def main():
myBroker = ALBroker("NaoAppBroker",
"0.0.0.0", # listen to anyone
0, # find a free port and use it
NAO_IP, # parent broker IP
NAO_PORT) # parent broker port
global SpeechDetection, NaoWorkingMode
SpeechDetection = SpeechDetectionModule("SpeechDetection")
basicVideoProcessing = BasicVideoProcessing()
basicVideoProcessing.connectToCamera()
try:
while True:
img = basicVideoProcessing.getImageFromCamera()
if (img == None):
print "Image from camera is empty!"
break
else:
if NaoWorkingMode == None:
sample.sample()
elif NaoWorkingMode == "color":
print 'Working in color detection mode...'
#cv2.imshow('Color', img) - cos nie dziala, pewnie kwestia tego, ze to kod ladowany jako modul
elif NaoWorkingMode == "text":
print 'Working in text detection mode...'
elif NaoWorkingMode == "hand":
print 'Working in hand gesture detection mode...'
elif NaoWorkingMode == "phone":
print 'Working in mobile phone detection mode...'
except KeyboardInterrupt:
print "Interrupted by user, shutting down"
except RuntimeError, err:
print "An error occured: " + str(err)
def evaluateCrossSections(): for tmpline in csFile: line=tmpline.split() if line==[]: continue if line[0]=='STOP': break if not (line[0]=='charm' or line[0]=='beauty'): continue Flavour=line[0] Q2Min=line[1] x=sample(Flavour, Q2Min, "AE", 0, '') x._verbose=False x._NumberOfEventsCrossSect=10000 print Flavour+', Q^2> '+Q2Min+' GeV^2, resAE' x._calculateCrossSect() y=sample(Flavour, Q2Min, "C", 0, '') y._verbose=False y._NumberOfEventsCrossSect=10000 print Flavour+', Q^2> '+Q2Min+' GeV^2, resC' y._calculateCrossSect() z=sample(Flavour, Q2Min, "BGF", 0, '') z._verbose=True z._NumberOfEventsCrossSect=10000 print Flavour+', Q^2> '+Q2Min+' GeV^2, BGF' z._calculateCrossSect()
async def on_message(message):
global training
if client.user.mentioned_in(message):
sp.sample(sampleArgs, re.sub('(<@|<@!)([0-9])+>', '', message.content))
tf.reset_default_graph()
with open('output/output.txt', 'r') as the_file:
lines = the_file.read().split('\\r\\n')
# the training data im using produced a lot of double-escaped unicode, e.g. \\xf012 or something like that, so it has to decode twice, but python is funky so this is the ugly, horrible fix
# remove the re.sub() to allow barry to tag people
await client.send_message(discord.Object(id=client_channel), re.sub('(<@|<@!)([0-9])+>', '', lines[1].encode('ascii').decode('unicode_escape').encode('ascii').decode('unicode_escape')), tts=bool(random.getrandbits(1)))
elif message.content.startswith('!record') and message.author.id == admin_id:
print('Recording...')
with open('data/input.txt', 'w') as the_file:
async for log in client.logs_from(message.channel, limit=1000000000000000):
messageEncode = str(log.content.encode("utf-8"))[2:-1]
template = '{message}\n'
try:
the_file.write(template.format(message=messageEncode))
except:
the_file.write(template.format(message=messageEncode))
print('Data Collected from ' + message.channel.name)
elif message.content.startswith('!train') and message.author.id == admin_id:
if training != True:
# status change doesnt work right now. starting the training turns off the discord bot, probably just due to how the training function works
await client.change_presence(game=None, status='with his brain', afk=False)
training = True
tr.train(trainArgs)
elif training == True:
await client.change_presence(game=None, status=None, afk=False)
training = False
elif message.content.startswith('!leave') and message.author.id == admin_id:
await client.disconnect()
def main():
myBroker = ALBroker(
"NaoAppBroker",
"0.0.0.0", # listen to anyone
0, # find a free port and use it
NAO_IP, # parent broker IP
NAO_PORT) # parent broker port
global SpeechDetection, NaoWorkingMode
SpeechDetection = SpeechDetectionModule("SpeechDetection")
basicVideoProcessing = BasicVideoProcessing()
basicVideoProcessing.connectToCamera()
try:
while True:
img = basicVideoProcessing.getImageFromCamera()
if (img == None):
print "Image from camera is empty!"
break
else:
if NaoWorkingMode == None:
sample.sample()
elif NaoWorkingMode == "color":
print 'Working in color detection mode...'
