def processWikipediaOscarFiles(procYear = None):
outdir = getWikipediaDir()
if procYear == None:
files = findSubExt(outdir, "data", ext=".p")
#files = glob(join(outdir, "data", "*.p"))
else:
files = findSubPatternExt(outdir, "data", pattern=str(procYear), ext=".p")
#files = glob(join(outdir, "data", str(procYear)+".p"))
movies = OrderedDict()
for ifile in files:
print ifile
year = getBaseFilename(ifile)
print year
#if year == "1985": continue
htmldata = get(ifile)
bsdata = getHTML(htmldata)
if int(year) <= 1984:
results = parseWikipediaOscarDataPre1985(bsdata, True)
elif int(year) >= 1986:
results = parseWikipediaOscarDataPost1987(bsdata, True)
else:
results = parseWikipediaOscarData1985(debug = True)
movies[year] = results
for k,v in results.iteritems():
print "====>",year,'\t',k
print " Winner :",results[k]["Winner"]
print " Nominees:",results[k]["Nominees"]
print ""
savename = setFile(outdir, "oscars.yaml")
print "Saving",len(movies),"years of wikipedia oscar data to",savename
save(savename, movies)
def getOscarData():
filename = setFile(getOscarDir(), "oscars.yaml")
data = get(filename)
yearlyData = {}
for year,ydata in data.iteritems():
movies = {}
for category,categorydata in ydata.iteritems():
if category.find("Song") != -1:
continue
sf = 1
if category.find("Song") != -1:
sf = 0
elif category.find("Picture") != -1:
sf = 40
elif category.find("Animated Feature") != -1:
sf = 35
elif category.find("Director") != -1:
sf = 30
elif category.find("Actor") != -1 or category.find("Actress") != -1:
sf = 25
elif category.find("Screenplay") != -1:
sf = 20
winner = categorydata.get("Winner")
if winner:
#print category,'\t',winner
if isinstance(winner, list):
movie = winner[0]
else:
movie = winner
#print category,'\t',10*sf,'\t',winner
if movies.get(movie) == None:
movies[movie] = 10*sf
else:
movies[movie] = max(10*sf, movies[movie])
nominees = categorydata.get("Nominees")
if nominees:
for nominee in nominees:
if isinstance(nominee, list):
movie = nominee[0]
else:
movie = nominee
#print category,'\t',sf,'\t',winner
if movies.get(movie) == None:
movies[movie] = sf
else:
movies[movie] = max(sf, movies[movie])
yearlyData[year] = sorted(movies.items(), key=operator.itemgetter(1), reverse=True)
print "---->",year,"<----"
for item in yearlyData[year][:15]:
print item
print '\n'
savename = setFile(getOscarDir(), "oscars.json")
print "Saving",len(yearlyData),"yearly results to",savename
save(savename, yearlyData)
def processBoxOfficeMojo():
outdir = getBoxOfficeDir()
savename = setFile(outdir, "results.json")
data = get(savename)
movies = {}
yearlyData = {}
for i,year in enumerate(data.keys()):
movies[year] = {}
ydata = data[year]
for wdata in ydata:
for mdata in wdata:
movie = mdata[2]
retval = search("\((\d+)\)",movie)
if retval:
stryear = retval.group()
movie = movie.replace(stryear, "").strip()
gross = convertCurrency(mdata[9])
weekly = convertCurrency(mdata[4])
money = max(gross, weekly)
if movies[year].get(movie) == None:
movies[year][movie] = money
else:
movies[year][movie] = max(money, movies[year][movie])
yearlyData[year] = sorted(movies[year].items(), key=operator.itemgetter(1), reverse=True)
print "---->",year,"<----"
for item in yearlyData[year][:25]:
print item
print '\n'
savename = setFile(outdir, "boxofficemojo.json")
print "Saving",len(yearlyData),"yearly results to",savename
save(savename, yearlyData)
def woerter_lernen(indizes, tdn, wbn): wb = fileio.wb_laden(wbn) worte = fehlerabschnitt.fa.worte_finden(indizes) wb.extend(worte) fileio.save(wbn, wb) fehlerabschnitt.fa.fehlerworte = [x for x in fehlerabschnitt.fa.fehlerworte if x not in worte] zeilendialog.anzeigen(fa = fehlerabschnitt.fa, tdn = tdn, wbn = wbn)
def new_task(args):
name = ' '.join(args.name)
if not name:
print(INVALID_NAME_MSG)
return
date_str = args.date
if date_str:
try:
date = interpret_date(date_str)
except ValueError as e:
print(e)
return
else:
date = None
estimated_hours = float(args.est_hours)
manager = TaskManager(fileio.load())
id = manager.new_task(name, estimated_hours, date)
fileio.save(manager.task_dict)
print_table(manager, id)
def correctOscarData():
print "Checking for unparsed oscar data."
