def plotCorr(ssubs,
compare,
rng=slice(0, T),
ca_or_spikes='ca',
labels=None,
title='Shapes obtained on',
colors=[orange, blue, vermillon],
ls='-'):
""" map based on spatial distance, removed neurons are set to corr=0 """
def foo(ssub, comp, idx=None):
N, T = comp.shape
cor = np.zeros((N, len(dsls)))
for i, ds in enumerate(dsls):
if len(ssub[ds][0]) == len(comp):
cor[:, i] = np.nan_to_num([
np.corrcoef(
ssub[ds][0 if ca_or_spikes == 'ca' else 2][n, rng],
comp[n, rng])[0, 1] for n in range(N)
])
else: # necessary if component has been removed on small batch
for n, k in enumerate(idx):
if k is not None: # component removed -> keep corr=0
cor[n, i] = np.nan_to_num(
np.corrcoef(
ssub[ds][0 if ca_or_spikes == 'ca' else 2][
k, rng], comp[n, rng])[0, 1])
return cor
for i, ssub in enumerate(ssubs):
cor = foo(ssub, compare, [None, idxH, idxQ][i])
plt.plot(dsls,
np.mean(cor, 0),
ls=ls,
lw=4,
c=colors[i],
label=None if labels is None else labels[i],
clip_on=False)
noclip(
plt.errorbar(dsls,
np.mean(cor, 0),
yerr=np.std(cor, 0) / np.sqrt(len(cor)),
lw=3,
capthick=2,
fmt='o',
c=colors[i],
clip_on=False))
plt.xlabel('Spatial decimation')
simpleaxis(plt.gca())
plt.xticks(dsls,
['1', '', '', '', '', '8x8', '', '16x16', '24x24', '32x32'])
plt.yticks(*[np.arange(np.round(plt.ylim()[1], 1),
plt.ylim()[0], -.2)] * 2)
plt.xlim(dsls[0], dsls[-1])
plt.ylim(.25, 1)
plt.ylabel('Correlation w/ undecimated $C_1$/$S_1$', y=.42, labelpad=1)
plt.subplots_adjust(.13, .15, .94, .96)
if labels is not None:
lg = plt.legend(loc=(-.01, -.01), title=title)
return lg
def adjust():
plt.xticks(*[[0, 5, 10]] * 2)
plt.yticks(*[[1, 1.01], ['1.00', 1.01]])
plt.xlim(0, 8.5)
plt.ylim(l, u)
plt.xlabel('Wall time [s]')
simpleaxis(plt.gca())
plt.subplots_adjust(.1, .155, .99, .99)
def plotCorr(ssubs,
compare,
dsls,
labels=None,
colors=[cyan, orange, green],
ca_or_spikes='ca',
ls='-'):
def foo(ssub, comp):
N, T = comp.shape
cor = np.zeros((N, len(dsls)))
for i, ds in enumerate(dsls):
if len(ssub[ds][0]) == len(comp):
cor[:, i] = np.nan_to_num([
np.corrcoef(ssub[ds][0 if ca_or_spikes == 'ca' else 2][n],
comp[n])[0, 1] for n in range(N)
])
else: # necessary if update_spatial_components removed a component
mapIdx = [
np.argmax([np.corrcoef(s, tC)[0, 1] for tC in comp])
for s in ssub[ds][0 if ca_or_spikes == 'ca' else 2]
]
for n in range(len(ssub[ds][0])):
cor[mapIdx[n], i] = np.nan_to_num(
np.corrcoef(
ssub[ds][0 if ca_or_spikes == 'ca' else 2][n],
comp[mapIdx[n]])[0, 1])
return cor
for i, ssub in enumerate(ssubs):
cor = foo(ssub, compare)
plt.plot(dsls,
np.mean(cor, 0),
ls=ls,
lw=4,
c=colors[i],
label=None if labels is None else labels[i],
clip_on=False)
noclip(
plt.errorbar(dsls,
np.mean(cor, 0),
yerr=np.std(cor, 0) / np.sqrt(len(cor)),