#cv2.imshow('Color', img) - cos nie dziala, pewnie kwestia tego, ze to kod ladowany jako modul
elif NaoWorkingMode == "text":
print 'Working in text detection mode...'
elif NaoWorkingMode == "hand":
print 'Working in hand gesture detection mode...'
elif NaoWorkingMode == "phone":
print 'Working in mobile phone detection mode...'
except KeyboardInterrupt:
print "Interrupted by user, shutting down"
except RuntimeError, err:
print "An error occured: " + str(err)
def sample(checkpoint_path, sampling_type, n_samples):
import sample
models_path = os.path.dirname(checkpoint_path)
models_path = os.path.join(models_path, 'models.pkl')
with open(models_path, 'rb') as f:
models = pkl.load(f)
for model in models:
sample.sample(checkpoint_path, model, sampling_type, n_samples)
def play(self):
abcpath = os.getcwd() + '\\music\\' + str(self.secnum) + '.abc'
midipath = os.getcwd() + '\\music\\' + str(self.secnum) + '.mid'
if not os.path.exists(midipath):
sample(self.secnum)
music21.converter.parse(abcpath).write('midi', midipath)
pygame.mixer.music.load(midipath)
pygame.mixer.music.play(-1)
def sentence_gen():
histo_text = get_words('siddhartha.txt')
histo = histogram(histo_text)
random_word = sample(histo)
random_words = []
for i in range(7):
random_words.append(sample(histo))
random_sentence = sentence_maker(random_words)
return random_sentence
def chain_traversal(self, length=10):
''' Creates a sentence using the Markov Chain'''
current_word = random.choice(list(self))
sentence = []
sentence.append(current_word)
for _ in range(length):
new_word = sample(self[current_word])
sentence.append(sample(self[current_word]))
current_word = new_word
return ' '.join(sentence)
def fig1(model, output_folder):
'''
This function makes two 2x10 images
showing the difference between conditioning
and intervening
'''
str_step = guess_model_step(model)
fname = os.path.join(output_folder, str_step + model.model_type)
for key in [
'Young', 'Smiling', 'Wearing_Lipstick', 'Male',
'Mouth_Slightly_Open', 'Narrow_Eyes'
]:
#for key in ['Mustache','Bald']:
#for key in ['Mustache']:
print 'Starting ', key,
#for key in ['Bald']:
p50, n50 = find_logit_percentile(model, key, 50)
do_dict = {key: np.repeat([p50], 10)}
eps = 3
cond_dict = {key: np.repeat([+eps], 10)}
out, _ = sample(model, do_dict=do_dict)
intv_images = out['G']
out, _ = sample(model, cond_dict=cond_dict)
cond_images = out['G']
images = np.vstack([intv_images, cond_images])
dc_file = fname + '_' + key + '_topdo1_botcond1.pdf'
save_figure_images(model.model_type, images, dc_file, size=[2, 10])
do_dict = {key: np.repeat([p50, n50], 10)}
cond_dict = {key: np.repeat([+eps, -eps], 10)}
dout, _ = sample(model, do_dict=do_dict)
cout, _ = sample(model, cond_dict=cond_dict)
itv_file = fname + '_' + key + '_topdo1_botdo0.pdf'
cond_file = fname + '_' + key + '_topcond1_botcond0.pdf'
eps = 3
save_figure_images(model.model_type, dout['G'], itv_file, size=[2, 10])
save_figure_images(model.model_type,
cout['G'],
cond_file,
size=[2, 10])
print '..finished ', key
#return images,cout['G'],dout['G']
return key
def mplot3d(f, var1, var2, show=True):
"""
Plot a 3d function using matplotlib/Tk.
