Merge pull request #115 from BenjaminHsieh/organize

Major Organization

Makefile

@@ -1,15 +1,15 @@
-.PHONY: all clean coverage test
+.PHONY: all clean coverage test verbose
 
 all: clean
 
 clean:
 	find . -name "*.so" -o -name "*.pyc" -o -name "*.pyx.md5" | xargs rm -f
 
 coverage:
-	nosetests code/utils data --with-coverage --cover-package=data  --cover-package=utils
+	nosetests code/utils/tests data/tests --with-coverage --cover-package=data/data.py  --cover-package=code/utils/functions
 
 test:
-	nosetests code/utils data
+	nosetests code/utils/tests data/tests
 
 verbose:
-	nosetests -v code/utils data
+	nosetests -v code/utils/tests data/tests

README.md

@@ -1,5 +1,13 @@
-# project-template
+# UC Berkeley Stat 159/259 Fall 2015
+## Project-Theta
+
 [![Build Status](https://travis-ci.org/berkeley-stat159/project-theta.svg?branch=master)](https://travis-ci.org/berkeley-stat159/project-theta?branch=master)
 [![Coverage Status](https://coveralls.io/repos/berkeley-stat159/project-theta/badge.svg?branch=master)](https://coveralls.io/r/berkeley-stat159/project-theta?branch=master)
 
-Fall 2015 group project
+_**Group members:**_ Siyao Chang ([`changsiyao`](https://github.com/changsiyao)) , Boying Gong ([`boyinggong`](https://github.com/boyinggong)), Benjamin Hsieh ([`BenjaminHsieh`](https://github.com/BenjaminHsieh)), Brian Qiu ([`brianqiu`](https://github.com/brianqiu)), Jiang Zhu ([`pigriver123`](https://github.com/pigriver123))
+
+_**Topic:**_ [The Neural Basis of Loss Aversion in Decision Making Under Risk] (https://openfmri.org/dataset/ds000005/)
+[SOME SUMMARY OF PAPER]
+
+## Instructions
+[NEED TO ADD]

code/Makefile

@@ -1,5 +1,7 @@
+.PHONY: test coverage
+
 test:
-	nosetests code/utils -w ..
+	nosetests code/utils/tests -w ..
 
 coverage:
-	nosetests code/utils -w .. --with-coverage --cover-package=utils 
+	nosetests code/utils/tests -w .. --with-coverage --cover-package=utils/functions

code/scripts/find_mask_threshold.py

@@ -2,7 +2,7 @@
 import matplotlib.pyplot as plt
 import numpy as np
 import sys
-sys.path.append('../utils')
+sys.path.append('../utils/functions')
 import smooth_gaussian
 import os
 

code/scripts/findoutlier.py

@@ -1,7 +1,7 @@
 import numpy as np
 import json
 import sys
-pathtofunction = '../utils'
+pathtofunction = '../utils/functions'
 # Append the sys path
 sys.path.append(pathtofunction)
 from outlierfunction import outlier

code/scripts/graph_lindiag_script.py

@@ -4,7 +4,7 @@
 import sys
 
 # Path to function
-pathtofunction = '../utils'
+pathtofunction = '../utils/functions'
 # Append path to sys
 sys.path.append(pathtofunction)
 

code/scripts/graphoutlier.py

@@ -9,11 +9,17 @@
 # Paths
 pathtodata = '../../data/'
 pathtofig = '../../paper/figures/'
-pathtofunction = '../utils'
+pathtofunction = '../utils/functions'
+pathtographing = '../utils/graphing'
 # Append fuction path
 sys.path.append(pathtofunction)
 # Import function
-from graphoutlier_functions import loadnib_dict, loadtxt_dict, plot_dvars, plot_fd, plot_meanSig
+from basic_util import loadnib_dict, loadtxt_dict
+
+# Append graphing path
+sys.path.append(pathtographing)
+# Import function
+from graphoutlier_functions import plot_dvars, plot_fd, plot_meanSig
 
 
 # load outlier files

code/scripts/lme_script.py

@@ -5,12 +5,12 @@
 import sys
 
 # Path to function
-pathtofunction = '../utils'
+pathtofunction = '../utils/functions'
 # Append path to sys
 sys.path.append(pathtofunction)
 
 from behavtask_tr import events2neural_extend, merge_cond
-from regression_functions import hrf, getRegressor, calcBeta, calcMRSS, deleteOutliers
+from regression_functions import hrf, getRegressor, deleteOutliers
 from lme_functions import calcBetaLme, calcSigProp, calcAnov, anovStat
 
 n_vols=240

code/scripts/logistic.py

@@ -5,7 +5,7 @@
 import sys
 
 # Path to function
-pathtofunction = '../utils'
+pathtofunction = '../utils/functions'
 # Append path to sys
 sys.path.append(pathtofunction)
 
