(only=['MaxSlope'])
+        >>> features, values = fs.extract(**lc_normal)
+        >>> dict(zip(features, values))
+        {'MaxSlope': 5.4943105823904741}
+
     """
 
     data = ['magnitude', 'time']
     features = ["MaxSlope"]
+    params = {"timesort": True}
 
-    def fit(self, magnitude, time):
-        slope = np.abs(magnitude[1:] - magnitude[:-1]) / (time[1:] - time[:-1])
-        np.max(slope)
+    def fit(self, magnitude, time, timesort):
+        if timesort:
+            sort = np.argsort(time)
+            time, magnitude = time[sort], magnitude[sort]
 
+        slope = np.abs(magnitude[1:] - magnitude[:-1]) / (time[1:] - time[:-1])
         return {"MaxSlope": np.max(slope)}

feets/extractors/ext_mean.py

@@ -52,6 +52,20 @@
 # =============================================================================
 
 class Mean(Extractor):
+    r"""
+    **Mean**
+
+    Mean magnitude. For a normal distribution it should take a value
+    close to zero:
+
+    .. code-block:: pycon
+
+        >>> fs = feets.FeatureSpace(only=['Mean'])
+        >>> features, values = fs.extract(**lc_normal)
+        >>> dict(zip(features, values))
+        {'Mean': 0.0082611563419413246}
+
+    """
 
     data = ['magnitude']
     features = ["Mean"]

feets/extractors/ext_mean_variance.py

@@ -52,7 +52,26 @@
 # =============================================================================
 
 class MeanVariance(Extractor):
-    """variability index"""
+    """
+    **Meanvariance** (:math:`\frac{\sigma}{\bar{m}}`)
+
+    This is a simple variability index and is defined as the ratio of the
+    standard deviation :math:`\sigma`, to the mean magnitude, :math:`\bar{m}`.
+    If a light curve has strong  variability, :math:`\frac{\sigma}{\bar{m}}`
+    of the light curve is generally  large.
+
+    For a uniform distribution from 0 to 1, the mean is equal to 0.5 and the
+    variance is equal to 1/12, thus the mean-variance should take a value
+    close to 0.577:
+
+    .. code-block:: pycon
+
+        >>> fs = feets.FeatureSpace(only=['Meanvariance'])
+        >>> features, values = fs.extract(**lc_uniform)
+        >>> dict(zip(features, values))
+        {'Meanvariance': 0.5816791217381897}
+
+    """
 
     data = ['magnitude']
     features = ['Meanvariance']

feets/extractors/ext_skew.py

@@ -52,7 +52,29 @@
 # =============================================================================
 
 class Skew(Extractor):
-    """Skewness of the magnitudes"""
+    r"""
+
+    **Skew**
+
+    The skewness of a sample is defined as follow:
+
+    .. math::
+
+        Skewness = \frac{N}{(N-1)(N-2)}
+            \sum_{i=1}^N (\frac{m_i-\hat{m}}{\sigma})^3
+
+    Example:
+
+    For a normal distribution it should be equal to zero:
+
+    .. code-block:: pycon
+
+        >>> fs = feets.FeatureSpace(only=['Skew'])
+        >>> features, values = fs.extract(**lc_normal)
+        >>> dict(zip(features, values))
+        {'Skew': -0.00023325826785278685}
+
+    """
 
     data = ['magnitude']
     features = ["Skew"]