Merge pull request #223 from timothy1191xa/lin_reg

updated path and moved files

code/utils/functions/linear_regression.py

@@ -2,8 +2,8 @@
 
 
 import pandas as pd
-import sys
-sys.path.append(".././utils")
+import sys, os
+sys.path.append(os.path.join(os.path.dirname(__file__), "./"))
 #import statsmodels.api as sm
 import statsmodels.formula.api as smf
 import pylab as pl
@@ -13,7 +13,8 @@
 from scipy.stats import t 
 import numpy.linalg as npl
 import math
-from log_regression import *
+import logistic_reg
+from logistic_reg import *
 
 
 
@@ -44,7 +45,7 @@ def load_data(subject, data_dir = "/Users/macbookpro/Desktop/stat159_Project/"):
 
 	except IOError:
 
-		print "Can't find files in such directory! Please enter the directory where you store ds005 dataset!"
+		print ("Can't find files in such directory! Please enter the directory where you store ds005 dataset!")
 		return
 
 	run_1=run1.append(run2)
@@ -182,7 +183,7 @@ def linear_regression(data, y, *arg):
 	for i in range(1,p):
 		print('==============================================================')
 		print(arg[i-1])
-		print 'Coefficient: ' + str(beta[i]), 'p-value: ' + str(pvalues[i])
+		print ('Coefficient: ' + str(beta[i]), 'p-value: ' + str(pvalues[i]))
 
 	return beta, pvalues
 
@@ -247,10 +248,6 @@ def linear_regression_fast(data, formula):
 
 	return est
 
-# My prediction for Response Time given ratio (gain/loss)
-def my_line(x, beta = B):
-    	return beta[0] + beta[1] * x
-
 
 
 def simple_regression_plot(data, dep_var, exp_var):
@@ -308,19 +305,20 @@ def plot_neural_and_behav_loss_aversion(data, subject):
 		logit_pars = x.params
 		ratio =  -logit_pars['loss'] / logit_pars['gain'] 
 		lambdas.append( math.log(ratio) )
-		loss_aversion.append( (-logit_pars['loss']) - logit_pars['gain'] )
+		loss_aversion.append( (-logit_pars['loss']) - logit_pars['gain'] ) # This will be changed!
+
 
 	X = np.column_stack((np.ones(16), loss_aversion))
 
 
 
 	B = npl.pinv(X).dot(lambdas)
 
-	def my_line(x):
+	def my_line(x, B = B):
     	# Best prediction 
-    	return B[0] + B[1] * x
-
-   	x_vals = [0, max(loss_aversion)]
+		return B[0] + B[1] * x
+		
+	x_vals = [0, max(loss_aversion)]
 	y_vals = [my_line(0), my_line(max(loss_aversion))]
 
 	plt.plot(loss_aversion, lambdas, '+')

code/utils/logistic_reg.py → code/utils/functions/logistic_reg.py

@@ -120,9 +120,9 @@ def plot_neural_and_behav_loss_aversion(data, subject):
 
 	def my_line(x):
     	# Best prediction 
-    	return B[0] + B[1] * x
+		return B[0] + B[1] * x
 
-   	x_vals = [0, max(loss_aversion)]
+	x_vals = [0, max(loss_aversion)]
 	y_vals = [my_line(0), my_line(max(loss_aversion))]
 
 	plt.plot(loss_aversion, lambdas, '+')

code/utils/tests/test_linear_regression.py

@@ -28,11 +28,12 @@
 #Specicy the path for functions
 sys.path.append(os.path.join(os.path.dirname(__file__), "../functions/"))
 import linear_regression
+from linear_regression import *
 
 
 
 
-def test_pearson_1d():
+def test_linear_regression():
 
 
 	# Create a data frame
@@ -63,3 +64,11 @@ def test_pearson_1d():
 	assert_almost_equal(expected_p2, pvalues2)
 	assert_almost_equal(expected_beta3, beta3)
 	assert_almost_equal(expected_p3, pvalues3)
+
+
+
+
+
+
+
+