In [1]:

    

%matplotlib inline
projecttitle = 'Analogy'
import sys, os
if sys.platform == 'darwin':
    
    sys.path.append(os.path.join("/Users", "njchiang", "GitHub", "task-fmri-utils"))
else:
    sys.path.append(os.path.join("D:\\", "GitHub", "task-fmri-utils"))


    

In [2]:

    

from fmri_core import analysis as pa
from fmri_core import utils as pu
import numpy as np
import matplotlib.pyplot as plt


    

In [3]:

    

os.chdir("D:\\CloudStation\\Grad\\Research\\montilab-ucla")


    

In [4]:

    

projectSettings = pu.loadConfig(os.path.join('analogy', 'config', 'project.json'))
analysisSettings = pu.loadConfig(os.path.join('analogy', 'config', 'analyses.json'))
# paths = projectSettings['filepaths']['osxPaths']
paths = projectSettings['filepaths']['winPaths']
analysisSettings['searchlight']


    

    




Loading JSON config from analogy\config\project.json
Loading JSON config from analogy\config\analyses.json






    
Out[4]:





{'n_jobs': -1, 'process_mask_img': None, 'radius': 5, 'verbose': 0}

In [5]:

    

paths


    

    
Out[5]:





{'cloud': 'D:\\CloudStation\\Grad\\Research\\Analogy',
 'code': 'D:\\CloudStation\\Grad\\Research\\montilab-ucla\\analogy',
 'root': 'D:\\fmri\\Analogy'}

In [6]:

    

sub = "sub-01"
run = "run-01"


    

In [7]:

    

funcFile = ("%s_task-analogy_%s_bold.nii.gz" % (sub, run))
eventFile = ("%s_task-analogy_%s_events.tsv" % (sub, run))
maskFile = "aal-LOFC-bin_mask.nii.gz"


    

In [8]:

    

img = pu.maskImg(os.path.join(paths['root'], 
                             'derivatives', 
                             sub, 'func',
                             funcFile),
                os.path.join(paths['root'],
                            'derivatives',
                            sub, 'masks',
                            maskFile))
labels = pu.loadLabels(os.path.join(paths['root'],
                                   'derivatives', 
                                   sub, 'func',
                                   eventFile), sep="\t")


    

    




Masking D:\fmri\Analogy\derivatives\sub-01\func\sub-01_task-analogy_run-01_bold.nii.gz
Loading label file from: D:\fmri\Analogy\derivatives\sub-01\func\sub-01_task-analogy_run-01_events.tsv

In [9]:

    

labels.keys()


    

    
Out[9]:





Index(['Onset', 'Duration', 'TrialTag', 'ABTag', 'CDTag', 'Match', 'ABMainRel',
       'CDMainRel', 'ABSubRel', 'CDSubRel', 'ProbeResp', 'ProbeCorr',
       'ProbeArr', 'AB', 'CD', 'Probe', 'Intensity', 'TrialType'],
      dtype='object')

In [10]:

    

label = [("%s_%s" % (labels.iloc[i]['TrialTag'], 
                     labels.iloc[i]['TrialType'])) 
         for i in range(len(labels))]


    

In [11]:

    

tr = 1.0
frametimes = np.arange(img.shape[0]) * tr


    

In [12]:

    

des = pa.make_designmat(frametimes, label, 
                  labels['Onset'].values,
                 labels['Duration'].values)


    

    




Creating design matrix...

In [13]:

    

des.show()


    

    
Out[13]:





<matplotlib.axes._subplots.AxesSubplot at 0x1be0269cd68>






    

In [14]:

    

sgf = pa.sgfilter(window_length=121, polyorder=3, **analysisSettings['SGF'])


    

    




Creating SG filter

In [15]:

    

from sklearn.linear_model import Ridge, LinearRegression
ols = LinearRegression()
ridge = Ridge()


    

In [16]:

    

ols.fit(des.matrix, sgf.fit_transform(img))


    

    
Out[16]:





LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)

In [17]:

    

ridge.fit(des.matrix, sgf.fit_transform(img))


    

    
Out[17]:





Ridge(alpha=1.0, copy_X=True, fit_intercept=True, max_iter=None,
   normalize=False, random_state=None, solver='auto', tol=0.001)

In [ ]:

    

ols.coef_.shap


    

In [ ]:

    

a = des.names
a.remove("constant")


    

In [ ]:

    

a


    

In [ ]:

    

plt.imshow(pa.rdm(arr.T, square=True))


    

In [ ]:

    

arr.shape


    

In [ ]:

    

trialtype, trialtag = [], []
for t in des.names:
    curr = t.split('_')
    trialtype += curr[0]
    trialtag += curr[1]


    

In [ ]:

    

des.names.index('constant')


    

In [ ]:

    

ols.coef_[:, des.names.index("constant")].sum()


    

In [ ]:

    

for i in des.names:
    print(i.split('_')[0], i.split('_')[1])


    

In [ ]: