def addDownload2(self):
start=timeit.default_timer()
global urlname
global urlsize
global urlstatus
global urltype
global model
global idx
url=str(self.dialog.lineEdit_6.text())
urlname = url.split('/')[-1]
response=urllib2.urlopen(url)
self.meta=response.info()
urlsize=int(self.meta.getheaders('Content-Length')[0])
if 1000<urlsize<1000000:
urlsize=str(urlsize/1000.0) + " KB"
elif 1000000<urlsize<10**9:
urlsize=str(1.0*urlsize/10**6) + " MB"
else:
urlsize=str(1.0*urlsize/10**9) + " GB"
self.dialog.label_17.setText(urlsize)
self.dialog.label_68.setText(urlname)
f=open(urlname,'wb')
f.write(response.read())
stop=timeit.default_timer()
speed=float(urlsize.split(' ')[0])/(stop-start)
speed=str(speed) + " " + urlsize.split(' ')[-1] + "/s"
self.dialog.label_67.setText("100 %")
self.dialog.label_69.setText(speed)
self.dialog.label_70.setText(str(stop-start))
def test(self, view):
"""
Calls the given view and measures the time for it to return. The
garbage collector is diabled during execution.
"""
gc_old = gc.isenabled()
gc.disable()
try:
start = timeit.default_timer()
if view.method == 'GET':
response = self.client.get(view.url, view.data)
elif view.method == 'POST':
response = self.client.post(view.url, view.data)
else:
raise ValueError('Unknown view method: %s' % view.method)
end = timeit.default_timer()
# Return result in milliseconds
time_ms = (end - start) * 1000
# Try to get version information
version = subprocess.check_output(['git', 'describe'])
from .models import TestResult
return TestResult(view=view, time=time_ms, result=response,
result_code=response.status_code, version=version)
finally:
if gc_old:
gc.enable()
def measured(self, msg, *args, **kwargs):
self.info(msg + '..', *args, **kwargs)
start_time = timeit.default_timer()
yield
end_time = timeit.default_timer()
self.info("%s took %.2fs." % (msg, end_time - start_time),
*args, **kwargs)
def __download_range(self, k, dst):
try:
_, ext = os.path.splitext(dst)
ds = []
parts = []
logging.info("Download %s start", k.name)
for startByte in range(0, k.size, self.splitMB):
output_part = self.new_temp_file(suffix=ext)
parts.append(output_part)
endByte = min(startByte + self.splitMB - 1, k.size)
logging.debug(
"deferToThreadPool %s start=%d end=%d size=%d cnt=%d",
k.name,
startByte,
endByte,
endByte - startByte,
len(ds),
)
d = twisted.internet.threads.deferToThreadPool(
reactor,
reactor.getThreadPool(), # @UndefinedVariable
self.__downloadOne,
k,
startByte,
endByte,
output_part,
len(ds),
)
ds.append(d)
if os.path.exists(dst):
os.remove(dst)
fout = file(dst, "wb")
start = timeit.default_timer()
for cnt, p in enumerate(parts):
yield ds[cnt]
shutil.copyfileobj(file(p, "rb"), fout)
size = min(k.size, (cnt + 1) * self.splitMB)
elapsed = timeit.default_timer() - start
speedstr = formatFileSize(size / elapsed)
sizestr = formatFileSize(size)
percent = (float(cnt) / len(parts)) * 100.0
logging.info(
"%03d/%03d (%.2f%%) speed=%s/s, elapsed=%.2f, size=%s",
cnt,
len(parts),
percent,
speedstr,
elapsed,
sizestr,
)
except Exception:
logging.error("download error", exc_info=True)
raise
def main():
start_time = timeit.default_timer()
proxies = []
targets = ['http://www.google-proxy.net/','http://free-proxy-list.net/']
for i in range(len(targets)):
proxy = proxy_scraper(targets[i])
for u in range(len(proxy)):
proxy_found = str(proxy[u]['ip'])+":"+str(proxy[u]['port'])
if proxy_found not in proxies:
if is_proxy_existed(proxy[u]['ip']) != True:
print proxy[u]['ip'] +" - "+ proxy[u]['port'] +" - "+ proxy[u]['hostname']
create_proxy(proxy[u]['ip'], proxy[u]['port'], proxy[u]['hostname'], proxy[u]['service'], proxy[u]['latitude'], proxy[u]['longitude'], proxy[u]['city'], proxy[u]['country'])
proxies.append(proxy_found)
# save to a file
file_name = "data_proxies.cfg"
write_file( file_name, "\n".join(proxies) )
print("\n%s proxies found. File saved. You can find it under '%s'." % (len(proxies), file_name))
# measure time
print "\nElapsed time: %d sec" % (timeit.default_timer() - start_time)
def create_features(features):
import timeit
source = load_source()
start = timeit.default_timer()
compute_features(source, features)
end = timeit.default_timer()
print("save all features takes ", (end-start))
def __execEvent__(self, eventName, ntime, commandHandler):
last = self.__events__[eventName]["lastExecTime"]
timeInterval = self.__events__[eventName]["timeInterval"]
if ntime - last >= timeInterval:
start = default_timer()
self.__events__[eventName]["function"](commandHandler, self.__events__[eventName]["channels"])
timeTaken = default_timer() - start
stats = self.__events__[eventName]["stats"]
if stats["average"] == None:
stats["average"] = timeTaken
stats["min"] = timeTaken
stats["max"] = timeTaken
else:
stats["average"] = (stats["average"]+timeTaken) / 2.0
if timeTaken < stats["min"]:
stats["min"] = timeTaken
if timeTaken > stats["max"]:
stats["max"] = timeTaken
self.__events__[eventName]["lastExecTime"] = time.time()
def main():
print "[Facebook Album Downloader v1]"
start = timeit.default_timer()
# hide images
prefs = {"profile.managed_default_content_settings.images": 2}
extensions = webdriver.ChromeOptions()
extensions.add_experimental_option("prefs", prefs)
browser = webdriver.Chrome(executable_path="chromedriver", chrome_options=extensions)
findAlbum(browser)
createAlbumPath()
queue = Queue()
for x in range(max_workers):
worker = DownloadWorker(queue)
worker.daemon = True
worker.start()
print "[Getting Image Links]"
linkImages = getImageLinks(browser)
print "[Found: " + str(len(linkImages)) + "]"
for fullRes in linkImages:
queue.put(fullRes)
print "[Downloading...]"
queue.join()
browser.quit()
stop = timeit.default_timer()
print "[Time taken: %ss]" % str(stop - start)
raw_input("Press any key to continue...")
def read_features(features):
"""
read all the features in the 'features' array and return a numpy array
currently only compute the grand mean and std
"""
start = timeit.default_timer()
x = []
y = []
for fn in glob.glob(os.path.join(FT_DIR, "*.npy")):
start = fn.rfind('/')
end = fn.rfind('.')