"""
import warnings
warnings.filterwarnings("ignore", "Could not match \S")
try:
import pylab as p
import matplotlib.axes3d as p3
except ImportError:
raise ImportError("Matplotlib is required to use mplot3d.")
x, y, z = sample(f, var1, var2)
fig = p.figure()
ax = p3.Axes3D(fig)
#ax.plot_surface(x,y,z) #seems to be a bug in matplotlib
ax.plot_wireframe(x, y, z)
ax.set_xlabel('X')
ax.set_ylabel('Y')
ax.set_zlabel('Z')
if show:
p.show()
def create_name(start_token, dictionary):
""" Takes dictionary, start and end tokens and makes a sentence. """
# create sentence and add first word
name = []
# this is hard coded; must be changed to fit the order number; currently second
(letter1, letter2) = start_token
name.append(letter1)
name.append(letter2)
current_token = start_token
# stop when current_token is a stop token
while not current_token[1].isspace() and len(name) <= 10:
for key, value in dictionary.items():
if key == current_token:
# sample from histogram of values
cumulative = sample.cumulative_distribution(value)
sample_letter = sample.sample(cumulative)
# add new sample to name_list
name.append(sample_letter)
# assign second word of key and value to current token
# this is hard coded; must be changed to fit the order number
# unpacking the current token
(current_token_one, current_token_two) = current_token
current_token = (current_token_two, sample_letter)
# get out of for loop and start process over
break
return name
def hello_world():
his = histogram_dict(original_text)
p = s.get_probability(his)
s_ = s.sample(p, his)
# print("W: %s" % s)
return "W: %s" % s_
def encoder(message_str, context, enc, model):
unicode_enc = True
mode = 'arithmetic'
block_size = 3 # for huffman and bins
temp = 0.9 # for arithmetic
precision = 26 # for arithmetic
sample_tokens = 100 # for sample
topk = 300
finish_sent=True
context_tokens = encode_context(context, enc)
if mode not in ['arithmetic', 'huffman', 'bins', 'sample']:
raise NotImplementedError
if unicode_enc:
ba = bitarray.bitarray()
ba.frombytes(message_str.encode('utf-8'))
message = ba.tolist()
else:
message_ctx = [enc.encoder['<|endoftext|>']]
message_str += '<eos>'
message = decode_arithmetic(model, enc, message_str, message_ctx, precision=40, topk=60000)
# Next encode bits into cover text, using arbitrary context
Hq = 0
if mode == 'arithmetic':
out, nll, kl, words_per_bit, Hq = encode_arithmetic(model, enc, message, context_tokens, temp=temp, finish_sent=finish_sent, precision=precision, topk=topk)
elif mode == 'huffman':
out, nll, kl, words_per_bit = encode_huffman(model, enc, message, context_tokens, block_size, finish_sent=finish_sent)
elif mode == 'bins':
out, nll, kl, words_per_bit = encode_block(model, enc, message, context_tokens, block_size, bin2words, words2bin, finish_sent=finish_sent)
elif mode == 'sample':
out, nll, kl, Hq = sample(model, enc, sample_tokens, context_tokens, temperature=temp, topk=topk)
words_per_bit = 1
text = enc.decode(out)
return text
def mplot2d(f, var, show=True):
"""
Plot a 2d function using matplotlib/Tk.
"""
import warnings
warnings.filterwarnings("ignore", "Could not match \S")
p = import_module("pylab")
if not p:
sys.exit("Matplotlib is required to use mplot2d.")
if not is_sequence(f):
f = [
f,
]
for f_i in f:
x, y = sample(f_i, var)
p.plot(x, y)
p.draw()
if show:
p.show()
def mplot3d(f, var1, var2, show=True):
"""
Plot a 3d function using matplotlib/Tk.
"""
import warnings
warnings.filterwarnings("ignore", "Could not match \S")
p = import_module('pylab')
# Try newer version first
p3 = import_module('mpl_toolkits.mplot3d',
__import__kwargs={
'fromlist': ['something']
}) or import_module('matplotlib.axes3d')
if not p or not p3:
sys.exit("Matplotlib is required to use mplot3d.")