lw=3,
capthick=2,
fmt='o',
c=colors[i],
clip_on=False))
plt.xlabel('Spatial decimation')
simpleaxis(plt.gca())
plt.xticks(dsls,
['1', '', '', '', '', '8x8', '', '16x16', '24x24', '32x32'])
plt.yticks(*[[.4, .6, .8, 1.0]] * 2)
plt.xlim(dsls[0], dsls[-1])
plt.ylabel('Correlation w/ undecimated $C_1$/$S_1$', y=.42, labelpad=1)
if labels is not None:
lg = plt.legend(loc=(0, 0))
return lg
def plotChoice(w): # arg which 3b or 3e
offers = eval('offers' + w)
ratio = eval('ratio' + w)
pB = eval('pB' + w)
v = np.array([offers[1:-1, 0], ratio * offers[1:-1, 1]]).T
logratio = np.log(v[:, 0] / v[:, 1])
x = np.transpose(s[w], (1, 0, 2, 3))
rts = np.array([[get_t(a) for a in x[r]]
for r in range(len(x))], dtype=int)
choice = rts[:, :, 1] > rts[:, :, 2]
pBsim = np.mean(choice, 0)
# plot
pl.figure()
pl.plot(logratio, pB[1:-1], '--', marker='s', ms=30, mfc='w',
mec=col[0], mew=3, color=col[0], zorder=10, clip_on=False)
pl.plot(logratio, pBsim[1:-1], color=col[1], marker='s', ms=22,
mec=col[1], clip_on=False, zorder=10)
pl.plot([logratio[0] - .5, logratio[-1] + .5], [pB[0], pB[-1]], 's', ms=30,
mfc='w', mec=col[0], mew=3, color=col[0], zorder=10, clip_on=False)
pl.plot([logratio[0] - .5, logratio[-1] + .5], [pBsim[0], pBsim[-1]], 's',
color=col[1], ms=22, mec=col[1], clip_on=False, zorder=10)
pl.xticks([logratio[0] - .5, -1, 0, 1, logratio[-1] + .5, ],
[r"$-\infty$", -1, 0, 1, r"$\infty$"])
tt = pl.gca().get_xaxis().majorTicks
for i in range(len(tt)):
tt[i].set_pad(17)
pl.yticks([0, .5, 1], [0, 50, 100])
pl.xlim(logratio[0] - .65, logratio[-1] + .5)
pl.ylim(0, 1)
pl.xlabel('log(V(B)/V(A))', labelpad=-2)
pl.ylabel('B choice [\%]', labelpad=-15)
pl.text(.05, .83, 'A=' + str(ratio) + 'B', transform=pl.gca().transAxes)
pl.subplots_adjust(.17, .23, .96, .95)
simpleaxis(pl.gca())
# broken axis
pl.plot([logratio[0] - .27, logratio[0] - .23],
[0, 0], c='w', lw=2, clip_on=False, zorder=11)
pl.plot([logratio[0] - .29, logratio[0] - .25],
[-.02, .02], c='k', lw=2, clip_on=False, zorder=11)
pl.plot([logratio[0] - .25, logratio[0] - .21],
[-.02, .02], c='k', lw=2, clip_on=False, zorder=11)
pl.plot([logratio[-1] + .28, logratio[-1] + .24],
[0, 0], c='w', lw=2, clip_on=False, zorder=11)
pl.plot([logratio[-1] + .3, logratio[-1] + .26],
[.02, -.02], c='k', lw=2, clip_on=False, zorder=11)
pl.plot([logratio[-1] + .26, logratio[-1] + .22],
[.02, -.02], c='k', lw=2, clip_on=False, zorder=11)
if savefig:
pl.savefig('Padoa' + w + '.pdf', dpi=600)
else:
pl.show()
return choice
def plotCorr(ssub, ssub0, r=pearsonr, ds1phase=[1, 2, 3, 4], loc=(.1, .01)):
def foo(ssub, comp, dsls=dsls, ca_or_spikes='ca'):
N, T = comp.shape
cor = np.zeros((N, len(dsls))) * np.nan
for i, ds in enumerate(dsls):
if len(ssub[ds][0]) == len(comp):