backupfilename = setFile(getWikipediaDir(), "oscars.yaml.backup")
filename = setFile(getWikipediaDir(), "oscars.yaml")
copyFile(filename, backupfilename)
data = get(filename)
#fixes = {}
for year,ydata in data.iteritems():
print "\n==>",year
for cat,catdata in ydata.iteritems():
winner = catdata["Winner"]
if isinstance(winner, list):
if winner[0].find(",") != -1:
print "\t",cat,"\t",winner[0]
nominees = catdata["Nominees"]
for nominee in nominees:
if isinstance(nominee, list):
if nominee[0].find(",") != -1:
print "\t",cat,"\t",nominee[0]
savename = setFile(getOscarDir(), "oscars.yaml")
print "Saving",len(data),"yearly results to",savename
save(savename, data)
def woerter_lernen(indizes, tdn, wbn):
wb = fileio.wb_laden(wbn)
worte = fehlerabschnitt.fa.worte_finden(indizes)
wb.extend(worte)
fileio.save(wbn, wb)
fehlerabschnitt.fa.fehlerworte = [
x for x in fehlerabschnitt.fa.fehlerworte if x not in worte
]
zeilendialog.anzeigen(fa=fehlerabschnitt.fa, tdn=tdn, wbn=wbn)
def save(self, fname='save.dcb'):
'''Cleans everything up (play back, recording, etc), and saves the current
lecture and audio. If there is no set filename, uses the GUI to ask the
user where they would like the file saved. (Their answer will effect the
format of the save.'''
# TODO check if this is dirty.
if self.is_recording():
self.record(False)
fileio.save(fname, self.lec)
self.gui.canvas.dirty = False
def save(self, fname = 'save.dcb'):
'''Cleans everything up (play back, recording, etc), and saves the current
lecture and audio. If there is no set filename, uses the GUI to ask the
user where they would like the file saved. (Their answer will effect the
format of the save.'''
# TODO check if this is dirty.
if self.is_recording():
self.record(False)
fileio.save(fname, self.lec)
self.gui.canvas.dirty = False
def mergeBoxOfficeMojoResults():
outdir = getBoxOfficeDir()
retval = {}
files = findSubExt(outdir, "results", ext=".json")
for ifile in files:
year = getBaseFilename(ifile)
data = get(ifile)
retval[year] = data
savename = setFile(outdir, "results.json")
print "Saving",len(retval),"years of movie data to",savename
save(savename, retval)
def mergeYearlyMovies(outdir, yearlyMovies):
data = OrderedDict()
for year,ymovies in yearlyMovies.iteritems():
data[year] = {}
for movie,movieType in ymovies.iteritems():
if data[year].get(movieType) == None:
data[year][movieType] = {}
data[year][movieType][movie] = None
savename = setFile(outdir, "officialMovies.yaml")
print "Saving",len(data),"to",savename
save(savename, data)
def clock(args):
manager = TaskManager(fileio.load())