x, y, z = sample(f, var1, var2)
fig = p.figure()
ax = p3.Axes3D(fig)
# ax.plot_surface(x, y, z, rstride=2, cstride=2)
ax.plot_wireframe(x, y, z)
ax.set_xlabel('X')
ax.set_ylabel('Y')
ax.set_zlabel('Z')
if show:
p.show()
def minning(request):
sample=sample.sample('F:/2016Graduate/20160922.xlsx',u'Python')
sup=118
fsetdict={}
# fItems=mining.findFItem(sample.recipeTrans, sup)
# f2set=mining.findFSet(sample.recipeTrans, fItems.keys(), 2, sup)
# fsetdict.update(f2set)
# f3set=mining.findFSet(sample.recipeTrans, mining.getItemInSet(f2set.keys()), 3, sup)
# fsetdict.update(f3set)
# f4set=mining.findFSet(sample.recipeTrans, mining.getItemInSet(f3set.keys()), 4, sup)
# fsetdict.update(f4set)
# r=creatrules.rulesbuilder(sample=sample,fset=fsetdict,itemcount=fItems,conf=0.7)
# r.export2excel('F:/2016Graduate/python_result02.xls')
# dr=drawrule.drawrule(r.rulesdict)
# dr.drawscatter(fn='scatter02.png')
# dr.drawbubble(fn='bubble02.png')
# dr.drawnetwork(fn='network02.png')
fItems=mining_adv.findFItem(sample.recipeTrans, sample.weights, sup)
f2set=mining_adv.findFSet(sample.recipeTrans, fItems.keys(), sample.weights, 2, sup)
fsetdict.update(f2set)
f3set=mining_adv.findFSet(sample.recipeTrans, mining.getItemInSet(f2set.keys()), sample.weights, 3, sup)
fsetdict.update(f3set)
f4set=mining_adv.findFSet(sample.recipeTrans, mining.getItemInSet(f3set.keys()), sample.weights, 4, sup)
fsetdict.update(f4set)
r=creatrules.rulesbuilder(sample=sample,fset=fsetdict,itemcount=fItems,conf=0.7)
r.export2excel('F:/2016Graduate/python_result_y0.xls')
def create_sentence(self, start_token, stop_tokens, dictionary):
""" takes dictionary, start and end tokens and makes a sentence """
# create sentence and add first word
sentence = []
# this is hard coded; must be changed to fit the order number; currently third
(word1, word2, word3) = start_token
sentence.append(word1)
sentence.append(word2)
sentence.append(word3)
# print("There should be three words", sentence)
current_token = start_token
# print("This is my dictionary", dictionary)
# stop when current_token is a stop token
while current_token not in stop_tokens or len(sentence) <= 8:
for key, value in dictionary.items():
if key == current_token:
#
# sample from histogram of values
cumulative = sample.cumulative_distribution(value)
sample_word = sample.sample(cumulative)
# add new sample to sentence_list
sentence.append(sample_word)
# assign second word of key and value to current token
# this is hard coded; must be changed to fit the order number
# I am unpacking the current token
(current_token_one, current_token_two,
current_token_three) = current_token
current_token = (current_token_two, current_token_three,
sample_word)
# get out of for loop and start process over
break
return sentence
def testSample():
model = WrappedModel(TestDistributionSystem().getModel())
print(model.getDistribution(DistrCall('decideBias', [])))
for i in range(20):
res = sample(model, DistrCall('main', []))
print(res)
print(list(map(model.refToJSON, res)))
def newsample(self):
try:
dlg = sampleDialog()
if (dlg.exec_() == QDialog.Accepted):
name, pt = dlg.getValues()
if (pt is None or np.equal(pt[:-1], None).any()):
raise TypeError
self.fig.clear()
if self.sample:
self.sample.disconnect_plot()
self.sample = sample(name=name,
bl=pt[0],
br=pt[1],
tl=pt[2],
tr=pt[3],
c=pt[4],
temp=self.temp)
self.sample.show(self.fig)
except TypeError:
logger.exception(
"Can't create sample : At least one point is invalid")
QMessageBox.critical(
self, "Error",
"Can't create sample : At least one point is invalid")
except Exception as e:
logger.exception("Can't create sample")
QMessageBox.critical(self, "Error", "Can't create sample")
def update_valid(n, p):
n = unwrap(n)
# this grid square must have been created already in locally_early()
q, t, r = grid[n]
# no change if this square is already output or covered
if isinf(t):
return False
p = closest_p(p, n)
r = r - disk(p)
r.simplify()
# update
grid[n] = (q, t, r)
# Did p invalidate q?
if too_close(q, p):
A = r.area()
if A == 0:
# q's square has been covered
grid[n] = (q, inf, False)
else:
# generate a new q
q = sample.sample(r)
t = t + random.expovariate(A)
grid[n] = (q, t, r)
return True
return False
def mplot3d(f, var1, var2, show=True):
"""
Plot a 3d function using matplotlib/Tk.