cor[:, i] = np.array(
[r(ssub[ds][0 if ca_or_spikes == 'ca' else 2][n],
comp[n])[0] for n in range(N)])
else: # necessary if update_spatial_components removed a component
mapIdx = [np.argmax([np.corrcoef(s, tC)[0, 1] for tC in comp])
for s in ssub[ds][0 if ca_or_spikes == 'ca' else 2]]
for n in range(len(ssub[ds][0])):
cor[mapIdx[n], i] = np.array(
r(ssub[ds][0 if ca_or_spikes == 'ca' else 2][n],
comp[mapIdx[n]])[0])
return np.nan_to_num(cor)
cor = foo(ssub0, trueC, ds1phase)
l1, = plt.plot(ds1phase, np.median(np.nan_to_num(cor), 0), lw=4, c=cyan,
label='1 phase imaging')
IQRfill(cor, dsls, cyan)
cor = foo(ssub, trueC)
l2, = plt.plot(dsls, np.median(cor, 0), lw=4, c=orange, label='2 phase imaging')
IQRfill(cor, dsls, orange)
cor = foo(ssub0, trueSpikes, ds1phase, ca_or_spikes='spikes')
plt.plot(ds1phase, np.median(cor, 0), lw=4, c=cyan, ls='--')
IQRfill(cor, dsls, cyan, ls='--', hatch='///')
cor = foo(ssub, trueSpikes, ca_or_spikes='spikes')
plt.plot(dsls, np.median(cor, 0), lw=4, c=orange, ls='--')
IQRfill(cor, dsls, orange, ls='--', hatch='\\\\\\')
l3, = plt.plot([0, 1], [-1, -1], lw=4, c='k', label='denoised')
l4, = plt.plot([0, 1], [-1, -1], lw=4, c='k', ls='--', label='deconvolved')
plt.xlabel('Spatial decimation')
simpleaxis(plt.gca())
plt.xticks(dsls, ['1', '', '', '', '', '8x8', '', '16x16', '24x24', '32x32'])
plt.ylim(.3, 1)
plt.yticks(
*[np.round(np.arange(np.round(plt.ylim()[1], 1), plt.ylim()[0], -.2), 1)] * 2)
plt.xlim(dsls[0], dsls[-1])
plt.ylabel('Correlation w/ undecimated $C_1$/$S_1$', y=.42, labelpad=1)
plt.legend(handles=[l3, l4, l1, l2], loc=loc, ncol=1)
plt.subplots_adjust(.13, .15, .94, .96)
return l1, l2, l3, l4
def plot_trace(n=0, lg=False):
plt.plot(trueC[n], c=col[2], clip_on=False, zorder=5, label='Truth')
plt.plot(solution, c=col[0], clip_on=False, zorder=7, label='Estimate')
plt.plot(y, c=col[7], alpha=.7, clip_on=False, zorder=-10, label='Data')
if lg:
plt.legend(frameon=False, ncol=3, loc=(.1, .62), columnspacing=.8)
spks = np.append(0, solution[1:] - g * solution[:-1])
plt.text(800,
2.2,
'Correlation: %.3f' % (np.corrcoef(trueSpikes[n], spks)[0, 1]),
size=24)
plt.gca().set_xticklabels([])
simpleaxis(plt.gca())
plt.ylim(0, 2.85)
plt.xlim(0, 1500)
plt.yticks([0, 2], [0, 2])
plt.xticks([300, 600, 900, 1200], ['', ''])
def cb(y, active_set, counter, current):
solution = np.empty(len(y))
for i, (v, w, f, l) in enumerate(active_set):
solution[f:f + l] = (v if i else max(v, 0)) / w * g**np.arange(l)
color = y.copy()
ax1.plot(solution, c='k', zorder=-11, lw=1.3, clip_on=False)
ax1.scatter(np.arange(len(y)), solution, s=40, cmap=plt.cm.Spectral,
c=color, clip_on=False, zorder=11)
ax1.scatter([np.arange(len(y))[current]], [solution[current]],