task_id = args.id
to_deduct = args.to_deduct
# TODO Complete if less than/eq 0?
try:
old_task: task.Task = manager.task_dict[task_id]
old_task.hours_remaining -= to_deduct
fileio.save(manager.task_dict)
print_table(manager, task_id)
except KeyError:
print(INVALID_ID_MSG)
def parseBoxOfficeMojoResults(startYear = 1982, endYear = 2017):
outdir = getBoxOfficeDir()
if endYear == None: endYear = startYear
years = range(int(startYear), int(endYear)+1)
for year in years:
retval = []
files = findSubPatternExt(outdir, "data", pattern=str(year), ext=".p")
for ifile in files:
result = parseBoxOfficeMojo(ifile)
retval.append(result)
savename = setSubFile(outdir, "results", str(year)+".json")
print "Saving",len(retval),"weekends of movie data to",savename
save(savename, retval)
def complete_task(args):
manager = TaskManager(fileio.load())
ids = args.id
if not ids:
print(INVALID_ID_MSG)
else:
completed_tasks_dict = dict()
for i in ids:
try:
completed_tasks_dict[i] = manager.task_dict[i]
manager.complete(i)
fileio.save(manager.task_dict)
except KeyError:
print(INVALID_ID_MSG)
print_table(manager)
print()
print_completed_tasks(completed_tasks_dict)
def modify_task(args):
manager = TaskManager(fileio.load())
task_id = args.id
try:
old_task: task.Task = manager.task_dict[task_id]
new_name = args.name if args.name else old_task.name
if args.date:
new_date = interpret_date(args.date)
elif args.floating or isinstance(old_task, task.FloatingTask):
new_date = None
else:
new_date = old_task.due_date
new_hours_remaining = args.est_hours if args.est_hours else old_task.hours_remaining
manager.new_task(new_name, new_hours_remaining, new_date, task_id)
fileio.save(manager.task_dict)
print_table(manager, task_id)
except KeyError:
print(INVALID_ID_MSG)
except ValueError as e:
print(e)
def estimate(respfile,
covfile,
maskfile=None,
cvfolds=None,
testcov=None,
testresp=None,
saveoutput=True,
outputsuffix=None):
""" Estimate a normative model
This will estimate a model in one of two settings according to the
particular parameters specified (see below):
* under k-fold cross-validation
* estimating a training dataset then applying to a second test dataset
The models are estimated on the basis of data stored on disk in ascii or
neuroimaging data formats (nifti or cifti). Ascii data should be in
tab or space delimited format with the number of subjects in rows and the
number of variables in columns. Neuroimaging data will be reshaped
into the appropriate format
Basic usage::
estimate(respfile, covfile, [extra_arguments])
where the variables are defined below. Note that either the cfolds
parameter or (testcov, testresp) should be specified, but not both.
:param respfile: response variables for the normative model
:param covfile: covariates used to predict the response variable
:param maskfile: mask used to apply to the data (nifti only)
:param cvfolds: Number of cross-validation folds
:param testcov: Test covariates
:param testresp: Test responses
:param saveoutput: Save the output to disk? Otherwise returned as arrays
:param outputsuffix: Text string to add to the output filenames
All outputs are written to disk in the same format as the input. These are:
:outputs: * yhat - predictive mean
* ys2 - predictive variance
* Z - deviance scores
* Rho - Pearson correlation between true and predicted responses
* pRho - parametric p-value for this correlation
* rmse - root mean squared error between true/predicted responses
* smse - standardised mean squared error
The outputsuffix may be useful to estimate multiple normative models in the
same directory (e.g. for custom cross-validation schemes)
"""
