removed resample (#93)

* removed resample * updated version number * updated changelog
arundo · Mar 9, 2020 · 33395fa · 33395fa
1 parent de098cf
commit 33395fa
Show file tree

Hide file tree

Showing 8 changed files with 8 additions and 280 deletions.
diff --git a/docs/conf.py b/docs/conf.py
@@ -70,7 +70,7 @@
 # The short X.Y version.
 version = "0.6"
 # The full version, including alpha/beta/rc tags.
-release = "0.6.0-dev.26+pr.92"
+release = "0.6.0-dev.27+pr.93"
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

diff --git a/docs/notebooks/demo.ipynb b/docs/notebooks/demo.ipynb
@@ -3282,130 +3282,6 @@
     "precision(expanded_true_anomaly_list, detected_anomaly_list)"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### Resample a time series with new spacing"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Function `resample` samples time series with given new spacing. Resampled values are determined by time-weighted linear interpolation."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "In following example, we resample the given time series with spacing of 1 day."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 98,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "s = pd.Series([1, 2.5, 3, 4.25, 6, 9],\n",
-    "              index=pd.DatetimeIndex(['2017-1-1 00:00:00', '2017-1-2 12:00:00', '2017-1-3 00:00:00',\n",
-    "                                      '2017-1-4 06:00:00', '2017-1-6 00:00:00', '2017-1-9 00:00:00']))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 99,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "2017-01-01    1.0\n",
-       "2017-01-02    2.0\n",
-       "2017-01-03    3.0\n",
-       "2017-01-04    4.0\n",
-       "2017-01-05    5.0\n",
-       "2017-01-06    6.0\n",
-       "2017-01-07    7.0\n",
-       "2017-01-08    8.0\n",
-       "2017-01-09    9.0\n",
-       "Freq: D, dtype: float64"
-      ]
-     },
-     "execution_count": 99,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "from adtk.data import resample\n",
-    "resample(s, dT=\"1 day\")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "If no spacing is specified, the greatest common divider of original time steps will be used, which makes the refinement of time line a minimal refinement subject to keeping all original time points still included in the resampled time series. Please note that this may dramatically increase the size of time series and memory usage.\n",
-    "\n",
-    "For the previous example, such a refinement yields constant spacing of 6 hours. All original time points are kept in the refined series."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 100,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "2017-01-01 00:00:00    1.00\n",
-       "2017-01-01 06:00:00    1.25\n",
-       "2017-01-01 12:00:00    1.50\n",
-       "2017-01-01 18:00:00    1.75\n",
-       "2017-01-02 00:00:00    2.00\n",
-       "2017-01-02 06:00:00    2.25\n",
-       "2017-01-02 12:00:00    2.50\n",
-       "2017-01-02 18:00:00    2.75\n",
-       "2017-01-03 00:00:00    3.00\n",
-       "2017-01-03 06:00:00    3.25\n",
-       "2017-01-03 12:00:00    3.50\n",
-       "2017-01-03 18:00:00    3.75\n",
-       "2017-01-04 00:00:00    4.00\n",
-       "2017-01-04 06:00:00    4.25\n",
-       "2017-01-04 12:00:00    4.50\n",
-       "2017-01-04 18:00:00    4.75\n",
-       "2017-01-05 00:00:00    5.00\n",
-       "2017-01-05 06:00:00    5.25\n",
-       "2017-01-05 12:00:00    5.50\n",
-       "2017-01-05 18:00:00    5.75\n",
-       "2017-01-06 00:00:00    6.00\n",
-       "2017-01-06 06:00:00    6.25\n",
-       "2017-01-06 12:00:00    6.50\n",
-       "2017-01-06 18:00:00    6.75\n",
-       "2017-01-07 00:00:00    7.00\n",
-       "2017-01-07 06:00:00    7.25\n",
-       "2017-01-07 12:00:00    7.50\n",
-       "2017-01-07 18:00:00    7.75\n",
-       "2017-01-08 00:00:00    8.00\n",
-       "2017-01-08 06:00:00    8.25\n",
-       "2017-01-08 12:00:00    8.50\n",
-       "2017-01-08 18:00:00    8.75\n",
-       "2017-01-09 00:00:00    9.00\n",
-       "Freq: 6H, dtype: float64"
-      ]
-     },
-     "execution_count": 100,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "resample(s)"
-   ]
-  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -3432,7 +3308,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 101,
+   "execution_count": 98,
    "metadata": {},
    "outputs": [
     {
@@ -3463,7 +3339,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 102,
+   "execution_count": 99,
    "metadata": {},
    "outputs": [
     {
@@ -3491,7 +3367,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 103,
+   "execution_count": 100,
    "metadata": {},
    "outputs": [
     {
@@ -3519,7 +3395,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 104,
+   "execution_count": 101,
    "metadata": {},
    "outputs": [
     {

