cp-model.py

import matplotlib.pyplot as plt
from matplotlib.path import Path
import matplotlib.patches as patches
import pandas as pd
import sweat

x = [6, 180, 720]
X = sweat.array_1d_to_2d(x) 
y = [831,436,332]

#cp = round(((12*y[2])-(3*y[1]))/9)
#w_prime = 0.24*(y[1] - y[2])*1000

pdr = sweat.PowerDurationRegressor(model="3 param")
pdr.fit(X, y)

durations = range(0, 3600)
durations_2d = sweat.array_1d_to_2d(durations)
data = pd.DataFrame({
    "3 param": pdr.predict(durations_2d)
})

print(f"CP: {pdr.cp_}")
print(f"W': {pdr.w_prime_}")
#watts is joules per seconden. controle watts = p3min - cp
print(f"W' in watts: {pdr.w_prime_/x[1]}")

# plot cp
fig = plt.figure(figsize=(15, 10))

ax = fig.add_subplot(1, 1, 1)
  
pp1 = plt.Rectangle((0, pdr.cp_),
                    x[1], y[1]-pdr.cp_,color='orange',alpha=0.4)
  
pp2 = plt.Rectangle((0, pdr.cp_),
                    x[2], y[2]-pdr.cp_,color='orange',alpha=0.4)
  
ax.add_patch(pp1)
ax.add_patch(pp2)

#asymtoot
plt.text(3000, pdr.cp_-10, "cp asymtoot")
plt.plot([0,3600], [pdr.cp_, pdr.cp_], 'r--')
# linear regressie
plt.scatter(x, y, color="black")
plt.plot(durations, data['3 param'], label="cp curve")

plt.show()