# load data
print("Processing data in " + respfile)
X = fileio.load(covfile)
Y, maskvol = load_response_vars(respfile, maskfile)
if len(Y.shape) == 1:
Y = Y[:, np.newaxis]
if len(X.shape) == 1:
X = X[:, np.newaxis]
Nmod = Y.shape[1]
if testcov is not None:
# we have a separate test dataset
Xte = fileio.load(testcov)
Yte, testmask = load_response_vars(testresp, maskfile)
testids = range(X.shape[0], X.shape[0] + Xte.shape[0])
if len(Yte.shape) == 1:
Yte = Yte[:, np.newaxis]
if len(Xte.shape) == 1:
Xte = Xte[:, np.newaxis]
# treat as a single train-test split
splits = CustomCV((range(0, X.shape[0]), ), (testids, ))
Y = np.concatenate((Y, Yte), axis=0)
X = np.concatenate((X, Xte), axis=0)
# force the number of cross-validation folds to 1
if cvfolds is not None and cvfolds != 1:
print("Ignoring cross-valdation specification (test data given)")
cvfolds = 1
else:
# we are running under cross-validation
splits = KFold(n_splits=cvfolds)
testids = range(0, X.shape[0])
# find and remove bad variables from the response variables
# note: the covariates are assumed to have already been checked
nz = np.where(
np.bitwise_and(np.isfinite(Y).any(axis=0),
np.var(Y, axis=0) != 0))[0]
# starting hyperparameters. Could also do random restarts here
covfunc = CovSum(X, ('CovLin', 'CovSqExpARD'))
hyp0 = np.zeros(covfunc.get_n_params() + 1)
# run cross-validation loop
Yhat = np.zeros_like(Y)
S2 = np.zeros_like(Y)
Z = np.zeros_like(Y)
nlZ = np.zeros((Nmod, cvfolds))
Hyp = np.zeros((Nmod, len(hyp0), cvfolds))
for idx in enumerate(splits.split(X)):
fold = idx[0]
tr = idx[1][0]
te = idx[1][1]
# standardize responses and covariates, ignoring invalid entries
iy, jy = np.ix_(tr, nz)
mY = np.mean(Y[iy, jy], axis=0)
sY = np.std(Y[iy, jy], axis=0)
Yz = np.zeros_like(Y)
Yz[:, nz] = (Y[:, nz] - mY) / sY
mX = np.mean(X[tr, :], axis=0)
sX = np.std(X[tr, :], axis=0)
Xz = (X - mX) / sX
# estimate the models for all subjects
for i in range(0, len(nz)): # range(0, Nmod):
print("Estimating model ", i + 1, "of", len(nz))
gpr = GPR(hyp0, covfunc, Xz[tr, :], Yz[tr, nz[i]])
Hyp[nz[i], :, fold] = gpr.estimate(hyp0, covfunc, Xz[tr, :],
Yz[tr, nz[i]])
yhat, s2 = gpr.predict(Hyp[nz[i], :, fold], Xz[tr, :],
Yz[tr, nz[i]], Xz[te, :])
Yhat[te, nz[i]] = yhat * sY[i] + mY[i]
S2[te, nz[i]] = np.diag(s2) * sY[i]**2
Z[te, nz[i]] = (Y[te, nz[i]] - Yhat[te, nz[i]]) / \
np.sqrt(S2[te, nz[i]])
nlZ[nz[i], fold] = gpr.nlZ
# compute performance metrics
MSE = np.mean((Y[testids, :] - Yhat[testids, :])**2, axis=0)
RMSE = np.sqrt(MSE)
# for the remaining variables, we need to ignore zero variances
SMSE = np.zeros_like(MSE)
Rho = np.zeros(Nmod)
pRho = np.ones(Nmod)
iy, jy = np.ix_(testids, nz) # ids for tested samples with nonzero values
SMSE[nz] = MSE[nz] / np.var(Y[iy, jy], axis=0)
Rho[nz], pRho[nz] = compute_pearsonr(Y[iy, jy], Yhat[iy, jy])
# Set writing options
if saveoutput:
print("Writing output ...")
if fileio.file_type(respfile) == 'cifti' or \
fileio.file_type(respfile) == 'nifti':
exfile = respfile
else:
exfile = None
if outputsuffix is not None:
ext = str(outputsuffix) + fileio.file_extension(respfile)
else:
ext = fileio.file_extension(respfile)
# Write output
fileio.save(Yhat[testids, :].T,
'yhat' + ext,
example=exfile,
mask=maskvol)
fileio.save(S2[testids, :].T,
'ys2' + ext,
example=exfile,
mask=maskvol)
fileio.save(Z[testids, :].T, 'Z' + ext, example=exfile, mask=maskvol)
fileio.save(Rho, 'Rho' + ext, example=exfile, mask=maskvol)
fileio.save(pRho, 'pRho' + ext, example=exfile, mask=maskvol)
fileio.save(RMSE, 'rmse' + ext, example=exfile, mask=maskvol)
fileio.save(SMSE, 'smse' + ext, example=exfile, mask=maskvol)
if cvfolds is None:
