Basic core model with calcs that use source funcs

specialsauce/core.py

@@ -0,0 +1,78 @@
+import math
+
+import numpy as np
+import pandas as pd
+
+from specialsauce import util
+from specialsauce.sources import minetti
+
+
+def power_met_ss(speed, grade=0.0):
+  return speed * minetti.cost_of_running(grade)
+
+
+def power_met_ss(speed_series, grade_series=None):
+  """Calculate steady-state metabolic power in the moderate domain.
+
+  For more info, see `heartandsole_local/heartandsole/powerutils.py`.
+  """
+  if grade_series is None:
+    grade_series = pd.Series([0 for i in range(len(speed_series))])
+
+  c_i_series = grade_series.apply(minetti.cost_of_running)
+
+  # Instantaneous running power (W/kg) is simply cost of running 
+  # (J/kg/m) multiplied by speed (m/s).
+  power_series = c_i_series * speed_series
+
+  # Updated to account for the fact that the horizontal speed is
+  # measured, but cost of running relates to the distance along the
+  # incline.
+  power_series = power_series / np.cos(np.arctan(grade_series))
+
+  return power_series
+
+
+def power_met(speed_series, grade_series=None, time_series=None, tau=20):
+  """Calculate metabolic power in the moderate domain as a time series."""
+  if time_series is None:
+    time_series = pd.Series([i for i in range(len(speed_series))])
+
+  # Calculate the theoretical steady-state power associated with the
+  # speed and grade value at each timestep.
+  power_met_inst = power_met_ss(speed_series, grade_series=grade_series)
+
+  halflife = tau * math.log(2)
+
+  # Even if we have uniform 1-second samples, I would still need this
+  # util function. It makes the average start at 0 and trend up, rather
+  # than letting the first value have all the weight.
+
+  return util.ewma_halflife(
+    power_met_inst,
+    halflife,
+    time_series=time_series,
+  )
+
+
+def run_cost(grade=0.0):
+  """Calculates the metabolic cost of running.
+
+  See the documentation for powerutils.o2_power_ss for information
+  on the scientific basis for this calculation.
+
+  Args:
+    speed (float): Running speed in meters per second. 
+    grade (float): Decimal grade, i.e. 45% = 0.45.
+
+  Returns:
+    float: Cost of running on an inclined treadmill, in Joules/kg/m,
+      with distance measured along the incline slope.
+  """
+  c_i = minetti.cost_of_running(grade)
+
+  return c_i
+
+
+def run_power(speed, grade=0.0):
+  return run_cost(speed, grade=grade) * speed / math.cos(math.atan(grade))