import numpy as np
import matplotlib.pylab as plt
from scipy import stats

datatable = np.genfromtxt('savage_metabolicrate_data.csv', delimiter=',',
                          dtype=None, names=True)

#DISTRIBUTIONS
x = np.arange(-5, 5, 0.01)
norm_pdf = stats.norm.pdf(x)
norm_cdf = stats.norm.cdf(x)
norm_rand_nums = stats.norm.rvs(size=100)
plt.subplot(1, 3, 1)
plt.plot(x, norm_pdf, 'b-')
plt.subplot(1, 3, 2)
plt.plot(x, norm_cdf, 'r-')
plt.subplot(1, 3, 3)
plt.hist(norm_rand_nums)

#DESCRIPTIVE STATISTICS
x = [1, 2, 3, 4, 5]
print stats.describe(x)

#REGRESSION
logmass = np.log10(datatable['Massg'])
logmetabolicrate = np.log10(datatable['BMRW'])
slope, intercept, rval, pval, stderr = stats.linregress(logmass, logmetabolicrate)
print(slope, rval ** 2, pval)

plt.figure()
plt.plot(logmass, logmetabolicrate, 'ro')
plt.plot([min(logmass), max(logmass)], [min(logmetabolicrate), max(logmetabolicrate)], 'k-', linewidth=2)
plt.xlabel('Log(Mass)', fontsize=20)
plt.ylabel('Log(Metabolic Rate)', fontsize=20)