def landOnHandle(trim = True, spacing = .5, exclOverlap = False, \
exclY = False, colList = ["#ef2929", "#3465a4", "#f57900", "#73d216"],\
lLegend = ["Exp1: relative to center", "Exp1: relative to CoG", \
"Exp2: relative to CoG", "Exp3: relative to CoG"], yLim = [-.3, .3],\
xLabels = ["1", "2", "3"], xTitle = "saccade", \
yTitle = "normalised landings towards handle"):
"""
Landing position as a funciton of orientation, across or per object group.
Arguments:
Keyword arguments:
trim ---
exclOverlap --- indicates whether or not to exclude gap-overlap trials
in the simulation.
"""
fig = plt.figure(figsize = (5,10))
title = "Average towards-handle landings - exclOverlap = %s - exclY = %s" \
% (exclOverlap, exclY)
plt.suptitle(title)
copyColList = colList[:]
dm_sim = getDM.getDM("004C")
## for handle in ["right", "left"]:
## handle_dm = dm_sim.select("handle_side == '%s'" % handle)
## print handle_dm["endX1NormToHandle"].mean()
#
# cm = dm_sim.collapse(['object'], 'endX1NormToHandle')
# cm._print(sign=5)
#
# dm_sim = dm_sim.select('object == "fork"')
# for _dm in dm_sim:
# print '%d: %.4f' % (_dm['count_trial_sequence'], \
# _dm['endX1NormToHandle'])
# cm = dm_sim.collapse(['gap', 'handle_side'], 'endX1NormToHandle')
# cm._print(sign=5)
# sys.exit()
for exp in ["004A", "004B", "004C"]:
if exp != "004B":
continue
# l = []
# if exp != "004A":
# continue
dm = getDM.getDM(exp)
for vf in dm.unique("visual_field"):
_dm = dm.select("visual_field == '%s'" % vf)
plt.hist(_dm["y_stim"], bins = 50)
plt.show()
sys.exit()
if exp == "004A":
dvList = ["abs", "corr"]
else:
dvList = ["abs"]
for dvType in dvList:
lMeans = []
for sacc in ["1", "2", "3"]:
sim_avg = float(dm_sim["endX%sNormToHandle" % sacc].mean())
print sim_avg
#raw_input()
if dvType == "corr":
dv = "endX%sCorrNormToHandle" % sacc
else:
dv = "endX%sNormToHandle" % sacc
# dv must not contain ''s:
on_dm = onObject.onObject(dm, sacc)
if trim:
trim_dm = on_dm.selectByStdDev(keys = ["file"], dv = dv)
else:
trim_dm = on_dm
pm = PivotMatrix(trim_dm, ["catch_trial"], ["file"], dv, \
colsWithin = False, err = 'se')
a = pm.asArray()
# print pm
M = float(a[-2][2])
SE = float(a[-1][2])
# print M
# print SE
#sys.exit()
#M = cdm["mean"].mean()
#SE = cdm["se"].mean()
#print M
#print SE
print "exp ", exp
print "sacc = ", sacc
print "dv = ", dv
am = AnovaMatrix(trim_dm, ["handle_side"], "endX%sNorm" % \
sacc, "file")#._print(ret=True)
# print am
p = am.asArray()[2][3]
p_corr = float(p) * 6
# print p_corr
print 'T test scipy'
cm = trim_dm.collapse(["file"], dv)
M = cm["mean"].mean()
SE = cm['mean'].std() / np.sqrt(len(cm))
ref = 0
t, p = scipy.stats.ttest_1samp(cm['mean'], ref)
p_corr = min(1, float(p * 6.))
# print p_corr
print "M = %.3f, SE = %.3f, t(17) = %.2f, p = %6.4f" % \
(M, SE, t, p_corr)
raw_input()