fileio.save(Hyp, 'Hyp' + ext, example=exfile, mask=maskvol)
else:
for idx in enumerate(splits.split(X)):
fold = idx[0]
fileio.save(Hyp[:, :, fold],
'Hyp_' + str(fold + 1) + ext,
example=exfile,
mask=maskvol)
else:
output = (Yhat, S2, Z, Rho, pRho, RMSE, SMSE)
return output
config = parse_args(sys.argv[1:])
if config.help_req:
Configuration.print_usage()
sys.exit(0)
if config.export_fmt is not None:
if config.export_fmt == 'swf':
lec = fileio.load(config.file_to_load)
exporter.to_swf(lec, lec.adats, config.file_to_load[:-4] + '.swf')
elif config.export_fmt == 'pdf':
lec = fileio.load(config.file_to_load)
exporter.to_pdf(lec, config.file_to_load[:-4] + '.swf')
elif config.export_fmt in ['dcd', 'dcb', 'dcx', 'dar', 'dct']:
lec = fileio.load(config.file_to_load)
fileio.save(config.file_to_load[:-3] + config.export_fmt, lec,
lec.adats)
else:
print 'Unknown flag "--exp-%s"' % config.export_fmt
sys.exit(0)
# Something was passed, so use that to
if config.audio_module is not None:
try:
Audio = __import__(config.audio_module).Audio
except AttributeError:
config.audio_module = None
print 'audio module "%s" not found' % config.audio_module
if config.audio_module is None:
for a in Configuration.VALID_AV_MODULES:
try:
config = parse_args(sys.argv[1:])
if config.help_req:
Configuration.print_usage()
sys.exit(0)
if config.export_fmt is not None:
if config.export_fmt == 'swf':
lec = fileio.load(config.file_to_load)
exporter.to_swf(lec, lec.adats, config.file_to_load[:-4] + '.swf')
elif config.export_fmt == 'pdf':
lec = fileio.load(config.file_to_load)
exporter.to_pdf(lec, config.file_to_load[:-4] + '.swf')
elif config.export_fmt in ['dcd', 'dcb', 'dcx', 'dar', 'dct']:
lec = fileio.load(config.file_to_load)
fileio.save(config.file_to_load[:-3] + config.export_fmt, lec, lec.adats)
else:
print 'Unknown flag "--exp-%s"' % config.export_fmt
sys.exit(0)
# Something was passed, so use that to
if config.audio_module is not None:
try:
Audio = __import__(config.audio_module).Audio
except AttributeError:
config.audio_module = None
print 'audio module "%s" not found' % config.audio_module
if config.audio_module is None:
for a in Configuration.VALID_AV_MODULES:
try:
def doSave(self):
fileio.save(self.__processedValues, unicode(QtGui.QFileDialog.getSaveFileName()))
def estimate(respfile, covfile, maskfile=None, cvfolds=None,
testcov=None, testresp=None, alg='gpr', configparam=None,
saveoutput=True, outputsuffix=None):
""" Estimate a normative model
This will estimate a model in one of two settings according to the
particular parameters specified (see below):
* under k-fold cross-validation
required settings 1) respfile 2) covfile 3) cvfolds>2
* estimating a training dataset then applying to a second test dataset
required sessting 1) respfile 2) covfile 3) testcov 4) testresp
* estimating on a training dataset ouput of forward maps mean and se
required sessting 1) respfile 2) covfile 3) testcov
The models are estimated on the basis of data stored on disk in ascii or
neuroimaging data formats (nifti or cifti). Ascii data should be in
tab or space delimited format with the number of subjects in rows and the
number of variables in columns. Neuroimaging data will be reshaped
into the appropriate format
Basic usage::
estimate(respfile, covfile, [extra_arguments])
where the variables are defined below. Note that either the cfolds
parameter or (testcov, testresp) should be specified, but not both.
:param respfile: response variables for the normative model
:param covfile: covariates used to predict the response variable
:param maskfile: mask used to apply to the data (nifti only)
:param cvfolds: Number of cross-validation folds