ext = fn[start+1:end]
genre, _= ext.split('_')
data = np.load(fn)
surface_ft = data[:-1] #5 features
ft_vec = [np.mean(ft) for ft in surface_ft] + [np.std(ft) for ft in surface_ft]
ceps = data[-1]#mfcc features
cep_len = len(ceps)
ft_vec += np.mean(ceps[int(cep_len / 10.):int(cep_len * 9 / 10.)], axis=0).tolist()
x.append(ft_vec)
y.append(GENRE_DICT[genre])
end = timeit.default_timer()
print("reading all features takes: ", (end - start))
return np.array(x), np.array(y)
def index_project(self, project_name):
project_data = self.watcher.projects[project_name]["project_data"]
cfc_folders = project_data.get(self.folder_key, [])
mappings = project_data.get("mappings", [])
project_file_dir = os.path.dirname(project_name)
if len(cfc_folders) == 0:
return
start_time = timeit.default_timer()
index = {}
print("CFML: indexing components in project '" + project_name + "'")
for cfc_folder in sorted(cfc_folders, key=lambda d: d["path"]):
root_path = utils.normalize_path(cfc_folder["path"], project_file_dir)
path_index = self.parser.parse_directory(root_path)
index.update(path_index)
self.data[project_name] = {
"index": index,
"cache": {file_path: {} for file_path in index}
}
self.build_project_data(project_name)
index_time = timeit.default_timer() - start_time
message = "CFML: indexing components in project '{}' completed - {} files indexed in {:.2f} seconds"
print(message.format(project_name, str(len(index)), index_time))
self.notify_listeners(project_name)
def trim_data(crime_data, part, total_parts):
print 'Trimming unnecessary data...',
time1 = tm.default_timer()
crime_data = crime_data[crime_data['YEAR'] >= 2006]
crime_data = crime_data[crime_data['YEAR'] <= 2015]
crime_data = crime_data[pd.notnull(crime_data['NEIGHBOURHOOD'])]
crime_data = crime_data.drop('HUNDRED_BLOCK', axis=1)
crime_data = crime_data.sort_index()
if TEST_VAL:
print 'Taking subset of crime data (1000 row sample)...',
crime_data = crime_data.head(1005)
if part is not None and total_parts is not None:
start_index = int(1.0*(part-1)/total_parts*crime_data['YEAR'].count())
end_index = int(1.0*part/total_parts*crime_data['YEAR'].count())
if part == total_parts: end_index = crime_data['YEAR'].count()
crime_data = crime_data[start_index:end_index]
print 'Start index, end index, size:',start_index,end_index, crime_data['YEAR'].count()
print 'Finished'
print 'Time taken:', tm.default_timer()-time1, ' seconds\n'
return crime_data
def launch_jobs(quandl_codes, num_workers, calc_date, authtoken="", freq='M', span=60):
job_queue = Queue.Queue()
for b in quandl_codes:
job_queue.put(b)
print "Length %d"%job_queue.qsize()
thlist = []
s_time = timeit.default_timer()
fp = open("output.csv","w")
heading = "Ticker, Date, "+",".join(Worker._itemlist)+"\n"
fp.write(heading)
s_date = dutil.shift_months(calc_date, -(span+6))
trim_start = s_date.strftime('%Y-%m-%d')
trim_end = calc_date.strftime('%Y-%m-%d')
calc_param = {"calc_date": calc_date, "freq" : freq, "span":60}
for i in range(num_workers):
th = Worker(job_queue, trim_start, trim_end, calc_param, authtoken, fp)
th.daemon = True
th.start()
thlist.append(th)
print "Finished launching jobs"
e_time = timeit.default_timer()
print "Time taken ",(e_time - s_time)
# block until the queue is empty
job_queue.join()
def spawn_runpy(cp, wait=60, cb=check_rst):
"as decorator to run job"
global WAITQ, RUNQ, CFG
pool = Pool(processes=CFG['MAXJOBS'])
while len(WAITQ) > 0 or len(RUNQ) > 0:
if len(RUNQ) <= CFG['MAXJOBS'] and len(WAITQ) > 0:
path, test = WAITQ.pop()
rst = pool.apply_async(call_runpy, (cp, path, test,))
RUNQ.append((rst, test, timeit.default_timer()))
else:
for r in RUNQ:
usec = float("%.2f" %(timeit.default_timer()-r[2]))
if r[0].successful:
print "[{0}] success used {1} usec".format(r[1], usec)
RUNQ.remove(r)
if cb:
cb(r[1], 'pass', usec)
else:
if usec > CFG['TIMEOUT']:
print "[{0}] unsuccess used timeout {1} usec".format(r[1], usec)
r[0].terminate()
if cb:
cb(r[1], 'fail', usec)
time.sleep(float(wait))
def test_exercise_6(self):
con = self.con
con.isolation_level = None
cur = con.cursor()
N = 30000
#############################
# Exercise 6
#
# Change the following schema to include an index on column "a".
cur.execute('CREATE TABLE "numbers" (a INTEGER)')
#
#
#############################
rows = []
for i in range(0, N):
rows.append( (i,) )
cur.executemany('INSERT INTO "numbers" VALUES (?)', rows)
start_time = timeit.default_timer()
cur.execute('select min(a) from numbers')
print("exercise_6: That took %f ms." % ((timeit.default_timer() - start_time) * 1000,))
data = cur.fetchall()
cur.close()
self.assertTrue(data[0][0] == 0)
def evaluate(im, algo, gt_illuminant, i, range_thresh, bin_num, dst_folder):
new_im = None
start_time = timeit.default_timer()
if algo=="grayworld":
new_im = cv2.xphoto.autowbGrayworld(im, 0.95)
elif algo=="nothing":
new_im = im
elif algo=="learning_based":
new_im = cv2.xphoto.autowbLearningBased(im, None, range_thresh, 0.98, bin_num)
elif algo=="GT":
gains = gt_illuminant / min(gt_illuminant)
g1 = float(1.0 / gains[2])
g2 = float(1.0 / gains[1])
g3 = float(1.0 / gains[0])
new_im = cv2.xphoto.applyChannelGains(im, g1, g2, g3)
time = 1000*(timeit.default_timer() - start_time) #time in ms
if len(dst_folder)>0:
if not os.path.exists(dst_folder):
os.makedirs(dst_folder)
im_name = ("%04d_" % i) + algo + ".jpg"
cv2.imwrite(os.path.join(dst_folder, im_name), stretch_to_8bit(new_im))
#recover the illuminant from the color balancing result, assuming the standard model:
estimated_illuminant = [0, 0, 0]
eps = 0.01
estimated_illuminant[2] = np.percentile((im[:,:,0] + eps) / (new_im[:,:,0] + eps), 50)
estimated_illuminant[1] = np.percentile((im[:,:,1] + eps) / (new_im[:,:,1] + eps), 50)
estimated_illuminant[0] = np.percentile((im[:,:,2] + eps) / (new_im[:,:,2] + eps), 50)
res = np.arccos(np.dot(gt_illuminant,estimated_illuminant)/
(np.linalg.norm(gt_illuminant) * np.linalg.norm(estimated_illuminant)))
return (time, (res / np.pi) * 180)
def execute(self):
start_time = timeit.default_timer()
response = self.svc.call()
end_time = timeit.default_timer()
self.elapsed_time = end_time - start_time
return self.validate(response)
def runTestCode(self):
"""
This function ties into the debug menu. It is meant to allow execution
of some test code. Feel free to change the contents of this function.
"""
start = timeit.default_timer()
monsters = []
lib = Libraries.MonsterLibrary()
stop = timeit.default_timer()
time = stop - start
print "Created library in " + str(time) + " seconds"
for i in range(0, 10000):
myRandom = lib.getRandomMonster(random.randint(0, 80))
monsters.append(myRandom)
# lib = Libraries.ItemLibrary()
# myItem = lib.createItem('heal')
# print myItem
# myItem = lib.createItem('sword')
# print myItem
# myItem = lib.createItem('cloak')
# print myItem
# myItem = lib.createItem('fireball')
# print myItem
stop = timeit.default_timer()
time = stop - start
print "Created " + str(len(monsters)) + " monsters in " + str(time) + " seconds"
def scan_vocab(self, documents, progress_per=10000, trim_rule=None):
logger.info("collecting all words and their counts")
document_no = -1
total_words = 0
min_reduce = 1
interval_start = default_timer() - 0.00001 # guard against next sample being identical
interval_count = 0
vocab = defaultdict(int)
for document_no, document in enumerate(documents):
if document_no % progress_per == 0:
interval_rate = (total_words - interval_count) / (default_timer() - interval_start)
logger.info("PROGRESS: at example #%i, processed %i words (%i/s), %i word types, %i tags",
document_no, total_words, interval_rate, len(vocab), len(self.docvecs))
interval_start = default_timer()
interval_count = total_words
document_length = len(document.words)
for tag in document.tags:
self.docvecs.note_doctag(tag, document_no, document_length)
for word in document.words:
vocab[word] += 1
total_words += len(document.words)
if self.max_vocab_size and len(vocab) > self.max_vocab_size:
utils.prune_vocab(vocab, min_reduce, trim_rule=trim_rule)
min_reduce += 1
logger.info("collected %i word types and %i unique tags from a corpus of %i examples and %i words",
len(vocab), len(self.docvecs), document_no + 1, total_words)
self.corpus_count = document_no + 1
self.raw_vocab = vocab
def run(self):
try:
self.running = True
self.logger.info("[Relay] Relay controller activated in "
"{}ms".format((timeit.default_timer()-self.thread_startup_timer)*1000))
while (self.running):
current_time = datetime.datetime.now()
for relay_id in self.relay_id:
if (self.relay_on_until[relay_id] < current_time and
self.relay_on_duration[relay_id] and
self.relay_pin[relay_id]):
# Use threads to prevent a slow execution of a
# process that could slow the loop
turn_relay_off = threading.Thread(
target=self.relay_on_off,
args=(relay_id, 'off',))
turn_relay_off.start()
if self.relay_last_duration[relay_id] > 0:
write_db = threading.Thread(
target=write_influxdb,
args=(self.logger, INFLUXDB_HOST,
INFLUXDB_PORT, INFLUXDB_USER,
INFLUXDB_PASSWORD, INFLUXDB_DATABASE,
'relay', relay_id, 'duration_sec',
float(self.relay_last_duration[relay_id]),))
write_db.start()
time.sleep(0.01)
finally:
self.all_relays_off()
self.running = False
self.logger.info("[Relay] Relay controller deactivated in "
"{}ms".format((timeit.default_timer()-self.thread_shutdown_timer)*1000))
def train(self, examples, cv_extract, epochs, learning_rate):
"""
Specializes the network for prediction on the given examples, using the
given center extract function, the given number of epochs, and the given
learning rate.