s=120, lw=2.5, marker='+', color='b', clip_on=False, zorder=11)
for a in active_set[::2]:
ax1.axvspan(a[2], a[2] + a[3], alpha=0.1, color='k', zorder=-11)
for x in np.where(trueSpikes)[0]:
ax1.plot([x, x], [0, 2.3], lw=1.5, c='r', zorder=-12)
ax1.set_xlim((0, len(y) - .5))
ax1.set_ylim((0, 2.3))
simpleaxis(ax1)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_ylabel('Fluorescence')
for i, s in enumerate(np.r_[[0], solution[1:] - g * solution[:-1]]):
ax2.plot([i, i], [0, s], c='k', zorder=-11, lw=1.4, clip_on=False)
ax2.scatter(np.arange(len(y)), np.r_[[0], solution[1:] - g * solution[:-1]],
s=40, cmap=plt.cm.Spectral, c=color, clip_on=False, zorder=11)
ax2.scatter([np.arange(len(y))[current]], [np.r_[[0], solution[1:] - g * solution[:-1]][current]],
s=120, lw=2.5, marker='+', color='b', clip_on=False, zorder=11)
for a in active_set[::2]:
ax2.axvspan(a[2], a[2] + a[3], alpha=0.1, color='k', zorder=-11)
for x in np.where(trueSpikes)[0]:
ax2.plot([x, x], [0, 1.55], lw=1.5, c='r', zorder=-12)
ax2.set_xlim((0, len(y) - .5))
ax2.set_ylim((0, 1.55))
simpleaxis(ax2)
ax2.set_xticks([])
ax2.set_yticks([])
ax2.set_xlabel('Time', labelpad=35, x=.5)
ax2.set_ylabel('Spikes')
plt.subplots_adjust(left=0.032, right=.995, top=.995, bottom=0.19, hspace=0.22)
fig.canvas.draw()
if save_figs:
plt.savefig('video/%03d.pdf' % counter)
# import time
# time.sleep(.03)
ax1.clear()
ax2.clear()
def cb(y, active_set, counter, current):
solution = np.empty(len(y))
for v, w, f, l in active_set:
solution[f:f + l] = max(v, 0) / w * g**np.arange(l)
plt.figure(figsize=(3, 3))
color = y.copy()
plt.plot(solution, c='k', zorder=-11, lw=1.2)
plt.scatter(np.arange(len(y)), solution, s=60, cmap=plt.cm.Spectral,
c=color, clip_on=False, zorder=11)
plt.scatter([np.arange(len(y))[current]], [solution[current]],
s=200, lw=2.5, marker='+', color='b', clip_on=False, zorder=11)
for a in active_set[::2]:
plt.axvspan(a[2], a[2] + a[3], alpha=0.1, color='k', zorder=-11)
for x in np.where(trueSpikes)[0]:
plt.plot([x, x], [0, 1.65], lw=1.5, c='r', zorder=-12)
plt.xlim((0, len(y) - .5))
plt.ylim((0, 1.65))
simpleaxis(plt.gca())
plt.xticks([])
plt.yticks([])
# plt.tight_layout()
if save_figs:
plt.savefig('fig/%d.pdf' % counter)
plt.show()
def plotTrace(lg=False):
fig = plt.figure(figsize=(10, 9))
fig.add_axes([.13, .7, .86, .29])
plt.plot(c, c=col[0], label='L1')
plt.plot(c_t, c=col[1], label='Thresh.')
plt.plot(trueC[0], c=col[2], lw=3, label='Truth', zorder=-5)
plt.plot(y, c=col[7], alpha=.7, zorder=-10, label='Data')
if lg:
plt.legend(frameon=False, ncol=4, loc=(.05, .82))
plt.gca().set_xticklabels([])
simpleaxis(plt.gca())
plt.yticks([0, int(y.max())], [0, int(y.max())])
plt.xticks(range(150, 500, 150), [''] * 3)
plt.ylabel('Fluor.')