@@ -49,7 +49,16 @@
     logreg_proba = logreg.predict_proba(X)
     confusion = create_confusion(logreg_proba, y)
     addsub = fig.add_subplot(4, 4, i)
-    addsub, AUC = plot_roc(confusion, addsub, i)
+    addsub, ROC, AUC = plot_roc(confusion, addsub)
+    # Plot the ROC curve.
+    plt.plot(ROC[:,0], ROC[:,1], lw=2)
+    plt.xlim(-0.1,1.1)
+    plt.ylim(-0.1,1.1)
+    plt.xlabel('$FPR(t)$')
+    plt.ylabel('$TPR(t)$')
+    plt.grid()
+    plt.title('subject '+ str(i)+', AUC = %.4f'%AUC)
+    #------------------------------------------------------------------------#
     Min_thrs = np.append(Min_thrs, getMin_thrs(confusion))
     AUC_smr = np.append(AUC_smr, AUC)
 

code/scripts/make_smooth_graphs.py

@@ -2,7 +2,7 @@
 import matplotlib.pyplot as plt
 import nibabel as nib
 import sys
-sys.path.append('../utils')
+sys.path.append('../utils/functions')
 import smooth_gaussian
 
 #possibly put in os function to specify path later

code/scripts/regression_script.py

@@ -5,7 +5,7 @@
 import sys
 
 # Path to function
-pathtofunction = '../utils'
+pathtofunction = '../utils/functions'
 # Append path to sys
 sys.path.append(pathtofunction)
 

code/utils/functions/basic_util.py

@@ -0,0 +1,53 @@
+"""
+A collection of basic utility functions for use in outlier detection/graphing
+of fd, dvars, and meanSignal, see graphoutlier_functions.py
+"""
+
+import numpy as np
+import nibabel as nib
+
+
+def loadtxt_dict(file_name, fig_name):
+    """
+    Input: 
+        Txt file specified by file_name
+        Name of figure file wish to output
+    
+    Output:
+        Dictonary that attached to file_name the np array after loading 
+    file_name
+    """
+    data = np.loadtxt(file_name)
+    dict_out = {fig_name: data}
+    return dict_out
+
+def loadnib_dict(file_name, fig_name):
+    """
+    Input: 
+        Bold file (bold.nii, bold.nii.gz) specified by file_name 
+        Name of figure file wish to output
+    
+    Output:
+    Dictonary that attached to file_name the np array after loading file_name
+    """
+    img = nib.load(file_name)
+    data = img.get_data()
+    dict_out = {fig_name: data}
+    return dict_out
+
+
+# Calculate mean
+def vol_mean(data):
+    """ Return mean across voxels for $D `data`
+    
+    Input:
+        np array of data
+    Output: np array of dim (T,)
+        mean of data across all but the last dimension
+    """
+    mean_list = []
+    # Loop over the each volume and outputs the mean of each dimension
+    for i in range(data.shape[-1]):
+        mean = np.mean(data[...,i])
+        mean_list.append(mean)
+    return np.asarray(mean_list)

code/utils/behavtask_tr.py → code/utils/functions/behavtask_tr.py

@@ -70,10 +70,3 @@ def events2neural_extend(behav_task, tr, n_trs):
         time_course[on:on + dur,9] = RT
     return time_course
 
-def plot_time_course(time_course):
-    """
-    Simple function to plot time_course, an array from return of 
-    events2neural_extend
-    """
-    plt.plot(time_course[:,0])    
-    plt.show()

code/utils/lme_functions.py → code/utils/functions/lme_functions.py

@@ -3,8 +3,6 @@
 from scipy import stats  
 import pandas as pd
 import numpy as np
-from behavtask_tr import events2neural_extend, merge_cond
-from regression_functions import hrf, getRegressor, calcBeta, calcMRSS, deleteOutliers
 
 def calcBetaLme(data_full, gain_full, loss_full, linear_full, quad_full, run_group, thrshd=None):
     """ 

code/utils/logistic_function.py → code/utils/functions/logistic_function.py

@@ -8,8 +8,8 @@ def create_confusion(logreg_proba, y, thrs_inc=0.01):
         
         Parameters
         ----------
-        actual: Actual responses, 1-d array with values 0 or 1
-        fitted: Fitted probabilities, 1-d array with values between 0 and 1
+        y: Actual responses, 1-d array with values 0 or 1
+        logreg_proba: Fitted probabilities, 2-d array with values between 0 and 1 in second column
         thrs_inc: increment of threshold probability (default 0.05)
         
         Returns
@@ -52,15 +52,15 @@ def getMin_thrs(confusion):
         false_pos: number of incorrect trues
         false_neg: number of incorrect falses
         """
-    thrs_min = np.argmin(confusion[:,3]+ confusion[:,4])
+    thrs_min = np.min(confusion[:,3]+ confusion[:,4])
     col_out = confusion[thrs_min, :]
     thrs = col_out[0]
     false_pos = col_out[3]
     false_neg = col_out[4]
     return thrs, false_pos, false_neg
 