"""
input = T.vector(name="training_input", dtype=theano.config.floatX)
tf = self.get_specialization_function(input, cv_extract, learning_rate)
indices = list(range(examples.get_value(borrow=True).shape[0]))
start_time = timeit.default_timer()
# TODO: batches?
for epoch in range(epochs):
self.rng.shuffle(indices)
costs = []
for j in indices:
cost = tf(examples.get_value(borrow=True)[j].reshape(-1))
costs.append(cost)
debug(
"... [{}] epoch {: 3d} done {} ...".format(
str(datetime.timedelta(seconds=timeit.default_timer()-start_time)),
epoch + 1,
"(min/avg cost {:0.3f}/{:0.3f})".format(
float(float(min(costs))),
float(float(sum(costs)/float(len(costs))))
)
)
)
def simulate(new_N = N, new_R = R, new_D = D): global N N = new_N global R R = new_R global D D = new_D global distance_arr distance_arr = [([0] * N) for i in xrange(N)] global sensor_network sensor_network = [Sensor(i) for i in range(N)] for x in xrange(N): for y in xrange(N): if x != y and distance_arr[x][y] == 0: distance_arr[x][y] = sensor_distance(sensor_network[x].position, sensor_network[y].position) distance_arr[y][x] = distance_arr[x][y] start = timeit.default_timer() [s.start() for s in sensor_network] [s.join() for s in sensor_network] stop = timeit.default_timer() return stop - start
def evaluate(self, p, sim, plt):
start = timeit.default_timer()
sim.run(p.T)
end = timeit.default_timer()
speed = p.T / (end - start)
data = sim.data[self.p_ens]
last = []
for row in data.T:
nz = np.nonzero(row>0.05)[0]
if len(nz) == 0:
last.append(0)
else:
last.append(nz[-1])
time_to_inhibit = np.array(last)*p.dt
if plt:
plt.plot(sim.trange(), sim.data[self.p_ens])
for t in time_to_inhibit:
plt.axvline(t)
plt.axhline(0.05, linestyle='--', c='k')
plt.xlabel('time (s) with increasing inhibition')
plt.ylabel('decoded output')
return dict(time_to_inhibit=np.mean(time_to_inhibit),
speed=speed)
def load_indicators_to_mongo_zh(is_incremental):
print("start loading indicator data(zh) from JSON file to MongoDB...")
all_start = timeit.default_timer()
static = Static()
f = io.open(static.output_folder + '/worldbank_wdi_indicators_zh.json', 'r', encoding='utf8', errors='ignore')
json_str = f.readline()
indicator_array = json.loads(json_str)
f.close()
client = MongoClient(static.mongo_url, static.mongo_port)
db = client[static.database_name]
## print(db.collection_names())
indicator_col = db[static.indicator_col_name]
if not is_incremental:
indicator_col.drop()
for ind in indicator_array:
indicator_key = ind['id'].replace('.', '_') + '_ZH'
data_type = 'number'
if(ind['name'].find('百分比') > -1):
data_type = 'percentage'
topics = []
for topic in ind['topics']:
topics.append(topic['value'])
indicator_rec = {'indicator_key': indicator_key, 'original_id': ind['id'], 'indicator_text': ind['name'], 'data_type': data_type, 'sourceOrganization': ind['sourceOrganization'], 'sourceNote': ind['sourceNote'], 'topics': topics, 'data_source': '世界发展指标', 'dimension': [{'dimension_key': 'year', 'dimension_text': '年'}, {'dimension_key': 'region', 'dimension_text': '区域'}, {'dimension_key': 'country', 'dimension_text': '国家'}]}
pk = indicator_col.insert(indicator_rec)
print(indicator_key + ' ' + ind['name'] + ' inserted.')
print("job is complete.")
print("total records: " + str(indicator_col.count()))
print("total time cost: " + str(round(timeit.default_timer() - all_start)) + 's')
def _run_analyzers_on_event(self):
'''Run all analysers on the current event, self.event.
Returns a tuple (success?, last_analyzer_name).
'''
for i,analyzer in enumerate(self._analyzers):
if not analyzer.beginLoopCalled:
analyzer.beginLoop(self.setup)
start = timeit.default_timer()
if self.memReportFirstEvent >=0 and iEv >= self.memReportFirstEvent:
memNow=resource.getrusage(resource.RUSAGE_SELF).ru_maxrss
if memNow > self.memLast :
print "Mem Jump detected before analyzer %s at event %s. RSS(before,after,difference) %s %s %s "%( analyzer.name, iEv, self.memLast, memNow, memNow-self.memLast)
self.memLast=memNow
ret = analyzer.process( self.event )
if self.memReportFirstEvent >=0 and iEv >= self.memReportFirstEvent:
memNow=resource.getrusage(resource.RUSAGE_SELF).ru_maxrss
if memNow > self.memLast :
print "Mem Jump detected in analyzer %s at event %s. RSS(before,after,difference) %s %s %s "%( analyzer.name, iEv, self.memLast, memNow, memNow-self.memLast)
self.memLast=memNow
if self.timeReport:
self.timeReport[i]['events'] += 1
if self.timeReport[i]['events'] > 0:
self.timeReport[i]['time'] += timeit.default_timer() - start
if ret == False:
return (False, analyzer.name)
return (True, analyzer.name)
def load_rowdata_to_mongo_zh(is_incremental):
print("start loading row data(zh) from JSON file to MongoDB...")
all_start = timeit.default_timer()
static = Static()
bydim_dir = static.output_folder + static.dataset_bydim_folder
client = MongoClient(static.mongo_url, static.mongo_port)
db = client[static.database_name]
dataset_col = db[static.dataset_col_name]
if not is_incremental:
dataset_col.drop()
file_path_array = []
for idx, file in enumerate(os.listdir(bydim_dir)):
file_path = os.path.join(bydim_dir, file)
if os.path.isfile(file_path):
file_path_array.append(file_path)
print(str(len(file_path_array)) + " files are loaded")
counter = []
mapfunc = partial(insert_by_dim, counter=counter, dataset_col=dataset_col, all_start=all_start)
pool = ThreadPool(12)
pool.map(mapfunc, file_path_array)
pool.close()
pool.join()
print("All the threads are completed. Total number is " + str(len(counter)) + "\n")
print("total time cost: " + str(round(timeit.default_timer() - all_start)) + 's')
def loop_sd_mean(alphabet):
print("======== sd-mean test===========")
start = timeit.default_timer()
count = 0
letters_number_list = []
entropy_list = []
for i in list(range(1,101)): # this is sd
alphabet1 = eliminate_sd(alphabet,i)
for j in list(range(1,101)): # this is mean
alphabet2 = eliminate_mean(alphabet1,j)
letters_number = len(alphabet2)
letters_number_list.append((i,j,letters_number))
balanced_alphabet = rebalance(alphabet2)
entropy = calculate_entropy(balanced_alphabet)
entropy_list.append((i,j,entropy))
count = count+1
print(count)
stop = timeit.default_timer()
time = (stop - start)
print (letters_number_list)
print (entropy_list)
print("======== sd-mean test===========")
print('Running Time (s): %f' %time)
def worker(F, chargers, sensors, p_list, sensors_p, p_list_p):
"""worker function, used to create processing"""
result = {}
tic = timeit.default_timer()
anser = reconfiguration.iaa.solution(chargers, sensors, p_list, args['B'], sensors_p, p_list_p, F, args['p_min'])
toc = timeit.default_timer()
result['IAA'] = (toc - tic, anser)
if DEBUG:
print "============================================="
print "# solution IAA #"
print "============================================="
pprint(anser)
tic = timeit.default_timer()
anser = solution.solutionOpt.solution(chargers, sensors_p, p_list_p)
toc = timeit.default_timer()
result['Opt'] = (toc - tic, anser)
if DEBUG:
print "============================================="
print "# solution Opt #"
print "============================================="
pprint(anser)
return result
def main():
"""
"""
logging.info("Reading file:%s", "data/sample.avi")
vid = AoRecording.AoRecording(filepath="data/sample.avi")
vid.load_video()
logging.info("Starting parallel processing")
tic = timeit.default_timer()
vid.filter_frames()
vid.fixed_align_frames()
vid.complete_align_parallel()
vid.create_average_frame()
vid.create_stdev_frame()
toc = timeit.default_timer()
print "Parallel Process took {}:".format(toc - tic)
vid.create_stdev_frame()
logging.info("writing output")
vid.write_video("output/output_parallel.avi")
vid.write_average_frame("output/lucky_average_parallel.png")
vid.write_frame("output/lucky_stdev.png", "stdev")
logging.info("Starting serial processing")
tic = timeit.default_timer()
vid.filter_frames()
vid.fixed_align_frames()
vid.complete_align()
vid.create_average_frame()
toc = timeit.default_timer()
print "Serial Process took {}:".format(toc - tic)
logging.info("writing output")
vid.write_video("output/output_serial.avi")
vid.write_frame("output/lucky_average_serial.png", "average")
def analyze_files(self, iterCount, loci_classes, adapt_threshold):
Rmodel = VRmodel.VregMRmodel(iterCount, loci_classes, adapt_threshold)
print "len(Rmodel.rfmodels)=", len(Rmodel.rfmodels)
ofile = open("bkg_out.dat","a+")
Rmodel.set_bckgoutfile( ofile )
for species in self.speciesList:
fbar= self.S[species]["WGS"]
print fbar
outFile = self.outDir + os.path.basename(fbar).replace(".fasta", "_"+str(iterCount)+"_outRF.fasta")
ofile = open(outFile,"w")
Rmodel.set_outfile( ofile )
fb = self.outDir + os.path.basename(fbar).replace(".fasta", "_"+str(iterCount)+"_exon.fasta")
exfile1 = open(fb,"w")
Rmodel.set_exon_outfiles( exfile1 )
start_time = timeit.default_timer()
gene_cnt=0
for strand in [1, -1]:
qbar=deepcopy(self.contigs)
print "STRAND=", strand
for record in SeqIO.parse(fbar, "fasta"):
if self.check_contigs:
if ( record.id.split("|")[3] not in self.contigs):
continue
print "record.id=", record.id
print "cnts=",record.id.split("|")[3]
print "qbar=", qbar
if self.check_contigs:
qbar.remove(record.id.split("|")[3])
if strand == 1:
seq=record.seq
else:
seq=record.seq.reverse_complement()
Rmodel.set_record(record.id, record.name, record.description)
seq_size=len(seq)
res= self.mapper( divide_work(seq) )
"""
print "len(res)=", len(res)
for ix in range(2):
print res[ix][0], res[ix][1], type(res[ix][2])
"""
Elist=Rmodel.exon_MRprobabilities(res)
gene_cnt = Rmodel.V_exon_model(gene_cnt, seq, strand, Elist)
#res=None
#Elist=None
if len(qbar)==0:
break
ofile.close()
elapsed = timeit.default_timer() - start_time
print "ELAPSED TIME =", elapsed
def pretrain(self, examples, epoch_counts, corruption_rates, learning_rates):
"""
Trains the network for autoencoding on the given examples, given lists of
epoch counts, corruption rates, and learning rates each equal in length to
the number of layers in the stack.
