revert(models): restore forecast function

Author: christophertitchen

nbs/src/core/models.ipynb

@@ -61,7 +61,6 @@
     "from typing import Any, Dict, List, Optional, Tuple, Union\n",
     "\n",
     "import numpy as np\n",
-    "from coreforecast.exponentially_weighted import exponentially_weighted_mean\n",
     "from numba import njit\n",
     "from scipy.optimize import minimize_scalar\n",
     "from scipy.special import inv_boxcox\n",
@@ -3677,14 +3676,15 @@
     "    return sse\n",
     "\n",
     "\n",
+    "@njit(nogil=NOGIL, cache=CACHE)\n",
     "def _ses_forecast(x: np.ndarray, alpha: float) -> Tuple[float, np.ndarray]:\n",
     "    r\"\"\"Compute the one-step ahead forecast for a simple exponential smoothing fit.\n",
-    "    \n",
+    "\n",
     "    Parameters\n",
     "    ----------\n",
     "    x : numpy.array\n",
     "        Clean time series of shape (n, ).\n",
-    "    alpha : float \n",
+    "    alpha : float\n",
     "        Smoothing parameter.\n",
     "\n",
     "    Returns\n",
@@ -3693,9 +3693,16 @@
     "        One-step ahead forecast and in-sample fitted values.\n",
     "\n",
     "    \"\"\"\n",
-    "    fitted = exponentially_weighted_mean(x, alpha)\n",
-    "    forecast = fitted.item(-1)\n",
-    "    fitted[1:] = fitted[:-1]\n",
+    "    complement = 1 - alpha\n",
+    "    fitted = np.empty_like(x)\n",
+    "    fitted[0] = x[0]\n",
+    "    j = 0\n",
+    "\n",
+    "    for i in range(1, len(x)):\n",
+    "        fitted[i] = alpha * x[j] + complement * fitted[j]\n",
+    "        j += 1\n",
+    "\n",
+    "    forecast = alpha * x[j] + complement * fitted[j]\n",
     "    fitted[0] = np.nan\n",
     "    return forecast, fitted\n",
     "\n",

python/statsforecast/models.py

@@ -15,7 +15,6 @@
 from typing import Any, Dict, List, Optional, Tuple, Union
 
 import numpy as np
-from coreforecast.exponentially_weighted import exponentially_weighted_mean
 from numba import njit
 from scipy.optimize import minimize_scalar
 from scipy.special import inv_boxcox
@@ -2158,6 +2157,7 @@ def _ses_sse(alpha: float, x: np.ndarray) -> float:
     return sse
 
 
+@njit(nogil=NOGIL, cache=CACHE)
 def _ses_forecast(x: np.ndarray, alpha: float) -> Tuple[float, np.ndarray]:
     r"""Compute the one-step ahead forecast for a simple exponential smoothing fit.
 
@@ -2174,9 +2174,16 @@ def _ses_forecast(x: np.ndarray, alpha: float) -> Tuple[float, np.ndarray]:
         One-step ahead forecast and in-sample fitted values.
 
     """
-    fitted = exponentially_weighted_mean(x, alpha)
-    forecast = fitted.item(-1)
-    fitted[1:] = fitted[:-1]
+    complement = 1 - alpha
+    fitted = np.empty_like(x)
+    fitted[0] = x[0]
+    j = 0
+
+    for i in range(1, len(x)):
+        fitted[i] = alpha * x[j] + complement * fitted[j]
+        j += 1
+
+    forecast = alpha * x[j] + complement * fitted[j]
     fitted[0] = np.nan
     return forecast, fitted