plt.xlim(0, 452)
fig.add_axes([.13, .39, .86, .29])
for i, ss in enumerate(s[:500]):
if ss > 1e-2:
plt.plot([i, i], [2.5, 2.5 + ss], c=col[0], zorder=10)
plt.plot([0, 450], [2.5, 2.5], c=col[0], zorder=10)
for i, ss in enumerate(s_t[:500]):
if ss > 1e-2:
plt.plot([i, i], [1.25, 1.25 + ss], c=col[1], zorder=10)
plt.plot([0, 450], [1.25, 1.25], c=col[1], zorder=10)
for i, ss in enumerate(trueSpikes[0, :500]):
if ss > 1e-2:
plt.plot([i, i], [0, ss], c=col[2], clip_on=False, zorder=10)
plt.plot([0, 450], [0, 0], c=col[2], clip_on=False, zorder=10)
plt.gca().set_xticklabels([])
simpleaxis(plt.gca())
plt.yticks([0, 1.25, 2.5], ['Truth', 'Thresh.', 'L1'])
for tick in plt.gca().yaxis.get_major_ticks():
tick.label1.set_verticalalignment('bottom')
plt.xticks(range(150, 500, 150), [''] * 3)
plt.ylim(0, 3.5)
plt.xlim(0, 452)
fig.add_axes([.13, .08, .86, .29])
for i, r in enumerate(res):
for rr in r:
plt.plot([rr, rr], [.1 * i - .04, .1 * i + .04], c='k')
for rr in np.where(trueSpikes[0])[0]:
plt.plot([rr, rr], [-.08, -.16], c='r')
plt.gca().set_xticklabels([])
simpleaxis(plt.gca())
plt.yticks([0, .5, 1], [0, 0.5, 1.0])
plt.xticks(range(0, 500, 150), [0, 5, 10, ''])
plt.ylim(-.2, 1.1)
plt.xlim(0, 452)
plt.ylabel(r'$s_{\min}$')
plt.xlabel('Time [s]', labelpad=-10)
plt.show()
pl.figure()
for c in range(8):
pl.plot(Ratels.mean(axis=0)[c], color=col[c])
pl.xlabel('Time from target [ms]', labelpad=0)
pl.ylabel('Firing rate [Hz]', labelpad=10)
pl.xticks([0, 250 / step, 500 / step], [0, 250, 500])
pl.yticks([0, 20, 40, 60, 80], [0, 20, 40, 60, 80])
pl.xlim(0, 500 / step)
pl.ylim(0, 64)
pl.plot(smooth_spikes([step * 100 / 1000. *
(.1 + .65 * np.exp(-(t * step - 500) ** 2 / tau_r ** 2))
for t in range(int(1000 / step))], 40, .2)[250 / step:750 / step],
color='black', zorder=-1, lw=2)
pl.subplots_adjust(.18, .21, .945, .99)
simpleaxis(pl.gca())
if savefig:
pl.savefig('Sohn.pdf', dpi=600)
else:
pl.show()
pl.figure()
pl.plot(Ratels.mean(axis=0)[3], color=col[3])
l, = pl.plot(Ratels3.mean(axis=0)[3], '--', color=col[3])
l.set_dashes([10, 10])
l, = pl.plot(Ratels4.mean(axis=0)[3], ':', color=col[3])
l.set_dashes([3, 3])
pl.xlabel('Time from target [ms]', labelpad=0)
pl.ylabel('Firing rate [Hz]', labelpad=10)
pl.xticks([0, 250 / step, 500 / step], [0, 250, 500])
fig1.subplots_adjust(.04, .1, 1, 1)
if savefig:
pl.savefig('u.pdf', dpi=600, transparent=True)
else:
pl.show()
# Spikes
spikes, u = cfn.run(net.W, step, 400, 100, 20, 2, 20, 2)
pl.figure()
for i in xrange(8):
map(lambda a: pl.plot([a, a], [.95 + i, .05 + i],
c=col[i]), np.where(spikes[:, i] == 1)[0])
pl.yticks(np.arange(.4, 8), ['0L', '0R', '1L', '1R', '2L', '2R', '3L', '3R'])