"""
import warnings
warnings.filterwarnings("ignore", "Could not match \S")
p = import_module('pylab')
# Try newer version first
p3 = import_module('mpl_toolkits.mplot3d',
__import__kwargs={'fromlist':['something']}) or import_module('matplotlib.axes3d')
if not p or not p3:
sys.exit("Matplotlib is required to use mplot3d.")
x, y, z = sample(f, var1, var2)
fig = p.figure()
ax = p3.Axes3D(fig)
#ax.plot_surface(x,y,z) #seems to be a bug in matplotlib
ax.plot_wireframe(x,y,z)
ax.set_xlabel('X')
ax.set_ylabel('Y')
ax.set_zlabel('Z')
if show:
p.show()
def mplot3d(f, var1, var2, show=True):
"""
Plot a 3d function using matplotlib/Tk.
"""
import warnings
warnings.filterwarnings("ignore", "Could not match \S")
try:
import pylab as p
import matplotlib.axes3d as p3
except ImportError:
raise ImportError("Matplotlib is required to use mplot3d.")
x, y, z = sample(f, var1, var2)
fig = p.figure()
ax = p3.Axes3D(fig)
#ax.plot_surface(x,y,z) #seems to be a bug in matplotlib
ax.plot_wireframe(x,y,z)
ax.set_xlabel('X')
ax.set_ylabel('Y')
ax.set_zlabel('Z')
if show:
p.show()
def get_sample_freq(self,
selected_freq,
SPS=40000,
dispFFT=False,
FFTchannels=[1, 2, 3],
axis=None,
raw_file=""):
samples_count = 4096
bytes_in_block = samples_count * 16 #4 channels, 4B per sample
fftfreq = np.fft.rfftfreq(
samples_count,
d=1.0 / SPS) # /16 -> /4 channels /4 bytes per channel
selected_index = np.argmin(np.abs(fftfreq - selected_freq))
#self._tail = struct.unpack_from("l", self._data, self.PRU0_OFFSET_DRAM_HEAD)[0]
samples = self.get_sample_block(bytes_in_block)
#Invert dimensions
channels = np.transpose(samples)
ref_wave = waveform(
selected_freq,
np.fft.rfft(channels[0])[selected_index] / samples_count)
selected_sample = sample(ref_wave)
for chan in FFTchannels:
fft = np.fft.rfft(channels[chan]) / samples_count
chan_wave = waveform(selected_freq, fft[selected_index])
selected_sample.add_channel(chan_wave)
return selected_sample
def __init__(self, maxNumber=1):
self.n = 0
self.mu = 0
self.m2 = 0
self.lo = 10**32
self.hi = -10**32
self.sd = 0
self.same = sample(maxNumber)
def nystrom(net, all_idx, k, d):
k_set = sample(net, k, 'deg^2_prob')
mat = net.calc_matrix_sparse(k_set, k_set).toarray()
u, dd, v = np.linalg.svd(mat)
reconstruct_mat = u[:, :d] @ np.diag(mydiv(dd[:d])) @ v[:d, :]
left = net.calc_matrix_sparse(all_idx, k_set)
right = net.calc_matrix_sparse(k_set, all_idx)
return left @ reconstruct_mat @ right
def chain_traversal(self, length=20):
''' Creates a sentence using the Markov Chain'''
current_word = random.choice(list(self.keys()))
sentence = []
for _ in range(length):
new_word = sample(self[current_word])
sentence.append(current_word[1])
current_word = (current_word[1], new_word)
return ' '.join(sentence)
def __init__(self):
self.cli = sample.sample()
self.df = pd.read_csv('soybeans-texas-collin.csv')
self.df.columns = [
c.lower().replace(' ', '_') for c in self.df.columns
]
self.df['value'] = self.df['value'].str.replace(",", "")
self.df['value'] = self.df['value'].apply(pd.to_numeric,
errors='coerce')
def create_sample(dataset, model_name, prime, size=5000):
output_name = "%s_%s_%d.txt" % (model_name, prime, size)
output_file = os.path.join(dataset_dir, dataset, output_dir, output_name)
model_file = os.path.join(dataset_dir, dataset, models_dir, model_name)
sample_args = SampleArguments(model_file, size, prime)
with open(output_file, "w") as out:
tf.reset_default_graph()
out.write(sample.sample(sample_args))
return output_file
def generate():
parser = argparse.ArgumentParser(description='Sample some text from the')
parser.add_argument('epoch', type=int, help='epoch checkpoint to sample')
parser.add_argument('--seed', default='', help='initial seed for the text')
parser.add_argument('--len', type=int, default=512, help='no of character')