:param testcov: Test covariates
:param testresp: Test responses
:param alg: Algorithm for normative model
:param configparam: Parameters controlling the estimation algorithm
:param saveoutput: Save the output to disk? Otherwise returned as arrays
:param outputsuffix: Text string to add to the output filenames
All outputs are written to disk in the same format as the input. These are:
:outputs: * yhat - predictive mean
* ys2 - predictive variance
* Hyp - hyperparameters
* Z - deviance scores
* Rho - Pearson correlation between true and predicted responses
* pRho - parametric p-value for this correlation
* rmse - root mean squared error between true/predicted responses
* smse - standardised mean squared error
The outputsuffix may be useful to estimate multiple normative models in the
same directory (e.g. for custom cross-validation schemes)
"""
# load data
print("Processing data in " + respfile)
X = fileio.load(covfile)
Y, maskvol = load_response_vars(respfile, maskfile)
if len(Y.shape) == 1:
Y = Y[:, np.newaxis]
if len(X.shape) == 1:
X = X[:, np.newaxis]
Nmod = Y.shape[1]
if testcov is not None:
# we have a separate test dataset
Xte = fileio.load(testcov)
testids = range(X.shape[0], X.shape[0]+Xte.shape[0])
trainids = range(0, X.shape[0])
if len(Xte.shape) == 1:
Xte = Xte[:, np.newaxis]
if testresp is not None:
Yte, testmask = load_response_vars(testresp, maskfile)
if len(Yte.shape) == 1:
Yte = Yte[:, np.newaxis]
else:
sub_te = Xte.shape[0]
Yte = np.zeros([sub_te, Nmod])
# treat as a single train-test split
splits = CustomCV((range(0, X.shape[0]),), (testids,))
Y = np.concatenate((Y, Yte), axis=0)
X = np.concatenate((X, Xte), axis=0)
# force the number of cross-validation folds to 1
if cvfolds is not None and cvfolds != 1:
print("Ignoring cross-valdation specification (test data given)")
cvfolds = 1
else:
# we are running under cross-validation
splits = KFold(n_splits=cvfolds)
testids = range(0, X.shape[0])
# find and remove bad variables from the response variables
# note: the covariates are assumed to have already been checked
nz = np.where(np.bitwise_and(np.isfinite(Y).any(axis=0),
np.var(Y, axis=0) != 0))[0]
# Initialise normative model
nm = norm_init(X, alg=alg, configparam=configparam)
# run cross-validation loop
Yhat = np.zeros_like(Y)
S2 = np.zeros_like(Y)
Hyp = np.zeros((Nmod, nm.n_params, cvfolds))
Z = np.zeros_like(Y)
nlZ = np.zeros((Nmod, cvfolds))
for idx in enumerate(splits.split(X)):
fold = idx[0]
tr = idx[1][0]
te = idx[1][1]
# standardize responses and covariates, ignoring invalid entries
iy, jy = np.ix_(tr, nz)
mY = np.mean(Y[iy, jy], axis=0)
sY = np.std(Y[iy, jy], axis=0)
Yz = np.zeros_like(Y)
Yz[:, nz] = (Y[:, nz] - mY) / sY
mX = np.mean(X[tr, :], axis=0)
sX = np.std(X[tr, :], axis=0)
Xz = (X - mX) / sX
# estimate the models for all subjects
for i in range(0, len(nz)): # range(0, Nmod):
print("Estimating model ", i+1, "of", len(nz))
try:
nm = norm_init(Xz[tr, :], Yz[tr, nz[i]], alg=alg, configparam=configparam)
Hyp[nz[i], :, fold] = nm.estimate(Xz[tr, :], Yz[tr, nz[i]])
# Work around to get stats for subject in th emodel and out. Instead of te for all :
#yhat, s2 = nm.predict(Xz[tr, :], Yz[tr, nz[i]], Xz[te, :], Hyp[nz[i], :, fold])
yhat, s2 = nm.predict(Xz[tr, :], Yz[tr, nz[i]], Xz, Hyp[nz[i], :, fold])
#Yhat[te, nz[i]] = yhat * sY[i] + mY[i]
Yhat[:, nz[i]] = yhat * sY[i] + mY[i]
#S2[te, nz[i]] = s2 * sY[i]**2
S2[:, nz[i]] = s2 * sY[i] ** 2
nlZ[nz[i], fold] = nm.neg_log_lik
if testcov is None:
#Z[te, nz[i]] = (Y[te, nz[i]] - Yhat[te, nz[i]]) / np.sqrt(S2[te, nz[i]])
Z[:, nz[i]] = (Y[:, nz[i]] - Yhat[:, nz[i]]) / np.sqrt(S2[:, nz[i]])
