import timeit import numpy as np import pandas as pd import matplotlib.pyplot as plt from sklearn import datasets, linear_model X, y = datasets.load_diabetes(return_X_y=True) raw = X[:, None, 2]

# escalamiento max-min max_raw = max(raw) min_raw = min(raw) scaled = (2*raw - max_raw -min_raw)/(max_raw - min_raw) # normalización Z-score avg = np.average(raw) std = np.std(raw) z_scaled = (raw - avg)/std fig, axs = plt.subplots(3, 1, sharex=True, tight_layout=True) axs[0].hist(raw) axs[1].hist(scaled) axs[2].hist(z_scaled)

# modelos para entrenamiento def train_raw(): linear_model.LinearRegression().fit(raw, y) def train_scaled(): linear_model.LinearRegression().fit(scaled, y) def train_z_scaled(): linear_model.LinearRegression().fit(z_scaled, y)

raw_time = timeit.timeit(train_raw, number = 100) scaled_time = timeit.timeit(train_raw, number = 100) z_scaled_time = timeit.timeit(train_raw, number = 100) print('trainning time for raw data : {} '.format(raw_time)) print('trainning time for scaled data : {}'.format(scaled_time)) print('trainning time for z_scaled data : {}'.format(z_scaled_time))

df = pd.read_csv('cars.csv')

df.price_usd.hist()

p = 10000 df.price_usd.apply(lambda x: np.tanh(x/p)).hist()