def train(self, test_users=None, b=None):
block_size = 5000
self.model = {}
self.test_users = test_users
if test_users is None:
test_users = range(self.n_users)
tu = {}
if b is not None:
for u in test_users:
tu[u] = set(test_users[u][:b])
else:
for u in test_users:
tu[u] = set(test_users[u])
for i, u in enumerate(tu):
if self.verbose and (i + 1) % 100 == 0:
print("%d/%d" % (i + 1, len(test_users)))
utils.fl()
#Xp = list(self.data.get_items(u)) #get list of songs
Xp = [int(self.K_indexing[str(idx)]) for idx in tu[u]]
mtx_indexes = [a / block_size for a in Xp]
pos_indexes = [a % block_size for a in Xp]
npos = len(Xp)
kp = np.zeros((npos, npos))
for j in range(npos):
#kp[j] = self.K[Xp[j],Xp].todense()
kp[j] = self.K[mtx_indexes[j]][pos_indexes[j], Xp]
kp = co.matrix(kp)
kn = self.q_[Xp, :]
I = self.lambda_p * utc.identity(npos)
P = kp + I
q = -kn
G = -utc.identity(npos)
h = utc.zeroes_vec(npos)
A = utc.ones_vec(npos).T
b = co.matrix(1.0)
solver.options['show_progress'] = False
sol = solver.qp(P, q, G, h, A, b)
band = np.zeros((npos, len(self.inverse_indexing)))
for j in range(npos):
band[j] = self.K[mtx_indexes[j]][pos_indexes[j], :]
self.model[u] = (co.matrix(band.T) * (sol['x'])) - self.q_
#print self.model[u]
# endfor
return self
def train(self, test_users=None, b=None, selection=None):
self.model = {}
self.test_users = test_users
if test_users is None:
test_users = range(self.n_users)
tu = {}
if b is not None:
for u in test_users:
tu[u] = set(test_users[u][:b])
else:
for u in test_users:
tu[u] = set(test_users[u])
for i, u in enumerate(tu):
if self.verbose and (i + 1) % 100 == 0:
print("%d/%d" % (i + 1, len(test_users)))
utils.fl()
Xp = [idx for idx in tu[u]]
npos = len(Xp)
kp = np.zeros((npos, npos))
for j in range(npos):
kp[j] = self.K[Xp[j], Xp].todense()
kp = co.matrix(kp)
kn = self.q_[Xp, :]
I = self.lambda_p * utc.identity(npos)
P = kp + I
q = -kn
G = -utc.identity(npos)
h = utc.zeroes_vec(npos)
A = utc.ones_vec(npos).T
b = co.matrix(1.0)
solver.options['show_progress'] = False
sol = solver.qp(P, q, G, h, A, b)
#print(selection[u][0:5])
#utils.fl()
tmp = (self.K[np.array(Xp)[:, None], selection[u]].T).dot(
sol['x']) - self.q_[[a for a in selection[u]]]
self.model[u] = np.zeros(len(self.items_list)) - np.inf
#print(tmp.shape)
self.model[u][selection[u]] = tmp[:, 0]
#for i, v in enumerate(selection[u]):
# self.model[u][v] = tmp[i]
#print self.model[u]
# endfor
return self
def build_selection(k, p2s_test, global_popularity, t2t_id, S, p2t, lengths,
test_seed, P, power):
selection = {}
for user in p2s_test[test_seed]:
if (p2t[str(user)] == "" or p2t[str(user)] not in t2t_id):
selection[user] = global_popularity[:k]
else:
row = S[t2t_id[p2t[str(user)]], :]**power / lengths
selection[user] = utils.scores_sorter(P.dot(row), k)
print("selection built")
utils.fl()
return selection
def multiline_xy_plot(x, ys, xlabel, ylabel, legend_labels, filename, y_errorbars=None, save_figure = True):
assert isinstance(ys, list)
cmap = brewer2mpl.get_map('Set1', 'qualitative', 9)
p = Ppl(cmap, alpha=1)
fig, ax = plt.subplots(1)
ax.set_xlabel(fl(xlabel))
ax.set_ylabel(fl(ylabel))
ax.yaxis.get_major_formatter().set_powerlimits((0, 1))
for i, y in enumerate(ys):
#print "Errorbar: %s"%y_errorbar
#if y_errorbars is not None: ax.errorbar(x, y, yerr=y_errorbar, fmt='o')
#print legend_labels[i]
p.plot(ax, x, y, linewidth=2, label = legend_labels[i])
p.legend(ax, loc=0, frameon = False)
if save_figure: fig.savefig(filename)
return ax, fig
def plot_image_matrix(filename, jaccard_matrix, x_ticklabels = None, y_ticklabels = None):
colormap = brewer2mpl.get_map('Set1', 'qualitative', 9)
p = Ppl(colormap, alpha=1)
fig, ax = plt.subplots(1)
masked_jaccard = np.ma.masked_where(np.isnan(jaccard_matrix), jaccard_matrix)
p.pcolormesh(fig, ax, masked_jaccard)
print x_ticklabels
print y_ticklabels
#ax.imshow(jaccard_matrix, interpolation = 'none')
#ax.set_xticks([])
yticks = range(jaccard_matrix.shape[0]+1)
xticks = range(jaccard_matrix.shape[1]+1)
ax.set_yticks(yticks)
ax.set_xticks(xticks)
#x_ticklabels.reverse()
ax.set_yticklabels(y_ticklabels)
ax.set_xticklabels(x_ticklabels)
ax.set_ylabel(fl('ratioagent'))
ax.set_xlabel(fl('ratiolatency'))
fig.savefig(filename)
return fig, ax
def make_scatter_plot_for_labelled_data(data_frame, x_name, y_name, labels, filename, colormap, x_function = 'dummy', y_function = 'dummy', legend = False, point_size = 5, omit_largest = 0, labels_to_plot = []):
### Originally created for issue_28
if not labels_to_plot: labels_to_plot = set(labels)
assert omit_largest < max(set(labels)), "omit_largest must be smaller than number of clusters"
colors = colormap.mpl_colors
def dummy(a): return a
p = Ppl(colormap, alpha=1)
fig, ax = plt.subplots(1)
#ax.set_autoscale_on(False)
ax.set_xlim([eval(x_function)(min(data_frame[x_name])), eval(x_function)(max(data_frame[x_name]))])
ax.set_ylim([eval(y_function)(min(data_frame[y_name])), eval(y_function)(max(data_frame[y_name]))])
#x_label = x_name.capitalize().replace('_', ' ')
if x_function == 'log': x_label += ' (log)'
#y_label = y_name.capitalize().replace('_', ' ')
if y_function == 'log': y_label += ' (log)'
ax.set_xlabel(fl(x_name))
ax.set_ylabel(fl(y_name))
ax.xaxis.get_major_formatter().set_powerlimits((0, 1))
ax.yaxis.get_major_formatter().set_powerlimits((0, 1))
# Show the whole color range
cluster_size = map(lambda l: len(labels[labels == l]), set(labels))
sizes, groups = zip(*sorted(zip(cluster_size, set(labels)), reverse=True))
#print sizes, groups
for order_to_plot, group in enumerate(list(groups)[-(len(groups)-omit_largest):]):
#print order_to_plot, sizes[order_to_plot], group, cluster_size[group]
if group in labels_to_plot:
#print 'Plotting points in group %s'%group
x = eval(x_function)(data_frame[labels == group][x_name])
y = eval(y_function)(data_frame[labels == group][y_name])
p.scatter(ax, x, y, label='C%s: %s'%(group, list(sizes)[order_to_plot]), s=point_size, linewidth=0, zorder=order_to_plot, color=colors[group])
if legend:
legend = p.legend(ax, loc=0, fancybox=True, markerscale=5, frameon=False)
legend.set_zorder(100)
#ax.set_title('prettyplotlib `scatter` example\nshowing default color cycle and scatter params')
fig.savefig(filename)
def mkplot(fitness): mask = f[fitness] < f[fitness].quantile(0.95) mask &= f.overshoot < 5 masked_fit = f[mask][fitness] masked_par = p[mask] counts, bins = np.histogram(masked_fit, bins = 50) ws = (bins[1] - bins[0]) means = concat(map(lambda bin: masked_par[(masked_fit > bin - ws) & (masked_fit < bin + ws)].mean(), bins), axis=1).transpose() means_as_list = map(lambda x: list(means[x]), means.columns) xlabel = fl(fitness) ylabel = '' legend_labels = map(fl, means.columns) filename = folder + '%s.png'%fitness multiline_xy_plot(bins, means_as_list, xlabel, ylabel, legend_labels, filename)
def calc_and_plot(ratio_direction): for fitness in f.columns: ssmm_ys = list() sc_ys = list() legend_labels = list() for ratio_lower, ratio_upper in zip_to_tuples(ratio_range): ratio_mask = (ratio_lower < p.ratio) & (p.ratio < ratio_upper) ssmm_lat_range = concat(map(lambda l: f[get_ssmmlat_mask(l,l+20) & ratio_mask].mean(), ssmmlatencyrange), axis=1).transpose() ssmm_ys.append(ssmm_lat_range[fitness]) sc_lat_range = concat(map(lambda l: f[get_sclat_mask(l,l+20) & ratio_mask].mean(), sclatencyrange), axis=1).transpose() sc_ys.append(sc_lat_range[fitness]) legend_labels.append(r'$\displaystyle %s < %s < %s$'%(round(ratio_lower,1), fl(ratio_direction, mathmode = False), round(ratio_upper,1))) filename = '%s_%s_%s_mmlatency.png'%(folder, ratio_direction, fitness) multiline_xy_plot(ssmm_lat_range.index, ssmm_ys, xlabel = 'ssmm_latency_mu', ylabel = fitness, legend_labels = legend_labels, filename = filename) filename = '%s_%s_%s_sclatency.png'%(folder, ratio_direction, fitness) multiline_xy_plot(sc_lat_range.index, sc_ys, xlabel = 'sc_latency_mu', ylabel = fitness, legend_labels = legend_labels, filename = filename)
sys.path.append(
os.path.dirname(os.path.dirname(os.path.dirname(
os.path.abspath(__file__)))))
import utils
'''
File used to build the Kernel between songs: songs are represented with playlists they belongs to. In each
element of the kernel matrix we will have the cosine similarity between representation of two songs
'''
#s2p = utils.jload("../data/validation/s2p.json")
s2p = utils.jload("./s2p.json")
for s in s2p:
s2p[s] = set(s2p[s])
print "Creating s2d...",
utils.fl()
s2d = {}
for ss in s2p:
s = int(ss)
if s not in s2d:
s2d[s] = {}
for p in s2p[ss]:
if p not in s2d[s]:
s2d[s][p] = 1.
else:
s2d[s][p] += 1.
print "done!"
utils.fl()
print "Creating norms...",
utils.fl()
def train(self, test_users=None, bucket=None):
t = time.time()
self.model = {}
Kar = self.Klist[1].todense()
Kal = self.Klist[2].todense()
self.test_users = test_users
if test_users is None:
self.test_users = range(self.n_users)
if bucket is not None:
tu = {u: set(self.test_users[u][:bucket]) for u in self.test_users}
else:
tu = {u: set(self.test_users[u]) for u in self.test_users}
for i, u in enumerate(tu):
if self.verbose and (i + 1) % 10 == 0:
print("%d/%d" % (i + 1, len(self.test_users)))
print(str(time.time() - t))
utils.fl()
Xp = [int(self.s2k[idx]) for idx in tu[u]]
idxar = [self.k2ar[p] for p in Xp]
idxal = [self.k2al[p] for p in Xp]
npos = len(Xp)
kp = np.zeros((npos, npos))
for j in range(npos):
kp[j] = self.mu[0] * self.Klist[0][Xp[j],
Xp].todense().reshape(
(npos, ))
kp[j] = kp[j] + self.mu[1] * Kar[idxar[j], idxar].reshape(
(npos, ))
kp[j] = kp[j] + self.mu[2] * Kal[idxal[j], idxal].reshape(
(npos, ))
kp = co.matrix(kp)
kn = self.q_[Xp, :]
I = self.lambda_p * utc.identity(npos)
P = kp + I
q = -kn
G = -utc.identity(npos)
h = utc.zeroes_vec(npos)
A = utc.ones_vec(npos).T
b = co.matrix(1.0)
solver.options['show_progress'] = False
sol = solver.qp(P, q, G, h, A, b)
K = co.matrix(self.mu[0] * self.Klist[0][Xp, :].todense())
idxar = np.array(idxar).reshape((len(idxar), 1))
idxal = np.array(idxal).reshape((len(idxal), 1))
K = K + co.matrix(self.mu[1] * Kar[idxar, self.k2ar])
K = K + co.matrix(self.mu[2] * Kal[idxal, self.k2al])
self.model[u] = (K.T * sol['x']) - self.q_
return self
def train(self, test_users=None, b=None):
t = time.time()
self.model = {}
self.test_users = test_users
if test_users is None:
test_users = range(self.n_users)
tu = {}
if b is not None:
for u in test_users:
tu[u] = set(test_users[u][:b])
else:
for u in test_users:
tu[u] = set(test_users[u])
for i, u in enumerate(tu):
if self.verbose and (i + 1) % 10 == 0:
print("%d/%d" % (i + 1, len(test_users)))
print(str(time.time() - t))
utils.fl()
Xp = [int(self.K_indexing[str(idx)]) for idx in tu[u]]
npos = len(Xp)
kp = np.zeros((npos, npos))
for j in range(npos):
kp[j] = self.K[Xp[j], Xp].todense()
'''
for row in range(kp.shape[0]):
x = len(self.u2i[int(self.inverse_indexing[Xp[row]])])
sx = math.sqrt(self.identity_disjunctive(self.N, x))
xN = (x / N)
for col in range(row+1, kp.shape[0]):
if kp[row,col] == 0.:
y = len(self.u2i[int(self.inverse_indexing[Xp[col]])])
sz = math.sqrt(self.identity_disjunctive(self.N, z))
kp[row,col] = (2. / (sx*sz)) * xN * (z / (N-1.))
kp[row,col] = kp[col,row]
'''
for row in range(kp.shape[0]):
x = len(self.u2i[self.inverse_indexing[Xp[row]]])
for col in range(row + 1, kp.shape[0]):
if kp[row, col] == 0.:
z = len(self.u2i[self.inverse_indexing[Xp[col]]])
mx, mn = (x, z) if x >= z else (z, x)
kp[row, col] = self.imp_dic["%d,%d" % (mx, mn)]
kp[col, row] = kp[row, col]
kp = co.matrix(kp)
kn = self.q_[Xp, :]
I = self.lambda_p * utc.identity(npos)
P = kp + I
q = -kn
G = -utc.identity(npos)
h = utc.zeroes_vec(npos)
A = utc.ones_vec(npos).T
b = co.matrix(1.0)
solver.options['show_progress'] = False
sol = solver.qp(P, q, G, h, A, b)
K_imp = co.matrix(self.K[Xp, :].todense())
for row in range(K_imp.size[0]):
x = len(self.u2i[self.inverse_indexing[Xp[row]]])
for col in range(K_imp.size[1]):
if K_imp[row, col] == 0.:
z = len(self.u2i[self.inverse_indexing[col]])
mx, mn = (x, z) if x >= z else (z, x)
K_imp[row, col] = self.imp_dic["%d,%d" % (mx, mn)]
self.model[u] = (K_imp.T * sol['x']) - self.q_
#import numpy.random as rnd
#with open("../files/"+rnd.randint()+".txt") as F:
# endfor
return self
def train(self, test_users=None, b=None, test_titles=None, c=0):
self.model = {}
self.test_users = test_users
self.c = c
if test_users is None:
test_users = range(self.n_users)
tu = {}
if b is not None:
for u in test_users:
tu[u] = set(test_users[u][:b])
else:
for u in test_users:
tu[u] = set(test_users[u])
for i, u in enumerate(tu):
t = time.time()
beta = self.Beta[test_titles[u]]
beta_K = self.K.dot(beta.T).todense()
if self.verbose and (i + 1) % 1 == 0:
print("%d/%d" % (i + 1, len(test_users)))
utils.fl()
Xp = np.array([idx for idx in tu[u]])
npos = len(Xp) + 1
kp = np.zeros((npos, npos))
#print(self.K[Xp[:,None],Xp].todense())
kp[:npos - 1, :npos - 1] = self.K[Xp[:, None], Xp].todense()
kp[npos - 1, :] = np.concatenate(
(self.K[Xp, :].dot(beta.T).todense(), np.ones(
(1, 1)))).flatten()
kp[:npos - 1, npos - 1] = kp[npos - 1, :npos - 1]
q0 = (co.matrix(beta.todense()) * self.q_)[0, 0]
kp = co.matrix(kp)
kn = co.matrix(
np.concatenate(
(np.array(self.q_[list(Xp), :]), np.array([[q0]]))))
I = self.lambda_p * utc.identity(npos)
P = kp + I
q = -kn
G = np.vstack((-np.eye(npos), np.zeros(npos)))
G[npos, -1] = 1
G = co.matrix(G)
h = np.zeros(npos + 1)
h[-1] = self.c
h = co.matrix(h)
A = utc.ones_vec(npos).T
b = co.matrix(1.0)
solver.options['show_progress'] = False
sol = solver.qp(P, q, G, h, A, b)
self.model[u] = np.array((self.K[Xp, :].T).dot(sol['x'][:-1]) +
beta_K * sol['x'][-1] -
self.q_).flatten()
#print self.model[u]
print(time.time() - t)
# endfor
return self