"""
tfs = self.get_training_functions(corruption_rates, learning_rates)
indices = list(range(examples.get_value(borrow=True).shape[0]))
start_time = timeit.default_timer()
for i in range(len(self.layers)):
# TODO: batches?
for epoch in range(epoch_counts[i]):
self.rng.shuffle(indices)
costs = []
for j in indices:
cost = tfs[i](examples.get_value(borrow=True)[j].reshape(-1))
costs.append(cost)
debug(
"... [{}] epoch {: 3d} at layer {: 2d} done {} ...".format(
str(datetime.timedelta(seconds=timeit.default_timer()-start_time)),
epoch + 1,
i,
"(min/avg cost {:0.3f}/{:0.3f})".format(
float(min(costs)),
float(sum(costs)/float(len(costs))),
)
)
)
from sklearn.ensemble import RandomForestClassifier, BaggingClassifier, GradientBoostingClassifier, AdaBoostClassifier
from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVC
from sklearn.decomposition import PCA
from sklearn.pipeline import Pipeline
from sklearn.utils import resample
from sklearn.feature_selection import SelectFromModel
from sklearn.neural_network import MLPClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.impute import SimpleImputer
import timeit
# In[2]: Import data
startzeitDaten = timeit.default_timer()
data = pd.read_csv(r"Data\SP500_data_new.csv"
,parse_dates = ["adate", "qdate", "public_date"], dayfirst = True)#, index_col=["gvkey", "datadate"])
data_NaN = data.dropna()
data_y = data.dropna(subset =["splticrm"])
Names1 = pd.read_excel(r"Data\Names1.xlsx", header = 0)
Names1 = Names1.drop(["Data Type", "Help"], axis = 1)
Names1.columns = ["Name", "ExName"]
Names2 = pd.read_excel(r"Data\Names2.xlsx", header = 0)
Names2 = Names2.drop(["Data Type","Help"], axis = 1)
Names2.columns = ["Name", "ExName"]
features1RF_mean = pd.read_csv(r"Data\RF1_mean.csv", header = 0).dropna()
features1RF_mean.columns = ["Ort","Name", "Wert"]
def __exit__(self, _1, _2, _3):
self.t_end = timeit.default_timer()
self.dt = self.t_end - self.t_start
def __enter__(self):
self.t_start = timeit.default_timer()
return self
g.add_node(5, pos = (2,2))
g.add_node(6, pos = (3,1))
g.add_node(7, pos = (4,1))
# create the edges in the graph
g.add_edge(1,2, weight = 50)
g.add_edge(1,3, weight = 50)
g.add_edge(2,4, weight = 10)
g.add_edge(2,5, weight = 20)
g.add_edge(4,6, weight = 20)
g.add_edge(5,6, weight = 10)
g.add_edge(6,7, weight = 20)
g.add_edge(3,7, weight = 50)
if __name__ == "__main__":
startTime = timeit.default_timer()
path1 = bfs(g,1,7,0)[0]
cost1 = bfsPathCost(g, 1, 7)
endTime = timeit.default_timer()
calculations = bfs(g,1,7,0)[1]
# find the runtime of the program in seconds
runTime = (endTime-startTime)*10**6
print('The verticies found by BFS: ' + str(path1))
print('Cost of BFS: ' + str(cost1))
print('Runtime in microseconds: ' + str(runTime))
def process_dump(
input_file,
template_file,
out_file,
file_size,
file_compress,
process_count,
html_safe,
):
"""
:param input_file: name of the wikipedia dump file; '-' to read from stdin
:param template_file: optional file with template definitions.
:param out_file: directory where to store extracted data, or '-' for stdout
:param file_size: max size of each extracted file, or None for no max (one file)
:param file_compress: whether to compress files with bzip.
:param process_count: number of extraction processes to spawn.
"""
global knownNamespaces
global templateNamespace, templatePrefix
global moduleNamespace, modulePrefix
urlbase = "" # This is obtained from <siteinfo>
input = decode_open(input_file)
# collect siteinfo
for line in input:
line = line # .decode('utf-8')
m = tagRE.search(line)
if not m:
continue
tag = m.group(2)
if tag == "base":
# discover urlbase from the xml dump file
# /mediawiki/siteinfo/base
base = m.group(3)
urlbase = base[:base.rfind("/")]
elif tag == "namespace":
knownNamespaces.add(m.group(3))
if re.search('key="10"', line):
templateNamespace = m.group(3)
templatePrefix = templateNamespace + ":"
elif re.search('key="828"', line):
moduleNamespace = m.group(3)
modulePrefix = moduleNamespace + ":"
elif tag == "/siteinfo":
break
if expand_templates:
# preprocess
template_load_start = default_timer()
if template_file and os.path.exists(template_file):
logging.info(
"Preprocessing '%s' to collect template definitions: this may take some time.",
template_file,
)
file = decode_open(template_file)
templates = load_templates(file)
file.close()
else:
if input_file == "-":
# can't scan then reset stdin; must error w/ suggestion to specify template_file
raise ValueError(
"to use templates with stdin dump, must supply explicit template-file"
)
logging.info(
"Preprocessing '%s' to collect template definitions: this may take some time.",
input_file,
)
templates = load_templates(input, template_file)
input.close()
input = decode_open(input_file)
template_load_elapsed = default_timer() - template_load_start
logging.info("Loaded %d templates in %.1fs", templates,
template_load_elapsed)
if out_file == "-":
output = sys.stdout
if file_compress:
logging.warn(
"writing to stdout, so no output compression (use an external tool)"
)
else:
nextFile = NextFile(out_file)
output = OutputSplitter(nextFile, file_size, file_compress)
# process pages
logging.info("Starting page extraction from %s.", input_file)
extract_start = default_timer()
# Parallel Map/Reduce:
# - pages to be processed are dispatched to workers
# - a reduce process collects the results, sort them and print them.
maxsize = 10 * process_count
# output queue
output_queue = Queue(maxsize=maxsize)
# Reduce job that sorts and prints output
reduce = Process(target=reduce_process, args=(output_queue, output))
reduce.start()
# initialize jobs queue
jobs_queue = Queue(maxsize=maxsize)
# start worker processes
logging.info("Using %d extract processes.", process_count)
workers = []
for _ in range(max(1, process_count)):
extractor = Process(target=extract_process,
args=(jobs_queue, output_queue, html_safe))
extractor.daemon = True # only live while parent process lives
extractor.start()
workers.append(extractor)
# Mapper process
# we collect individual lines, since str.join() is significantly faster
# than concatenation
page = []
id = ""
revid = ""
last_id = ""
ordinal = 0 # page count
inText = False
redirect = False
for line in input:
if "<" not in line: # faster than doing re.search()
if inText:
page.append(line)
continue
m = tagRE.search(line)
if not m:
continue
tag = m.group(2)
if tag == "page":
page = []
redirect = False
elif tag == "id" and not id:
id = m.group(3)
elif tag == "id" and id: # <revision> <id></id> </revision>
revid = m.group(3)
elif tag == "title":
title = m.group(3)
elif tag == "redirect":
redirect = True
elif tag == "text":
inText = True
line = line[m.start(3):m.end(3)]
page.append(line)
if m.lastindex == 4: # open-close
inText = False
elif tag == "/text":
if m.group(1):
page.append(m.group(1))
inText = False
elif inText:
page.append(line)
elif tag == "/page":
colon = title.find(":")
if (colon < 0 or (title[:colon] in acceptedNamespaces)
and id != last_id and not redirect
and not title.startswith(templateNamespace)):
job = (id, revid, urlbase, title, page, ordinal)
jobs_queue.put(job) # goes to any available extract_process
last_id = id
ordinal += 1
id = ""
revid = ""
page = []
input.close()
# signal termination
for _ in workers:
jobs_queue.put(None)
# wait for workers to terminate
for w in workers:
w.join()
# signal end of work to reduce process
output_queue.put(None)
# wait for it to finish
reduce.join()
if output != sys.stdout:
output.close()
extract_duration = default_timer() - extract_start
extract_rate = ordinal / extract_duration
logging.info(
"Finished %d-process extraction of %d articles in %.1fs (%.1f art/s)",
process_count,
ordinal,
extract_duration,
extract_rate,
)
def sgd_optimization_mnist(
learning_rate=0.013,
n_epochs=100,
dataset='D:\JupyterWorkspace\DeepLearningTutorial\mnist.pkl.gz',
batch_size=50):
"""
Demonstrate stochastic gradient descent optimization of a log-linear
model
This is demonstrated on MNIST.