pl.gca().yaxis.set_tick_params(width=0)
pl.xticks(np.arange(0, len(u) + 1, 40 / step), [0, 40, 80])
simpleaxis(pl.gca())
pl.tight_layout(pad=.03)
if savefig:
pl.savefig('spikes.pdf', dpi=600, transparent=True)
else:
pl.show()
def f(W, x, ref):
y = (1 - 1. / 2000) * x + 1. / 2000\
* ((1 + ref) * np.dot(W, x * (x > 0)) - ref * x * (x > 0))
y[net.r] = 1
return y
x = np.zeros((1201, net.K))
np.random.seed(2)
color=col[1], mec=col[1], capthick=2, clip_on=False, zorder=11)
pl.xticks(range(len(offers3e)), [
str(a[0]) + ':' + str(a[1]) for i, a in enumerate(offers3e)])
tt = pl.gca().get_xaxis().majorTicks
for i in range(0, len(tt), 2):
tt[i].set_pad(40)
for i in range(1, len(tt), 2):
tt[i].set_pad(13)
pl.yticks([0, 10], [0, 10])
pl.xlim(-.3, len(offers3e) - .7)
pl.ylim(0, 14)
pl.xlabel('offers (\#B:\#A)', labelpad=-2)
pl.ylabel('Firing rate [Hz]', y=.45)
pl.text(.05, .83, 'A=' + str(ratio3e) + 'B', transform=pl.gca().transAxes)
pl.subplots_adjust(.17, .28, .98, .99)
simpleaxis(pl.gca())
lg = pl.legend(
['Experiment', 'Model'], bbox_to_anchor=(.63, .85), handlelength=2, frameon=False)
if savefig:
pl.savefig('Padoa3e2.pdf', dpi=600)
else:
pl.show()
x = s['3b']
y = 2 * np.transpose(x[:, :, :int(500 / step), :2].sum(axis=2), (1, 0, 2))
pl.figure()
pl.plot(range(1, len(offers3b) - 1),
f3b[1:-1], '--', color=col[0], zorder=10, clip_on=False)
pl.errorbar(range(len(offers3b)), f3b, yerr=fSEM3b, fmt='o', ms=30, color=col[0],
mfc='w', mec=col[0], mew=3, capthick=2, zorder=10, clip_on=False)
pl.errorbar(range(len(offers3b)), y.mean(axis=0).sum(axis=1) / pop3b,
zorder=11)
pl.xticks(range(len(offers3e)),
[str(a[0]) + ':' + str(a[1]) for i, a in enumerate(offers3e)])
tt = pl.gca().get_xaxis().majorTicks
for i in range(0, len(tt), 2):
tt[i].set_pad(40)
for i in range(1, len(tt), 2):
tt[i].set_pad(13)
pl.yticks([0, 10], [0, 10])
pl.xlim(-.3, len(offers3e) - .7)
pl.ylim(0, 14)
pl.xlabel('offers (\#B:\#A)', labelpad=-2)
pl.ylabel('Firing rate [Hz]', y=.45)
pl.text(.05, .83, 'A=' + str(ratio3e) + 'B', transform=pl.gca().transAxes)
pl.subplots_adjust(.17, .28, .98, .99)
simpleaxis(pl.gca())
lg = pl.legend(['Experiment', 'Model'],
bbox_to_anchor=(.63, .85),
handlelength=2,
frameon=False)
if savefig:
pl.savefig('Padoa3e2.pdf', dpi=600)
else:
pl.show()
x = s['3b']
y = 2 * np.transpose(x[:, :, :int(500 / step), :2].sum(axis=2), (1, 0, 2))
pl.figure()
pl.plot(range(1,
len(offers3b) - 1),
f3b[1:-1],
fig1.subplots_adjust(.04, .1, 1, 1)
if savefig:
pl.savefig('u.pdf', dpi=600, transparent=True)
else:
pl.show()
# Spikes
spikes, u = cfn.run(net.W, step, 400, 100, 20, 2, 20, 2)
pl.figure()
for i in xrange(8):