args = parser.parse_args()
print(args.epoch)
music = sample(args.epoch, args.seed, args.len)
# return sample(args.epoch, args.seed, args.len)
return render_template('generate.html', data=music)
def __init__(self, maxNumber, listNumbers):
self.n = 0
self.mu = 0
self.m2 = 0
self.lo = 10 ^ 32
self.hi = -10 ^ 32
self.sd = 0
self.same = sample(maxNumber)
for x in listNumbers:
self.numInc(x)
def train(self, round):
# Simulation Input Sampling
self.thetas = sample.sample(self.args,
self.args.simulation_budget_per_round,
self.netDiscriminator,
self.netPosterior,
self.observation,
self.prior,
self.sim,
round == 0,
self.args.posteriorInferenceMethod == 'no',
self.args.numChains,
SNLE=False).detach().to(self.args.device)
# Simulation Execution
simulated_output = self.sim.parallel_simulator(self.thetas)
print("simulated output : ", simulated_output.shape, self.thetas.shape)
# Likelihood Learning
self.training_theta, self.training_x, self.validation_theta, self.validation_x, self.netDiscriminator = \
RatioLearning.LikelihoodToEvidenceRatioLearning(args, round, self.thetas, simulated_output, self.training_theta,
self.training_x, self.validation_theta, self.validation_x,
self.netDiscriminator, self.optDiscriminator)
# Get Training Teacher Data for Implicit Surrogate Proposal (ISP) Learning
if self.args.posteriorInferenceMethod != 'no':
self.teacher_theta = sample.sample(self.args,
self.args.num_training,
self.netDiscriminator,
self.netPosterior,
self.observation,
self.prior,
self.sim,
round == -1,
True,
self.args.num_training,
parallel=True,
SNLE=False)
# Implicit Surrogate Proposal (ISP) Distribution Learning
self.netPosterior = PosteriorLearning.PosteriorLearning(
args, self.sim, self.teacher_theta)
def generate_sentence(markov_dict):
length = 10
first_word = start_word(markov_dict)
sentence = first_word.capitalize()
for word in range(random.randint(1, length)):
second_word = sample(markov_dict[first_word])
first_word = second_word
sentence += ' ' + second_word
return sentence
def main(config):
print('creating data')
dataLoader = Dataclass(dotdict(config.data_details))
training_data, valid_data = dataLoader.get_training_data()
# creat hmm model
# model = hmm.GaussianHMM(n_components=12, covariance_type="full")
# model.fit(training_data)
# feature_matrix , state_sequence = model.sample(100)
model = None
# The embedding dimension
if (not config.sampling_mode):
print('starting training')
model = train(training_data, valid_data, config)
if model == None:
model = load_model(config, training_data)
print('sampling, and generating midi')
sample(model, config, training_data)
def create_npz_files(n, name):
total_frames = np.zeros((TOTAL_FRAMES, n, FRAMES, SIDE, SIDE))
total_labels, total_sizes = np.zeros((TOTAL_FRAMES, n)), np.zeros((TOTAL_FRAMES, n))
for i in range(TOTAL_FRAMES):
photons = i + 1
index = photons_to_index(photons)
total_frames[i], total_labels[i] = sample(photons, n, index)
total_sizes[i] = [photons] * n
total_frames = total_frames.reshape((TOTAL_FRAMES * n, FRAMES, SIDE, SIDE), order = 'F')
total_labels = total_labels.reshape((TOTAL_FRAMES * n), order = 'F')
total_sizes = total_sizes.reshape((TOTAL_FRAMES * n), order = 'F')
np.savez(name, frames = total_frames, labels = total_labels, sizes = total_sizes)
def __init__(self, max=512, nums=[], func=lambda x: x):
self.max = max
self.n = 0
self.mu = 0
self.m2 = 0
self.sd = 0
self.lo = inf
self.hi = -inf
self.w = 1
self._some = sample(self.max)
for x in nums:
self.numInc(func(x))
def prompt():
prompt = request.args.get('text')
challenger = request.args.get('challenger')
n = int(request.args.get('n'))
args = Namespace(prime=prompt, n=n, save_dir="save/save", sample=1)
output = sample(args)
new_example = output[0]
for character in output[1:]:
# Append an underscore if the character is uppercase.
if character.isupper():
new_example += '\n'
new_example += character
return json.dumps({"text": new_example})
def gen_text():
prompt = request.args.get('text')
challenger = request.args.get('challenger')
n = int(request.args.get('n'))
args = Namespace(prime=prompt, n=n, save_dir="save/save", sample=1)
output = sample(args)
# output = clean_msg(output)
output = output.replace("\n", "<br>")
print output
# output = replace_w_rhymes(output)
print output
#output = uniform_syl(output)
return output
def fetch_grid((i, j)):
"""on demand generation of empty grid squares."""
(i, j) = unwrap((i, j))
if (i, j) not in grid:
xloc = i * grid_spacing
yloc = j * grid_spacing
r = Polygon(
[
(xloc, yloc),
(xloc + grid_spacing, yloc),
(xloc + grid_spacing, yloc + grid_spacing),
(xloc, yloc + grid_spacing),
]
)
p = sample.sample(r)
t = random.expovariate(r.area())
grid[i, j] = (p, t, r)
return grid[(i, j)]
def processConfigFile(confFileSuffix):
configFile=open('config'+confFileSuffix+'.cfg','r')
for tmpline in configFile:
line=tmpline.split()
if line==[]: continue
if line[0]=='STOP': break
if not (line[0]=='charm' or line[0]=='beauty'): #to enable comments in config file
continue
Flavour=line[0]
SubProcess=line[1]
Q2Min=line[2]
Luminosity=line[3]
Trigger=line[4]
print Flavour+', subprocess: '+SubProcess+', Q^2> '+Q2Min+' GeV^2'
x=sample(Flavour, Q2Min, SubProcess, float(Luminosity), Trigger)
x._NumberOfEventsCrossSect=1
x._LetterToAppend='submit.letter'+confFileSuffix
x._OutputPathPrefix=getPathPrefix(confFileSuffix)
x._generate()
def gibbs_iteration(self, init=False):
"""
Uses Gibbs sampling to draw a single sample from the posterior
distribution over token--component (i.e., token--topic)
assignments given this instance's corpus (i.e., document
tokens). By default (i.e., if keyword argument 'init' is set
to the value 'False') all token--component assignments (and
corresponding counts) are assumed to have been initialized
previously; otherwise, they are initialized.
Keyword arguments:
init -- whether to initialize token--component assignments
"""
corpus = self.corpus
Nvt_plus_beta_n = self.Nvt_plus_beta_n
Nt_plus_beta = self.Nt_plus_beta
Ntd_plus_alpha_m = self.Ntd_plus_alpha_m
Nd_plus_alpha = self.Nd_plus_alpha
z = self.z
for d, (doc, zd) in enumerate(iterview(zip(corpus, z), inc=200)):
for n, (v, t) in enumerate(zip(doc.w, zd)):
if not init:
Nvt_plus_beta_n[v, t] -= 1
Nt_plus_beta[t] -= 1
Ntd_plus_alpha_m[d, t] -= 1
t = sample((Nvt_plus_beta_n[v, :] / Nt_plus_beta) * Ntd_plus_alpha_m[d, :])
Nvt_plus_beta_n[v, t] += 1
Nt_plus_beta[t] += 1
Ntd_plus_alpha_m[d, t] +=1
if init:
Nd_plus_alpha[d] += 1
zd[n] = t
def mplot2d(f, var, show=True):
"""
Plot a 2d function using matplotlib/Tk.
"""
import warnings
warnings.filterwarnings("ignore", "Could not match \S")
p = import_module('pylab')
if not p:
sys.exit("Matplotlib is required to use mplot2d.")
if not ordered_iter(f):
f = [f,]
for f_i in f:
x, y = sample(f_i, var)
p.plot(x, y)
p.draw()
if show:
p.show()
def mplot2d(f, var, show=True):
"""
Plot a 2d function using matplotlib/Tk.