:type learning_rate: float
:param learning_rate: learning rate used (factor for the stochastic
gradient)
:type n_epochs: int
:param n_epochs: maximal number of epochs to run the optimizer
:type dataset: string
:param dataset: the path of the MNIST dataset file from
http://www.iro.umontreal.ca/~lisa/deep/data/mnist/mnist.pkl.gz
"""
'''
datasets = load_data(dataset)
train_set_x, train_set_y = datasets[0]
valid_set_x, valid_set_y = datasets[1]
test_set_x, test_set_y = datasets[2]
'''
data = sio.loadmat(
'D:\ResearchWork\Machine learning and MI\Code\MEMD-DL\subject_1_scheme8.mat'
)
processedEEG = data['processedEEG']
processedERD = data['processedERD']
processedSpectrum = data['processedSpectrum']
trainLabels = data['trainLabels']
trainLabels = trainLabels.T
def extractbands(processedSpectrum, trainLabels):
muband = processedSpectrum[:512, :, :]
muband = muband[10:60, :, :]
betaband = processedSpectrum[512:1024, :, :]
betaband = betaband[29:100, :, :]
totfeat = muband.shape[0] + betaband.shape[0]
trainData = numpy.zeros((processedSpectrum.shape[2], totfeat * 2))
labels = numpy.zeros((processedSpectrum.shape[2]))
for i in range(
processedSpectrum.shape[2]): #(processedSpectrum.shape[2]
trainData[i, :] = numpy.concatenate(
(muband[:, 0, i], muband[:, 2, i], betaband[:, 0,
i], betaband[:, 2,
i]))
labels[i] = trainLabels[i]
return trainData, labels
trainData, labels = extractbands(processedSpectrum, trainLabels)
indices = numpy.random.permutation(processedEEG.shape[2])
training_idx, test_idx = indices[:350], indices[350:]
train_set_x, test_set_x = trainData[training_idx, :], trainData[
test_idx, :]
train_set_y, test_set_y = labels[training_idx], labels[test_idx]
valid_set_x = test_set_x
valid_set_y = test_set_y
test_set_x = theano.shared(numpy.asarray(test_set_x,
dtype=theano.config.floatX),
borrow=True)
test_set_y = theano.shared(numpy.asarray(test_set_y,
dtype=theano.config.floatX),
borrow=True)
test_set_y = T.cast(test_set_y, 'int32')
train_set_x = theano.shared(numpy.asarray(train_set_x,
dtype=theano.config.floatX),
borrow=True)
train_set_y = theano.shared(numpy.asarray(train_set_y,
dtype=theano.config.floatX),
borrow=True)
train_set_y = T.cast(train_set_y, 'int32')
valid_set_x = theano.shared(numpy.asarray(valid_set_x,
dtype=theano.config.floatX),
borrow=True)
valid_set_y = theano.shared(numpy.asarray(valid_set_y,
dtype=theano.config.floatX),
borrow=True)
valid_set_y = T.cast(valid_set_y, 'int32')
# compute number of minibatches for training, validation and testing
'''
n_train_batches = train_set_x.get_value(borrow=True).shape[0] // batch_size
n_valid_batches = valid_set_x.get_value(borrow=True).shape[0] // batch_size
n_test_batches = test_set_x.get_value(borrow=True).shape[0] // batch_size
'''
n_train_batches = train_set_x.get_value(borrow=True).shape[0] // batch_size
n_valid_batches = 1
n_test_batches = 1
######################
# BUILD ACTUAL MODEL #
######################
print('... building the model')
# allocate symbolic variables for the data
index = T.lscalar() # index to a [mini]batch
# generate symbolic variables for input (x and y represent a
# minibatch)
x = T.matrix('x') # data, presented as rasterized images
y = T.ivector('y') # labels, presented as 1D vector of [int] labels
# construct the logistic regression class
# Each MNIST image has size 28*28
classifier = LogisticRegression(input=x, n_in=242, n_out=2)
# the cost we minimize during training is the negative log likelihood of
# the model in symbolic format
cost = classifier.negative_log_likelihood(y)
# compiling a Theano function that computes the mistakes that are made by
# the model on a minibatch
test_model = theano.function(
inputs=[index],
outputs=classifier.errors(y),
givens={
x: test_set_x[index * batch_size:(index + 1) * batch_size],
y: test_set_y[index * batch_size:(index + 1) * batch_size]
})
validate_model = theano.function(
inputs=[index],
outputs=classifier.errors(y),
givens={
x: valid_set_x[index * batch_size:(index + 1) * batch_size],
y: valid_set_y[index * batch_size:(index + 1) * batch_size]
})
# compute the gradient of cost with respect to theta = (W,b)
g_W = T.grad(cost=cost, wrt=classifier.W)
g_b = T.grad(cost=cost, wrt=classifier.b)
# start-snippet-3
# specify how to update the parameters of the model as a list of
# (variable, update expression) pairs.
updates = [(classifier.W, classifier.W - learning_rate * g_W),
(classifier.b, classifier.b - learning_rate * g_b)]
# compiling a Theano function `train_model` that returns the cost, but in
# the same time updates the parameter of the model based on the rules
# defined in `updates`
train_model = theano.function(
inputs=[index],
outputs=cost,
updates=updates,
givens={
x: train_set_x[index * batch_size:(index + 1) * batch_size],
y: train_set_y[index * batch_size:(index + 1) * batch_size]
})
# end-snippet-3
###############
# TRAIN MODEL #
###############
print('... training the model')
# early-stopping parameters
patience = 5000 # look as this many examples regardless
patience_increase = 2 # wait this much longer when a new best is
# found
improvement_threshold = 0.995 # a relative improvement of this much is
# considered significant
validation_frequency = min(n_train_batches, patience // 2)
# go through this many
# minibatche before checking the network
# on the validation set; in this case we
# check every epoch
best_validation_loss = numpy.inf
test_score = 0.
start_time = timeit.default_timer()
done_looping = False
epoch = 0
while (epoch < n_epochs) and (not done_looping):
epoch = epoch + 1
for minibatch_index in range(n_train_batches):
minibatch_avg_cost = train_model(minibatch_index)
# iteration number
iter = (epoch - 1) * n_train_batches + minibatch_index
if (iter + 1) % validation_frequency == 0:
# compute zero-one loss on validation set
validation_losses = [
validate_model(i) for i in range(n_valid_batches)
]
this_validation_loss = numpy.mean(validation_losses)
print('epoch %i, minibatch %i/%i, validation error %f %%' %
(epoch, minibatch_index + 1, n_train_batches,
this_validation_loss * 100.))
# if we got the best validation score until now
if this_validation_loss < best_validation_loss:
#improve patience if loss improvement is good enough
if this_validation_loss < best_validation_loss * improvement_threshold:
patience = max(patience, iter * patience_increase)
best_validation_loss = this_validation_loss
# test it on the test set
test_losses = [
test_model(i) for i in range(n_test_batches)
]
test_score = numpy.mean(test_losses)
print((' epoch %i, minibatch %i/%i, test error of'
' best model %f %%') %
(epoch, minibatch_index + 1, n_train_batches,
test_score * 100.))