else:
if testresp is not None:
#Z[te, nz[i]] = (Y[te, nz[i]] - Yhat[te, nz[i]]) / np.sqrt(S2[te, nz[i]])
Z[:, nz[i]] = (Y[:, nz[i]] - Yhat[:, nz[i]]) / np.sqrt(S2[:, nz[i]])
except Exception as e:
exc_type, exc_obj, exc_tb = sys.exc_info()
fname = os.path.split(exc_tb.tb_frame.f_code.co_filename)[1]
print("Model ", i+1, "of", len(nz),
"FAILED!..skipping and writing NaN to outputs")
print("Exception:")
print(e)
print(exc_type, fname, exc_tb.tb_lineno)
Hyp[nz[i], :, fold] = float('nan')
Yhat[te, nz[i]] = float('nan')
S2[te, nz[i]] = float('nan')
nlZ[nz[i], fold] = float('nan')
if testcov is None:
Z[te, nz[i]] = float('nan')
else:
if testresp is not None:
Z[te, nz[i]] = float('nan')
# compute performance metrics
if testcov is None:
MSE = np.mean((Y[testids, :] - Yhat[testids, :])**2, axis=0)
MSE_tr = np.mean((Y[0, X.shape[0], :] - Yhat[0, X.shape[0], :])**2, axis=0)
RMSE = np.sqrt(MSE)
# for the remaining variables, we need to ignore zero variances
SMSE = np.zeros_like(MSE)
Rho = np.zeros(Nmod)
pRho = np.ones(Nmod)
iy, jy = np.ix_(testids, nz) # ids for tested samples nonzero values
SMSE[nz] = MSE[nz] / np.var(Y[iy, jy], axis=0)
Rho[nz], pRho[nz] = compute_pearsonr(Y[iy, jy], Yhat[iy, jy])
else:
if testresp is not None:
MSE = np.mean((Y[testids, :] - Yhat[testids, :])**2, axis=0)
MSE_tr = np.mean((Y[0:X.shape[0], :] - Yhat[0:X.shape[0], :]) ** 2, axis=0)
RMSE = np.sqrt(MSE)
RMSE_tr = np.sqrt(MSE_tr)
# for the remaining variables, we need to ignore zero variances
SMSE = np.zeros_like(MSE)
SMSE_tr = np.zeros_like(RMSE_tr)
Rho = np.zeros(Nmod)
Rho_tr = np.zeros(Nmod)
pRho = np.ones(Nmod)
pRho_tr = np.ones(Nmod)
iy, jy = np.ix_(testids, nz) # ids tested samples nonzero values
iy_tr, jy_tr = np.ix_(range(0, X.shape[0]), nz)
SMSE[nz] = MSE[nz] / np.var(Y[iy, jy], axis=0)
SMSE_tr[nz] = MSE[nz] / np.var(Y[iy_tr, jy_tr], axis=0)
Rho[nz], pRho[nz] = compute_pearsonr(Y[iy, jy], Yhat[iy, jy])
Rho_tr[nz], pRho_tr[nz] = compute_pearsonr(Y[iy_tr, jy_tr], Yhat[iy_tr, jy_tr])
# Set writing options
if saveoutput:
print("Writing output ...")
if fileio.file_type(respfile) == 'cifti' or \
fileio.file_type(respfile) == 'nifti':
exfile = respfile
else:
exfile = None
if outputsuffix is not None:
ext = str(outputsuffix) + fileio.file_extension(respfile)
else:
ext = fileio.file_extension(respfile)
# Write output
if testcov is None:
fileio.save(Yhat[testids, :].T, 'yhat' + ext,
example=exfile, mask=maskvol)
fileio.save(S2[testids, :].T, 'ys2' + ext,
example=exfile, mask=maskvol)
fileio.save(Z[testids, :].T, 'Z' + ext, example=exfile,
mask=maskvol)
fileio.save(Rho, 'Rho' + ext, example=exfile, mask=maskvol)
fileio.save(pRho, 'pRho' + ext, example=exfile, mask=maskvol)
fileio.save(RMSE, 'rmse' + ext, example=exfile, mask=maskvol)
fileio.save(SMSE, 'smse' + ext, example=exfile, mask=maskvol)
if cvfolds is None:
fileio.save(Hyp[:,:,0], 'Hyp' + ext, example=exfile, mask=maskvol)
else:
for idx in enumerate(splits.split(X)):
fold = idx[0]
fileio.save(Hyp[:, :, fold], 'Hyp_' + str(fold+1) +
ext, example=exfile, mask=maskvol)
else:
if testresp is None:
fileio.save(Yhat[testids, :].T, 'yhat' + ext,
example=exfile, mask=maskvol)
fileio.save(S2[testids, :].T, 'ys2' + ext,
example=exfile, mask=maskvol)
fileio.save(Hyp[:,:,0], 'Hyp' + ext,
example=exfile, mask=maskvol)
else:
fileio.save(Yhat[testids, :].T, 'yhat' + ext, example=exfile, mask=maskvol)
fileio.save(Yhat[trainids, :].T, 'yhat_controls' + ext, example=exfile, mask=maskvol)
fileio.save(S2[testids, :].T, 'ys2' + ext, example=exfile, mask=maskvol)
fileio.save(S2[trainids, :].T, 'ys2_controls' + ext, example=exfile, mask=maskvol)