map(lambda a: pl.plot([a, a], [.95 + i, .05 + i], c=col[i]),
np.where(spikes[:, i] == 1)[0])
pl.yticks(np.arange(.4, 8), ['0L', '0R', '1L', '1R', '2L', '2R', '3L', '3R'])
pl.gca().yaxis.set_tick_params(width=0)
pl.xticks(np.arange(0, len(u) + 1, 40 / step), [0, 40, 80])
simpleaxis(pl.gca())
pl.tight_layout(pad=.03)
if savefig:
pl.savefig('spikes.pdf', dpi=600, transparent=True)
else:
pl.show()
def f(W, x, ref):
y = (1 - 1. / 2000) * x + 1. / 2000\
* ((1 + ref) * np.dot(W, x * (x > 0)) - ref * x * (x > 0))
y[net.r] = 1
return y
x = np.zeros((1201, net.K))
if i == 1:
l.set_dashes([14, 10])
plt.plot(z * ssubLR[4][2][neuronId] / scale - .7 * ub,
alpha=1.,
c=cyan,
zorder=-10)
plt.ylim(-.7 * ub, ub)
plt.legend(frameon=False,
ncol=len(dsls),
loc=[.073, .775],
columnspacing=14.2)
plt.xticks(range(0, T, ticksep * fps), ['', '', ''])
tmp = [500, 300, 100, 100][k]
plt.plot([0, 0], [0, 1], c='k', lw=5, clip_on=False, zorder=11)
plt.text(7, 1, '100\%', verticalalignment='center')
simpleaxis(plt.gca())
plt.gca().spines['left'].set_visible(False)
plt.yticks([])
for i, ds in enumerate(dsls):
ax = fig.add_axes([[.17, .467, .78][i], .838 - .244 * k, .035, .2625])
ss = downscale_local_mean(gaussian(shapes[neuronId], 1),
(ds, ds)).repeat(ds, 0).repeat(ds, 1)
ax.imshow(ss[map(lambda a: slice(*a), boxes[neuronId])],
cmap='hot',
interpolation='nearest')
ax.axis('off')
ax.text(
1.25,
.5,
'%.3f' %
(np.corrcoef(ssub[ds][0][neuronId], ssub[1][0][neuronId])[0, 1]),
def plotChoice(w): # arg which 3b or 3e
offers = eval('offers' + w)
ratio = eval('ratio' + w)
pB = eval('pB' + w)
v = np.array([offers[1:-1, 0], ratio * offers[1:-1, 1]]).T
logratio = np.log(v[:, 0] / v[:, 1])
x = np.transpose(s[w], (1, 0, 2, 3))
rts = np.array([[get_t(a) for a in x[r]] for r in range(len(x))],
dtype=int)
choice = rts[:, :, 1] > rts[:, :, 2]
pBsim = np.mean(choice, 0)
# plot
pl.figure()
pl.plot(logratio,
pB[1:-1],
'--',
marker='s',
ms=30,
mfc='w',
mec=col[0],
mew=3,
color=col[0],
zorder=10,
clip_on=False)
pl.plot(logratio,
pBsim[1:-1],
color=col[1],
marker='s',
ms=22,
mec=col[1],
clip_on=False,
zorder=10)
pl.plot([logratio[0] - .5, logratio[-1] + .5], [pB[0], pB[-1]],
's',
ms=30,
mfc='w',
mec=col[0],
mew=3,
color=col[0],
zorder=10,
clip_on=False)
pl.plot([logratio[0] - .5, logratio[-1] + .5], [pBsim[0], pBsim[-1]],
's',
color=col[1],
ms=22,
mec=col[1],
clip_on=False,
zorder=10)
pl.xticks([
logratio[0] - .5,
-1,
0,
1,
logratio[-1] + .5,
], [r"$-\infty$", -1, 0, 1, r"$\infty$"])
tt = pl.gca().get_xaxis().majorTicks
for i in range(len(tt)):
tt[i].set_pad(17)
pl.yticks([0, .5, 1], [0, 50, 100])
pl.xlim(logratio[0] - .65, logratio[-1] + .5)
pl.ylim(0, 1)