# save the best model
with open('best_model.pkl', 'wb') as f:
pickle.dump(classifier, f)
if patience <= iter:
done_looping = True
break
end_time = timeit.default_timer()
print(('Optimization complete with best validation score of %f %%,'
'with test performance %f %%') %
(best_validation_loss * 100., test_score * 100.))
print('The code run for %d epochs, with %f epochs/sec' %
(epoch, 1. * epoch / (end_time - start_time)))
'''
plt.axvline(i, color='k', lw=1)
for i in spk0[1].spike_times:
plt.axvline(i, color='r', lw=1)
ax.set_ylabel('$w$')
# ax = fig.add_subplot(4, 2, 8)
# ax.plot(wt[:, 1, 1], 'r', lw=3)
# ax.set_ylabel('$w$')
# fig = plt.figure(figsize=(10, 5))
# for i in spk[0].spike_times:
# plt.plot(wt[:, 1, 1], 'r', lw=3)
# plt.axvline(i, color='k', lw=1)
plt.savefig('/tmp/%s.png' %
(os.path.splitext(os.path.basename(__file__))[0]))
plt.close()
print('End %s:run()' % (os.path.splitext(os.path.basename(__file__))[0]))
if __name__ == '__main__':
print('Begin %s:main()' %
(os.path.splitext(os.path.basename(__file__))[0]))
start_t = timeit.default_timer()
setup()
run()
print("End %s:main() , running time: %f seconds" % (os.path.splitext(
os.path.basename(__file__))[0], timeit.default_timer() - start_t))
filename_in = "trans-out.csv"
user_f = "users-large.csv"
# Read csv
in_read = csv.reader(open(filename_in,"rb"), delimiter=',',quoting=csv.QUOTE_ALL)
users_read = csv.reader(open(user_f,"rb"), delimiter=',',quoting=csv.QUOTE_ALL)
in_read.next()
users_read.next()
# Create user dict
# Optimization problem here if user list doesn't fit in memory
users_dict = {}
for user in users_read:
users_dict[user[0]] = {"spending-limit" : user[1], "round-to" : user[2]}
start = timeit.default_timer()
for row in in_read:
# row[0] == time
# row[1] == account
# row[2] == transaction number
# row[3] == amount
# See if account number is our users
if row[1] in users_dict:
user_info = users_dict[row[1]]
if float(row[3]) <= float(user_info["spending-limit"]):
change = roundUp(float(row[3]), float(user_info["round-to"]))
if change != "0.0":
sendToBackend(row[0], change, row[2], row[1])
print "REQUEST TRANS AMNT " + change + " FROM " + row[1] + " REFNUM: " + row[2]
else:
def train(self,
sentences,
total_words=None,
word_count=0,
total_examples=None,
queue_factor=2,
report_delay=1.0):
""" Update the model's neural weights from a sequence of sentences (can be a
once-only generator stream).
For Word2Vec, each sentence must be a list of unicode strings.
(Subclasses may accept other examples.)
To support linear learning-rate decay from (initial) alpha to min_alpha,
either total_examples
(count of sentences) or total_words (count of raw words in sentences)
should be provided, unless the
sentences are the same as those that were used to initially build the
vocabulary.
"""
logger.info("Starting training.")
self.neg_labels = []
if self.negative > 0:
# precompute negative labels optimization for pure-python training
self.neg_labels = zeros(self.negative + 1)
self.neg_labels[0] = 1.
if FAST_VERSION < 0:
import warnings
warnings.warn(
"C extension not loaded for Word2Vec, training will be slow. "
"Install a C compiler and reinstall gensim for fast training.")
self.neg_labels = []
if self.negative > 0:
# precompute negative labels optimization for pure-python training
self.neg_labels = zeros(self.negative + 1)
self.neg_labels[0] = 1.
logger.info(
"training model with %i workers on %i vocabulary and %i features, "
"using sg=%s hs=%s sample=%s negative=%s window=%s", self.workers,
len(self.vocab), self.layer1_size, self.sg, self.hs, self.sample,
self.negative, self.window)
if not self.vocab:
raise RuntimeError(
"you must first build vocabulary before training the model")
if not hasattr(self, "syn0"):
raise RuntimeError(
"you must first finalize vocabulary before training the model")
if total_words is None and total_examples is None:
if self.corpus_count:
total_examples = self.corpus_count
logger.info(
"expecting %i sentences, matching count from corpus used for vocabulary survey",
total_examples)
else:
raise ValueError(
"you must provide either total_words or total_examples, to enable alpha and progress calculations"
)
job_tally = 0
if self.iter > 1:
sentences = utils.RepeatCorpusNTimes(sentences, self.iter)
total_words = total_words and total_words * self.iter
total_examples = total_examples and total_examples * self.iter
def worker_loop():
"""Train the model, lifting lists of sentences from the job_queue."""
work = matutils.zeros_aligned(
self.layer1_size, dtype=REAL) # per-thread private work memory
neu1 = matutils.zeros_aligned(self.layer1_size, dtype=REAL)
jobs_processed = 0
while True:
job = job_queue.get()
if job is None:
progress_queue.put(None)
break # no more jobs => quit this worker
sentences, pairwise, alpha = job
tally, raw_tally = self._do_train_job(sentences, pairwise, alpha,
(work, neu1))
progress_queue.put(
(len(sentences), tally, raw_tally)) # report back progress
jobs_processed += 1
logger.debug("worker exiting, processed %i jobs", jobs_processed)
def job_producer():
"""Fill jobs queue using the input `sentences` iterator."""
job_batch, batch_size = [], 0
pushed_words, pushed_examples = 0, 0
next_alpha = self.alpha
if next_alpha > self.min_alpha_yet_reached:
logger.warn("Effective 'alpha' higher than previous training cycles")
self.min_alpha_yet_reached = next_alpha
job_no = 0
for sent_idx, sentence in enumerate(sentences):
sentence_length = self._raw_word_count([sentence])
# can we fit this sentence into the existing job batch?
if batch_size + sentence_length <= self.batch_words:
# yes => add it to the current job
job_batch.append(sentence)
batch_size += sentence_length
else:
# no => submit the existing job
pair_idx = list(
numpy.random.choice(
range(len(self.pairwise_constraints)), int(batch_size * 0.2)))
pairwise_samples = [self.pairwise_constraints[x] for x in pair_idx]
logger.debug(
"queueing job #%i (%i words, %i sentences, %i constraints) at alpha %.05f",
job_no, batch_size, len(job_batch), len(pairwise_samples),
next_alpha)
job_no += 1
job_queue.put((job_batch, pairwise_samples, next_alpha))
# update the learning rate for the next job
if self.min_alpha < next_alpha:
if total_examples:
# examples-based decay
pushed_examples += len(job_batch)
progress = 1.0 * pushed_examples / total_examples
else:
# words-based decay
pushed_words += self._raw_word_count(job_batch)
progress = 1.0 * pushed_words / total_words
next_alpha = self.alpha - (self.alpha - self.min_alpha) * progress
next_alpha = max(self.min_alpha, next_alpha)
# add the sentence that didn't fit as the first item of a new job
job_batch, batch_size = [sentence], sentence_length
# add the last job too (may be significantly smaller than batch_words)
if job_batch:
logger.debug(
"queueing job #%i (%i words, %i sentences, %i constraints) at alpha %.05f",
job_no, batch_size, len(job_batch), len(self.pairwise_constraints),
next_alpha)
job_no += 1
job_queue.put((job_batch, self.pairwise_constraints, next_alpha))
if job_no == 0 and self.train_count == 0:
logger.warning(
"train() called with an empty iterator (if not intended, "
"be sure to provide a corpus that offers restartable "
"iteration = an iterable).")
# give the workers heads up that they can finish -- no more work!
for _ in xrange(self.workers):
job_queue.put(None)
logger.debug("job loop exiting, total %i jobs", job_no)
# buffer ahead only a limited number of jobs.. this is the reason we can't simply use ThreadPool :(
job_queue = Queue(maxsize=queue_factor * self.workers)
progress_queue = Queue(maxsize=(queue_factor + 1) * self.workers)
workers = [
threading.Thread(target=worker_loop) for _ in xrange(self.workers)
]
unfinished_worker_count = len(workers)
workers.append(threading.Thread(target=job_producer))
for thread in workers:
thread.daemon = True # make interrupting the process with ctrl+c easier
thread.start()
example_count, trained_word_count, raw_word_count = 0, 0, word_count
start, next_report = default_timer() - 0.00001, 1.0
while unfinished_worker_count > 0:
report = progress_queue.get() # blocks if workers too slow
if report is None: # a thread reporting that it finished
unfinished_worker_count -= 1
logger.info(
"worker thread finished; awaiting finish of %i more threads",
unfinished_worker_count)
continue
examples, trained_words, raw_words = report
job_tally += 1
# update progress stats
example_count += examples
trained_word_count += trained_words # only words in vocab & sampled
raw_word_count += raw_words
# log progress once every report_delay seconds
elapsed = default_timer() - start
if elapsed >= next_report:
if total_examples:
# examples-based progress %
logger.info(
"PROGRESS: at %.2f%% examples, %.0f words/s, in_qsize %i, out_qsize %i",
100.0 * example_count / total_examples,
trained_word_count / elapsed, utils.qsize(job_queue),
utils.qsize(progress_queue))
else:
# words-based progress %
logger.info(
"PROGRESS: at %.2f%% words, %.0f words/s, in_qsize %i, out_qsize %i",
100.0 * raw_word_count / total_words,
trained_word_count / elapsed, utils.qsize(job_queue),
utils.qsize(progress_queue))
next_report = elapsed + report_delay
# all done; report the final stats
elapsed = default_timer() - start
logger.info(
"training on %i raw words (%i effective words) took %.1fs, %.0f effective words/s",
raw_word_count, trained_word_count, elapsed,
trained_word_count / elapsed)
if job_tally < 10 * self.workers:
logger.warn(
"under 10 jobs per worker: consider setting a smaller `batch_words' for smoother alpha decay"
)
# check that the input corpus hasn't changed during iteration
if total_examples and total_examples != example_count:
logger.warn(
"supplied example count (%i) did not equal expected count (%i)",
example_count, total_examples)
if total_words and total_words != raw_word_count:
logger.warn(
"supplied raw word count (%i) did not equal expected count (%i)",
raw_word_count, total_words)
self.train_count += 1 # number of times train() has been called
self.total_train_time += elapsed
self.clear_sims()
return trained_word_count
def main():
"""Parse command line options/arguments and execute."""