fileio.save(Z[testids, :].T, 'Z' + ext, example=exfile, mask=maskvol)
fileio.save(Z[trainids, :].T, 'Z_controls' + ext, example=exfile, mask=maskvol)
fileio.save(Rho, 'Rho' + ext, example=exfile, mask=maskvol)
fileio.save(Rho_tr, 'Rho_controls' + ext, example=exfile, mask=maskvol)
fileio.save(pRho, 'pRho' + ext, example=exfile, mask=maskvol)
fileio.save(pRho_tr, 'pRho_controls' + ext, example=exfile, mask=maskvol)
fileio.save(RMSE, 'rmse' + ext, example=exfile, mask=maskvol)
fileio.save(RMSE_tr, 'rmse_controls' + ext, example=exfile, mask=maskvol)
fileio.save(SMSE, 'smse' + ext, example=exfile, mask=maskvol)
fileio.save(SMSE_tr, 'smse_controls' + ext, example=exfile, mask=maskvol)
if cvfolds is None:
fileio.save(Hyp[:,:,0], 'Hyp' + ext,
example=exfile, mask=maskvol)
else:
for idx in enumerate(splits.split(X)):
fold = idx[0]
fileio.save(Hyp[:, :, fold], 'Hyp_'+ str(fold+1) +
ext, example=exfile, mask=maskvol)
else:
if testcov is None:
output = (Yhat, S2, Hyp, Z, Rho, pRho, RMSE, SMSE)
else:
if testresp is None:
output = (Yhat, S2, Hyp)
else:
output = (Yhat, S2, Hyp, Z, Rho, pRho, RMSE, SMSE)
return output
def processSundanceData():
files = findSubExt(getSundanceDir(), "data", ext=".p")
data = OrderedDict()
for ifile in files:
htmldata = get(ifile)
bsdata = getHTML(htmldata)
years = []
for h2 in bsdata.findAll("h2"):
span = h2.find("span")
try:
year = int(span.string)
except:
continue
years.append(year)
for j,ul in enumerate(bsdata.findAll("ul")):
try:
year = years[j]
except:
break
data[year] = {}
lis = ul.findAll("li")
for li in lis:
try:
txt = li.text
txt = re.sub("\xe2\x80\x93", " :: ", txt)
txt = re.sub(u"(\u2018|\u2013)", " :: ", txt)
except:
print "Error with",li
continue
vals = txt.split(" :: ")
if len(vals) > 2:
vals[1] = "-".join(vals[1:])
vals = vals[:2]
vals = [x.strip() for x in vals]
if len(vals) != 2:
raise ValueError(vals)
cat = vals[0]
movie = vals[1]
if cat.find("Piper-Heidsieck") != -1:
continue
if cat.find("Alfred P. Sloan") != -1:
cat = "Alfred P. Sloan Prize"
if cat in ["World Cinema Dramatic Screenwriting Award",
"Sundance Institute/Mahindra Global Filmmaking Awards",
"World Cinema Documentary Editing Award",
"Excellence in Cinematography Award: Documentary",
"Excellence in Cinematography Award: Dramatic",
"World Cinema Cinematography Award: Documentary",
"World Cinema Cinematography Award: Dramatic",
"World Cinema Directing Award: Dramatic",
"World Cinema Directing Award: Documentary",
"World Dramatic Special Jury Prizes for Breakout Performances",
"Dramatic Special Jury Prize for Breakout Performance",
"Excellence in Cinematography Award Dramatic",
"xcellence in Cinematography Award Documentary",
"Documentary Editing Award",
"Waldo Salt Screenwriting Award: Dramatic",
"World Cinema Screenwriting Award",
"Directing Award Documentary",
"Directing Award Dramatic"]:
vals = movie.split(" for ")
if len(vals) == 2:
movie = vals[1]
elif len(vals) == 1:
movie = vals[0]
else:
print "Error in",cat,"===>",movie
continue
if cat in ["Special Jury Prize for Acting"]:
movie = movie.replace("for her performance ", "")
vals = movie.split(" in ")
if len(vals) == 2:
movie = vals[1]
vals = movie.split(" for ")
if len(vals) == 2:
movie = vals[1]
if movie.find("retitled") != -1:
movie = movie.split("retitled ")[1]
movie = movie[:-1]
movie = movie.replace(" (tie)", "")
if movie.find(" director of ") != -1:
movie = movie.split(" director of ")[1]
print years[j],'\t',cat,'\t\t',movie,'\t\t'
try:
data[year][str(cat)] = str(movie)
except:
data[year][str(cat)] = movie
savename = setFile(getSundanceDir(), "winners.yaml")
print "Saving",len(data),"yearly results to",savename
save(savename, data)