"""
import warnings
warnings.filterwarnings("ignore", "Could not match \S")
try:
import pylab as p
except ImportError:
raise ImportError("Matplotlib is required to use mplot2d.")
if not ordered_iter(f):
f = [f,]
for f_i in f:
x, y = sample(f_i, var)
p.plot(x, y)
p.draw()
if show:
p.show()
def processConfigFile(confFileSuffix):
configFile=open('config'+confFileSuffix+'.cfg','r')
for tmpline in configFile:
line=tmpline.split()
if line==[]: continue
if line[0]=='STOP': break
if not (line[0]=='charm' or line[0]=='beauty'): # to enable any comments in config file
continue
Flavour=line[0]
SubProcess=line[1]
Q2Min=line[2]
Luminosity=line[3]
Trigger=line[4]
print Flavour+', subprocess: '+SubProcess+', Q^2> '+Q2Min+' GeV^2'
x = sample(Flavour, Q2Min, SubProcess, float(Luminosity), Trigger)
x._NumberOfEventsCrossSect=NumberOfEventsCrossSect
x._LetterToAppend='submit.letter'+confFileSuffix # For each entry in the config file there will be a separate
# submission letter file. In addition, content of this submission letter
# will be appended to file specified by this variable (will be created
# if does not exist). There will be one such file per config file.
x._OutputPathPrefix=getOutputPathPrefix(confFileSuffix)
x._generate()
x._InfoForWeb(webInfoFile)
from sample import sample import numpy as np #x = np.empty(10) x = sample(10,2) print(x)
def test_cab(self):
self.assertEqual(1, sample(3, 1, 2))
def classFactory(iface):
# load sample class from file sample
from sample import sample
return sample(iface)
def log_predictive_prob(self, new_corpus, num_samples):
"""
Returns an approximation of the log probability of the
specified new corpus given this instance's corpus (i.e.,
document tokens) AND current set of token--component (i.e.,
token--topic) assignments according to LDA.
Arguments:
new_corpus -- new corpus of documents
num_samples -- ...
"""
V, T = self.V, self.T
D_new = len(new_corpus)
alpha, alpha_m = self.alpha, self.alpha_m
Nvt_plus_beta_n = self.Nvt_plus_beta_n
Nt_plus_beta = self.Nt_plus_beta
Nvt_new, Nt_new, Ntd_new, z_new = [], [], [], []
for r in xrange(num_samples):
Nvt_new.append(zeros((V, T), dtype=int))
Nt_new.append(zeros(T, dtype=int))
Ntd_new.append(zeros((D_new, T), dtype=int))
z_r = []
for doc in new_corpus:
z_r.append(zeros(len(doc), dtype=int))
z_new.append(z_r)
log_p = 0
for d, doc in enumerate(iterview(new_corpus)):
for n, v in enumerate(doc.w):
tmp = zeros(num_samples, dtype=float)
for r in xrange(num_samples):
# for efficiency, resample only those
# token--component assignments belonging to
# previous tokens in the current document
for prev_n in xrange(0, n):
prev_v = doc.w[prev_n]
t = z_new[r][d][prev_n]
Nvt_new[r][prev_v, t] -= 1
Nt_new[r][t] -= 1
Ntd_new[r][d, t] -= 1
t = sample((Nvt_new[r][prev_v, :] + Nvt_plus_beta_n[prev_v, :]) / (Nt_new[r] + Nt_plus_beta) * (Ntd_new[r][d, :] + alpha_m))
Nvt_new[r][prev_v, t] += 1
Nt_new[r][t] += 1
Ntd_new[r][d, t] += 1
z_new[r][d][prev_n] = t
pass # YOUR CODE GOES HERE
Nvt_new[r][v, t] += 1
Nt_new[r][t] += 1
Ntd_new[r][d, t] += 1
z_new[r][d][n] = t
log_p += log_sum_exp(tmp) - log(num_samples)
return log_p
def sample(self):
if len(self.noteQueue) == 0:
self.noteQueue.extend(lstm.sample(40, 1))
next_note = self.noteQueue[0]
self.noteQueue = self.noteQueue[1:] # shift the first element
return next_note
def test_abc(self):
self.assertEqual(1, sample(1, 2, 3))
def test_acb(self):
self.assertEqual(1, sample(1, 3, 2))
def test_bac(self):
self.assertEqual(1, sample(2, 1, 3))
def test_bca(self):
self.assertEqual(1, sample(2, 3, 1))
def __init__(self):
self.lastNotePlayed = None
self.noteQueue = lstm.sample(40, 1)
self.bpm = 120
def test_cba(self):
self.assertEqual(1, sample(3, 2, 1))