jimpea

from scipy import sparse
from scipy.sparse.linalg import spsolve
import numpy as np

def als(self, y, lam, p, niter):
        '''Asymmetric Least Squares Smoothing algorithm of Eilers and Boelens (2005)

        From Stackoverflow 29185844 by Sparrowcide as modified by jpatina
        
        Good choices for the parameters are:

jimpea

from matplotlib import pyplot as plt

fractions = range(1, 16)
mcgs = [0,0,0,0,0,190,145,10,0,0,0,0,0,0,0]
dpms = [0,0,0,0,0,290,390,0,0,0,0,5,20,80,350]

#check lists for length
for list in [dpms, mcgs]:
    assert len(list) == len(fractions), "lists should be same length"

jimpea

import numpy as np
figsize = (10, 4) # set the fgsize for both the box plots and histograms
# make boxplots

# notch shape box plot

def make_boxplot(ax, data, title):
    labels = [d.name for d in data]
    ax.boxplot(data,
                notch=True,  # notch shape

jimpea

import logging

logger = logging.getLogger(__name__)
logger.setLevel(logging.DEBUG)
formatter = logging.Formatter('%(asctime)s:%(levelname)s:%(name)s:%(message)s')
file_handler = logging.FileHandler('combined.log')
file_handler.setFormatter(formatter)

stream_handler = logging.StreamHandler()
stream_handler.setFormatter(formatter)

jimpea

import matplotlib.pyplot as plt
import numpy as np

x_data = np.arange(0, 10)
y_data = [n + np.random.normal() for n in x_data]

def poly_model(x_data, coeffs):
    '''Fit data to a linear polynomial model
    
    Args:

jimpea

# Use the recomended numpy array notation
import numpy as np
y = np.array([6.95,5.22,6.36,7.03,9.71,9.67,10.69,13.88,13.21,14.82])
X0 = np.array(np.ones(10))
X1 = np.array(range(0, 10, 1))
X = np.column_stack((X0.T, X1.T))
yhat = np.linalg.inv(X.T.dot(X)).dot(X.T).dot(y.T)
print yhat 
'''
    > [ 4.97309091  1.06242424]