pl.xlabel('log(V(B)/V(A))', labelpad=-2)
pl.ylabel('B choice [\%]', labelpad=-15)
pl.text(.05, .83, 'A=' + str(ratio) + 'B', transform=pl.gca().transAxes)
pl.subplots_adjust(.17, .23, .96, .95)
simpleaxis(pl.gca())
# broken axis
pl.plot([logratio[0] - .27, logratio[0] - .23], [0, 0],
c='w',
lw=2,
clip_on=False,
zorder=11)
pl.plot([logratio[0] - .29, logratio[0] - .25], [-.02, .02],
c='k',
lw=2,
clip_on=False,
zorder=11)
pl.plot([logratio[0] - .25, logratio[0] - .21], [-.02, .02],
c='k',
lw=2,
clip_on=False,
zorder=11)
pl.plot([logratio[-1] + .28, logratio[-1] + .24], [0, 0],
c='w',
lw=2,
clip_on=False,
zorder=11)
pl.plot([logratio[-1] + .3, logratio[-1] + .26], [.02, -.02],
c='k',
lw=2,
clip_on=False,
zorder=11)
pl.plot([logratio[-1] + .26, logratio[-1] + .22], [.02, -.02],
c='k',
lw=2,
clip_on=False,
zorder=11)
if savefig:
pl.savefig('Padoa' + w + '.pdf', dpi=600)
else:
pl.show()
return choice
[[net.R4Pi2(net.get_policy2(step, i, T), net.pstart_state) for i in S]
for T in Tls])
perf = (rew - R0) / (Rmax - R0)
perf[0] = 0
np.save('results/performance', perf)
pl.figure()
errorfill(range(len(perf)),
np.mean(perf, axis=1),
yerr=np.std(perf, axis=1) / np.sqrt(len(perf[0])))
pl.xticks([0, 200, 400], [0, 200, 400])
pl.yticks([0, .5, 1.0], [0, .5, 1.0])
pl.xlim([0, 400])
pl.ylim([0, 1])
pl.xlabel('Time [ms]')
pl.ylabel('Performance')
simpleaxis(pl.gca())
pl.tight_layout(0)
pl.savefig('performance.pdf', dpi=600)
# learning via parallel sampling
eta, T, r0 = .01, 1000, 0
try:
perf = np.load('results/learn_performance.npy')
except IOError:
try:
rew = np.load('results/learn.npz')['R']
except IOError:
res = np.array([
net.parallel_sampling_keepU(step,
eta,
run,
fig = plt.figure(figsize=(6.5, 6))
for ssub, col in ((activityDSlr, cyan), (activityDS, orange)):
cor, mx = foo(ssub, activityDS[1], shapes)
if col == cyan:
mapIdx = mx
plt.plot(dsls,
np.median(cor, 0),
lw=3,
c=col,
clip_on=False,
label='1 phase\n imaging' if col == cyan else '2 phase\n imaging')
IQRfill(cor, dsls, col)
plt.xlabel('Spatial decimation')
ax = plt.gca()
simpleaxis(ax)
ax.patch.set_visible(False)
plt.xticks(dsls, ['1x1', '', '', '4x4', '6x6', '8x8', '12x12'])
plt.ylim(.55, 1)
plt.yticks(*[[.6, .8, 1.]] * 2)
plt.ylabel('Correlation', labelpad=0)
plt.legend(loc=(.01, .25), ncol=1)
plt.subplots_adjust(.145, .15, .885, .97)
ax2 = ax.twinx()
ax2.set_zorder(-1)
ax2.spines['top'].set_visible(False)
z = np.mean((activityDS[1].T.dot(shapes.reshape(len(shapes), -1)) -
data.reshape(activity.shape[1], -1))**2)
ax2.plot(dsls, [
np.mean((activityDSlr[ds].T.dot(shapesDSlr[ds].repeat(ds, 1).repeat(
ds, 2).reshape(len(shapes), -1)) - data.reshape(activity.shape[1], -1))