 
-def plot_roc(confusion, fig, sub_i):
+def plot_roc(confusion, fig):
     """
         function to plot the ROC (receiver operating characteristic) curve and
         calculate the corresponding AUC (Area Under Curve).
@@ -83,19 +83,10 @@ def plot_roc(confusion, fig, sub_i):
         # Compute true  positive rate for current threshold.
         TPR_t = confusion[i, 1] / float(confusion[i, 1] + confusion[i, 4])
         ROC[i,1] = TPR_t
-
-    # Plot the ROC curve.
-    plt.plot(ROC[:,0], ROC[:,1], lw=2)
-    plt.xlim(-0.1,1.1)
-    plt.ylim(-0.1,1.1)
-    plt.xlabel('$FPR(t)$')
-    plt.ylabel('$TPR(t)$')
-    plt.grid()
 
     AUC = 0.
     for i in range(confusion.shape[0]-1):
         AUC += (ROC[i+1,0]-ROC[i,0]) * (ROC[i+1,1]+ROC[i,1])
     AUC *= -0.5
 
-    plt.title('subject '+ str(sub_i)+', AUC = %.4f'%AUC)
-    return fig, AUC
+    return fig, ROC,  AUC

code/utils/regression_functions.py → code/utils/functions/regression_functions.py

@@ -65,8 +65,7 @@ def calcBeta(data, gain, loss, linear_dr, quad_dr, threshold=None):
     design[:, 3] = quad_dr
     designp = npl.pinv(design)
     if threshold!=None:
-        mean_data = np.mean(data, axis=-1)
-        mask = mean_data > threshold
+        mask = np.mean(data, axis=-1) > threshold
         data[~mask]=0
     T = data.shape[-1]
     time_by_vox = np.reshape(data, (-1, T)).T

code/utils/graphoutlier_functions.py → code/utils/graphing/graphoutlier_functions.py

@@ -1,38 +1,28 @@
 """
-A collection of utility functions for outlier detection/graphing
+A collection of  graphing  functions for outlier detection/graphing
 of fd, dvars, and meanSignal
 """
 import numpy as np
 import matplotlib.pyplot as plt
-import nibabel as nib
 
-def loadtxt_dict(file_name, fig_name):
-    """
-    Input: 
-        Txt file specified by file_name
-        Name of figure file wish to output
+# Calculate mean this is same as the one in basic_util.py 
+# Test already written, included here because graphing functions not tested
+# but need to use above: 
+def vol_mean(data):
+    """ Return mean across voxels for $D `data`
     
-    Output:
-        Dictonary that attached to file_name the np array after loading 
-    file_name
+    Input:
+        np array of data
+    Output: np array of dim (T,)
+        mean of data across all but the last dimension
     """
-    data = np.loadtxt(file_name)
-    dict_out = {fig_name: data}
-    return dict_out
+    mean_list = []
+    # Loop over the each volume and outputs the mean of each dimension
+    for i in range(data.shape[-1]):
+        mean = np.mean(data[...,i])
+        mean_list.append(mean)
+    return np.asarray(mean_list)
 
-def loadnib_dict(file_name, fig_name):
-    """
-    Input: 
-        Bold file (bold.nii, bold.nii.gz) specified by file_name 
-        Name of figure file wish to output
-    
-    Output:
-    Dictonary that attached to file_name the np array after loading file_name
-    """
-    img = nib.load(file_name)
-    data = img.get_data()
-    dict_out = {fig_name: data}
-    return dict_out
 
 # Graphing dvars: RMS signal derivative    
 def plot_dvars(dvars_dict, dvars_outliers, saveit=False):
@@ -64,6 +54,7 @@ def plot_dvars(dvars_dict, dvars_outliers, saveit=False):
         plt.savefig(dvars_dict.keys()[0])
         plt.close()
 
+
 # Graphing fd: Framewise displacement
 def plot_fd(fd_dict, fd_outliers, saveit=False):
     """
@@ -94,21 +85,6 @@ def plot_fd(fd_dict, fd_outliers, saveit=False):
         plt.savefig(fd_dict.keys()[0])
         plt.close()
 
-# Calculate mean
-def vol_mean(data):
-    """ Return mean across voxels for $D `data`
-    
-    Input:
-        np array of data
-    Output: np array of dim (T,)
-        mean of data across all but the last dimension
-    """
-    mean_list = []
-    # Loop over the each volume and outputs the mean of each dimension
-    for i in range(data.shape[-1]):
-        mean = np.mean(data[...,i])
-        mean_list.append(mean)
-    return np.asarray(mean_list)
 
 # Graphing mean signal	
 def plot_meanSig(bdata_dict, saveit = False):
@@ -141,4 +117,3 @@ def plot_meanSig(bdata_dict, saveit = False):
         plt.savefig(bdata_dict.keys()[0])
         plt.close()
 
-