diff --git a/docs/releasehistory.rst b/docs/releasehistory.rst
@@ -103,6 +103,7 @@ Version 0.6.0-dev
 - Added consistency check between training and testing inputs in multivariate models (0.6.0-dev.23+pr.89)
 - Improved time index check in time-dependent models (0.6.0-dev.24+pr.90, 0.6.0-dev.25+pr.91)
 - Changed the output type of `adtk.data.split_train_test` from a 2-tuple of lists to a list of 2-tuples (0.6.0-dev.26+pr.92)
+- Removed `adtk.data.resample` because its functionality is highly overlapped with pandas resampler module (0.6.0-dev.27+pr.93)
 
 Version 0.5.5 (Feb 24, 2020)
 ===================================

diff --git a/setup.cfg b/setup.cfg
@@ -1,6 +1,6 @@
 [metadata]
 name = adtk
-version = 0.6.0-dev.26+pr.92
+version = 0.6.0-dev.27+pr.93
 author = Arundo Analytics, Inc.
 maintainer = Tailai Wen
 maintainer_email = [email protected]

diff --git a/src/adtk/__init__.py b/src/adtk/__init__.py
@@ -20,4 +20,4 @@
 
 """
 
-__version__ = "0.6.0-dev.26+pr.92"
+__version__ = "0.6.0-dev.27+pr.93"
diff --git a/src/adtk/data/__init__.py b/src/adtk/data/__init__.py
@@ -2,7 +2,6 @@
 
 from ._data import (
     expand_events,
-    resample,
     split_train_test,
     to_events,
     to_labels,
@@ -16,6 +15,5 @@
     "to_labels",
     "expand_events",
     "validate_events",
-    "resample",
     "split_train_test",
 ]
diff --git a/src/adtk/data/_data.py b/src/adtk/data/_data.py
@@ -537,73 +537,6 @@ def expand_events(
         raise TypeError("Arugment `lists` must be a list or a dict of lists.")
 
 
-def resample(
-    ts: Union[pd.Series, pd.DataFrame], dT: pd.Timedelta = None
-) -> Union[pd.Series, pd.DataFrame]:
-    """Resample the time points of a time series with given constant spacing.
-    The values at new time points are calcuated by time-weighted linear
-    interpolation.
-
-    Parameters
-    ----------
-    ts: pandas Series or DataFrame
-        Time series to resample.
-
-    dT: pandas Timedelta, str, or int, optional
-        The new constant time step.
-
-        - If str, it must be able to be converted into a pandas Timedelta
-          object.
-        - If int, it must be in nanosecond.
-
-        If not given, the greatest common divider of original time steps will
-        be used, which makes the refinement a minimal refinement subject to
-        keeping all original time points still included in the resampled time
-        series. Please note that this may dramatically increase the size of
-        time series and memory usage.
-
-        Default: None.
-
-    Returns
-    -------
-    pandas Series or DataFrame
-        Resampled time series.
-
-    """
-
-    def gcd_of_array(arr: np.array) -> int:
-        """Get the GCD of an array of integer"""
-        return reduce(gcd, arr)
-
-    ts = validate_series(ts)
-
-    isSeries = False
-    if isinstance(ts, pd.Series):
-        isSeries = True
-        seriesName = ts.name
-        ts = ts.to_frame()
-
-    if dT is None:
-        dT = pd.Timedelta(
-            np.timedelta64(gcd_of_array([int(dt) for dt in np.diff(ts.index)]))
-        )
-    elif not isinstance(dT, pd.Timedelta):
-        dT = pd.Timedelta(dT)
-
-    rdf = pd.DataFrame(index=pd.date_range(ts.index[0], ts.index[-1], freq=dT))
-
-    rdf = rdf.join(ts, how="outer")
-    rdf = rdf.interpolate("index")
-
-    rdf = rdf.reindex(pd.date_range(ts.index[0], ts.index[-1], freq=dT))
-
-    if isSeries:
-        rdf = rdf[rdf.columns[0]]
-        rdf.name = seriesName
-
-    return rdf
-
-
 def split_train_test(
     ts: Union[pd.Series, pd.DataFrame],
     mode: int = 1,

diff --git a/tests/test_resample.py b/tests/test_resample.py