try:
arg_names = [
"help", "version", "quick", "strict", "debug", "stop-tag="
]
opts, args = getopt.getopt(sys.argv[1:], "hvqsdct:v", arg_names)
except getopt.GetoptError:
usage(2)
detailed = True
stop_tag = DEFAULT_STOP_TAG
debug = False
strict = False
color = False
for option, arg in opts:
if option in ("-h", "--help"):
usage(0)
if option in ("-v", "--version"):
show_version()
if option in ("-q", "--quick"):
detailed = False
if option in ("-t", "--stop-tag"):
stop_tag = arg
if option in ("-s", "--strict"):
strict = True
if option in ("-d", "--debug"):
debug = True
if option in ("-c", "--color"):
color = True
if not args:
usage(2)
exif_log.setup_logger(debug, color)
# output info for each file
for filename in args:
file_start = timeit.default_timer()
try:
img_file = open(str(filename), 'rb')
except IOError:
logger.error("'%s' is unreadable", filename)
continue
logger.info("Opening: %s", filename)
tag_start = timeit.default_timer()
# get the tags
data = process_file(img_file,
stop_tag=stop_tag,
details=detailed,
strict=strict,
debug=debug)
tag_stop = timeit.default_timer()
if not data:
logger.warning("No EXIF information found\n")
continue
if 'JPEGThumbnail' in data:
logger.info('File has JPEG thumbnail')
del data['JPEGThumbnail']
if 'TIFFThumbnail' in data:
logger.info('File has TIFF thumbnail')
del data['TIFFThumbnail']
tag_keys = list(data.keys())
tag_keys.sort()
for i in tag_keys:
try:
logger.info('%s (%s): %s', i,
FIELD_TYPES[data[i].field_type][2],
data[i].printable)
except:
logger.error("%s : %s", i, str(data[i]))
file_stop = timeit.default_timer()
logger.debug("Tags processed in %s seconds", tag_stop - tag_start)
logger.debug("File processed in %s seconds", file_stop - file_start)
print("")
ds_model = ds_model.sel(nregions=cR)
# # Take mean of ensemble
# ds_model = ds_model.mean(dim='ensemble')
# Get model plotting specs
cc = E.model_color[cmod]
cl = E.model_linestyle[cmod]
# Plot Model
if i == 0: # Control only one initiailzation label in legend
no_init_label = False
else:
no_init_label = True
import timeit
start_time = timeit.default_timer()
ice_plot.plot_reforecast(ds=ds_model, axin=ax1,
labelin=E.model[cmod]['model_label'],
color=cc, marker=None,
linestyle=cl,
no_init_label=no_init_label)
print( (timeit.default_timer() - start_time), ' seconds.' )
# Memeory clean up
ds_model = None
cxlims = ax1.get_xlim()
# add obs and climotrend
if (cR==99):
def transposonmapper(bamfile=bam_arg,
gfffile=None,
essentialfiles=None,
genenamesfile=None):
'''
This function is created for analysis of SATAY data using the species Saccharomyces Cerevisiae.
It outputs the following files that store information regarding the location of all insertions:
- .bed-file: Includes all individual basepair locations of the whole genome where at least one transposon has been mapped and the number of insertions for each locations (the number of reads) according to the Browser Extensible Data (bed) format.
A distinction is made between reads that had a different reading orientation during sequencing. The number of reads are stored using the equation #reads*20+100 (e.g. 2 reads is stored as 140).
- .wig-file: Includes all individual basepair locations of the whole genome where at least one transposon has been mapped and the number of insertions for each locations (the number of reads) according to the Wiggle (wig) format.
In this file no distinction is made between reads that had a different reading orientation during sequencing. The number of reads are stored as the absolute count.
- _pergene.txt-file: Includes all genes (currently 6600) with the total number of insertions and number of reads within the genomic region of the gene.
- _peressential.txt-file: Includes all annotated essential genes (currently 1186) with the total number of insertions and number of reads within the genomic region of the gene.
- _pergene_insertions.txt-file: Includes all genes with their genomic location (i.e. chromosome number, start and end position) and the locations of all insertions within the gene location. It also include the number number of reads per insertions.
- _peressential_insertions.txt-file: Includes all essential genes with their genomic location (i.e. chromosome number, start and end position) and the locations of all insertions within the gene location. It also include the number number of reads per insertions.
(note that in the latter two files, the genomic locations are continous, for example chromosome II does not start at 0, but at 'length chromosome I + 1' etc.).
The output files are saved at the location of the input file using the same name as the input file, but with the corresponding extension.
The function assumes that the reads are already aligned to a reference genome.
The input data should be a .bam-file and the location where the .bam-file is stored should also contain an index file (.bam.bai-file, which for example can be created using sambamba).
This function takes the following inputs:
- bamfile [required]: Path to the bamfile. This location should also contain the .bam.bai index file (does not need to be input in this function).
- gfffile [optional]: Path to a .gff-file including all gene information (e.g. downloaded from SGD). Default file is 'Saccharomyces_cerevisiae.R64-1-1.99.gff3'.
- essentialfiles [optional]: Path to a .txt file containing a list all essential genes. Every line should consist of a single essential gene and the file should have one header line. Ideally this file is created using 'Create_EssentialGenes_list.py'. Default file is 'Cerevisiae_AllEssentialGenes_List.txt'.
- genenamesfile [optional]: Path to text file that includes aliases for all genes. Default file is 'Yeast_Protein_Names.txt'.
When the arguments for the optional files are not given, the files are used that are stored at the following location:
"path_current_pythonscript/../data_files"
The function uses the pysam package for handling bam files (see pysam.readthedocs.io/en/latest/index.html) and therefore this function only runs on Linux systems with SAMTools installed.
'''
#%% LOADING BAM FILE
if bamfile is None:
path = os.path.join('/home', 'gregoryvanbeek', 'Documents',
'data_processing')
filename = 'SRR062634.filt_trimmed.sorted.bam'
bamfile = os.path.join(path, filename)
else:
filename = os.path.basename(bamfile)
path = bamfile.replace(filename, '')
if os.path.isfile(bamfile):
print('Running: ', bamfile)
else:
raise ValueError('Bam file not found at: ', bamfile)
#%% LOADING ADDITIONAL FILES
files_path = os.path.join(dirname, '..', '..', 'data_files')
#LOADING GFF-FILE
if gfffile is None:
gfffile = os.path.join(files_path,
'Saccharomyces_cerevisiae.R64-1-1.99.gff3')
if not os.path.isfile(gfffile):
raise ValueError('Path to GFF-file does not exist.')
#LOADING TEXT FILES WITH ESSENTIAL GENES
if essentialfiles is None:
essentialfiles = os.path.join(files_path,
'Cerevisiae_AllEssentialGenes_List.txt')
if not os.path.isfile(essentialfiles):
raise ValueError('Following path does not exist: ' + essentialfiles)
del essentialfiles
#LOADING TEXT FILE WITH GENE NAME ALIASES
if genenamesfile is None:
genenamesfile = os.path.join(files_path, 'Yeast_Protein_Names.txt')
if not os.path.isfile(genenamesfile):
raise ValueError('Following path does not exist: ' + genenamesfile)
#%% READ BAM FILE
bam = pysam.AlignmentFile(bamfile,
'rb') #open bam formatted file for reading
#%% GET NAMES OF ALL CHROMOSOMES AS STORED IN THE BAM FILE
ref_tid_dict = {} # 'I' | 0, 'II' | 1, ...
ref_name_list = [] # 'I', 'II', ...
for i in range(
bam.nreferences
): #if bam.nreferences does not work, use range(17) #16 chromosomes and the mitochondrial chromosome
ref_name = bam.get_reference_name(i)
ref_tid_dict[ref_name] = bam.get_tid(ref_name)
ref_name_list.append(ref_name)
del (ref_name, i)
#%% CONVERT CHROMOSOME NAMES IN DATA FILE TO ROMAN NUMERALS
ref_romannums = chromosomename_roman_to_arabic()[0]
ref_tid_roman_dict = {}
for key, val in ref_tid_dict.items():
ref_tid_roman_dict[ref_romannums[int(val) + 1]] = key
del (key, val, ref_romannums)
#%% GET SEQUENCE LENGTHS OF ALL CHROMOSOMES
chr_length_dict = {} # 'I' | 230218, 'II' | 813184, ...
chr_summedlength_dict = {} # 'I' | 0, 'II' | 230218, 'III' | 1043402, ...
ref_summedlength = 0
for key in ref_tid_dict:
ref_length = bam.get_reference_length(key)
chr_length_dict[key] = ref_length
chr_summedlength_dict[key] = ref_summedlength
ref_summedlength += ref_length
del (key, ref_length, ref_summedlength)
#%% GET NUMBER OF MAPPED, UNMAPPED AND TOTAL AMOUNT OF READS PER CHROMOSOME
# total_reads = bam.mapped
stats = bam.get_index_statistics()
chr_mappedreads_dict = {} # 'I' | [mapped, unmapped, total reads]
for stat in stats:
chr_mappedreads_dict[stat[0]] = [stat[1], stat[2], stat[3]]
if stat[2] != 0:
warnings.warn('Unmapped reads found in chromosome ' + stat[0])
del (stat, stats)
#%% GET ALL READS WITHIN A SPECIFIED GENOMIC REGION
tnnumber_dict = {}
ll = 0 #Number of unique insertions in entire genome
for kk in ref_name_list:
timer_start = timeit.default_timer()
read_counter = 0
N_reads_kk = chr_mappedreads_dict[kk][2]
start_array = np.empty(shape=(N_reads_kk), dtype=int)
flag_array = np.empty(shape=(N_reads_kk), dtype=int)
readlength_array = np.empty(shape=(N_reads_kk), dtype=int)
#RETREIVING ALL THE READS FROM THE CURRENT CHROMOSOME.
print('Getting reads for chromosome %s ...' % kk)
for reads in bam.fetch(kk, 0, chr_length_dict[kk], until_eof=True):
read = str(reads).split('\t')
start_array[read_counter] = int(read[3]) + 1
flag_array[read_counter] = int(read[1])
readlength_array[read_counter] = int(len(read[9]))
read_counter += 1
#CORRECT STARTING POSITION FOR READS WITH REVERSED ORIENTATION
flag0coor_array = np.where(
flag_array == 0) #coordinates reads 5' -> 3'
flag16coor_array = np.where(
flag_array == 16) # coordinates reads 3' -> 5'
startdirect_array = start_array[flag0coor_array]
flagdirect_array = flag_array[flag0coor_array]
startindirect_array = start_array[flag16coor_array] + readlength_array[
flag16coor_array]
flagindirect_array = flag_array[flag16coor_array]
start2_array = np.concatenate((startdirect_array, startindirect_array),
axis=0)
flag2_array = np.concatenate((flagdirect_array, flagindirect_array),
axis=0)
del (flag0coor_array, flag16coor_array, startdirect_array,
flagdirect_array, startindirect_array, flagindirect_array)
start2_sortindices = start2_array.argsort(
kind='mergesort') #use mergesort for stable sorting
start2_array = start2_array[start2_sortindices]
flag2_array = flag2_array[start2_sortindices]
del start2_sortindices
#CREATE ARRAY OF START POSITION AND FLAGS OF ALL READS IN GENOME
ref_tid_kk = int(ref_tid_dict[kk] + 1)
if ll == 0:
tncoordinates_array = np.array([])
mm = 0 # Number of unique reads per insertion
jj = 1 # Number of unique reads in current chromosome (Number of transposons in current chromosome)
for ii in range(1, len(start2_array)):
if abs(
start2_array[ii] - start2_array[ii - 1]
) <= 2 and flag2_array[ii] == flag2_array[
ii -
1]: #If two subsequent reads are within two basepairs and have the same orientation, add them together.
mm += 1
else:
avg_start_pos = abs(
round(np.mean(start2_array[ii - mm - 1:ii])))
if tncoordinates_array.size == 0: #include first read
tncoordinates_array = np.array([
ref_tid_kk,
int(avg_start_pos),
int(flag2_array[ii - 1])
])
readnumb_list = [mm + 1]
else:
tncoordinates_array = np.vstack((tncoordinates_array, [
ref_tid_kk,
int(avg_start_pos),
int(flag2_array[ii - 1])
]))
readnumb_list.append(mm + 1)
mm = 0
jj += 1
ll += 1
if ii == len(start2_array) - 1: #include last read
avg_start_pos = abs(
round(np.mean(start2_array[ii - mm - 1:ii])))
tncoordinates_array = np.vstack((tncoordinates_array, [
ref_tid_kk,
int(avg_start_pos),
int(flag2_array[ii - 1])
]))
readnumb_list.append(mm + 1)
tnnumber_dict[kk] = jj
del (jj, start_array, flag_array, readlength_array, flag2_array,
start2_array, ref_tid_kk)
timer_end = timeit.default_timer()
print('Chromosome %s completed in %.3f seconds' %
(kk, (timer_end - timer_start)))
print('')
readnumb_array = np.array(readnumb_list)
del readnumb_list
tncoordinatescopy_array = np.array(tncoordinates_array, copy=True)
#%% GET LIST OF ALL GENES AND ALL ESSENTIAL GENES
print('Getting coordinates of all genes ...')
# GET POSITION GENES
gff_path = os.path.join(files_path,
'Saccharomyces_cerevisiae.R64-1-1.99.gff3')
genecoordinates_dict = gene_position(
gff_path) #'YAL069W' | ['I', 335, 649], ...
# GET ALL ANNOTATED ESSENTIAL GENES
essential_path = os.path.join(files_path,
'Cerevisiae_AllEssentialGenes_List.txt')
essentialcoordinates_dict = {}
with open(essential_path, 'r') as f:
genes = f.readlines()[1:]
for gene in genes:
name = gene.strip('\n')
essentialcoordinates_dict[name] = genecoordinates_dict.get(
name).copy()
# GET ALIASES OF ALL GENES
names_path = os.path.join(files_path, 'Yeast_Protein_Names.txt')
aliases_designation_dict = gene_aliases(names_path)[
0] #'YMR056C' \ ['AAC1'], ...
del (gff_path, gene, genes, name, essential_path)
#%% CONCATENATE ALL CHROMOSOMES
#FOR EACH INSERTION LOCATION, ADD THE LENGTH OF ALL PREVIOUS CHROMOSOMES.
ll = 0
for ii in range(1, len(ref_name_list)):
ll += chr_length_dict[ref_name_list[ii - 1]]
aa = np.where(tncoordinatescopy_array[:, 0] == ii + 1)
tncoordinatescopy_array[aa, 1] = tncoordinatescopy_array[aa, 1] + ll
#FOR EACH GENE LOCATION, ADD THE LENGTH OF ALL PREVIOUS CHROMOSOMES.
for key in genecoordinates_dict:
gene_chrom = ref_tid_roman_dict.get(genecoordinates_dict.get(key)[0])
genecoordinates_dict[key][1] = genecoordinates_dict.get(
key)[1] + chr_summedlength_dict.get(gene_chrom)
genecoordinates_dict[key][2] = genecoordinates_dict.get(
key)[2] + chr_summedlength_dict.get(gene_chrom)
#FOR EACH ESSENTIAL GENE LOCATION, ADD THE LENGTH OF ALL PREVIOUS CHROMOSOMES.
for key in essentialcoordinates_dict:
gene_chrom = ref_tid_roman_dict.get(
essentialcoordinates_dict.get(key)[0])
essentialcoordinates_dict[key][1] = essentialcoordinates_dict.get(
key)[1] + chr_summedlength_dict.get(gene_chrom)
essentialcoordinates_dict[key][2] = essentialcoordinates_dict.get(
key)[2] + chr_summedlength_dict.get(gene_chrom)
del (ii, ll, aa, key, gene_chrom)
#%% GET NUMBER OF TRANSPOSONS AND READS PER GENE
print('Get number of insertions and reads per gene ...')
#ALL GENES
tnpergene_dict = {}
readpergene_dict = {}
tncoordinates_pergene_dict = {}
# readpergenecrude_dict = {}
for gene in genecoordinates_dict:
xx = np.where(
np.logical_and(
tncoordinatescopy_array[:, 1] >=
genecoordinates_dict.get(gene)[1],
tncoordinatescopy_array[:, 1] <= genecoordinates_dict.get(gene)
[2])) #get all insertions within range of current gene
tnpergene_dict[gene] = np.size(xx)
readpergene_dict[gene] = sum(readnumb_array[xx]) - max(
readnumb_array[xx],
default=0) #REMOVE LARGEST VALUE TO REDUCE NOISE
# readpergenecrude_dict[gene] = sum(readnumb_array[xx])
if np.size(xx) > 0:
tncoordinates_pergene_dict[gene] = [
genecoordinates_dict.get(gene)[0],
genecoordinates_dict.get(gene)[1],
genecoordinates_dict.get(gene)[2],
list(tncoordinatescopy_array[xx[0][0]:xx[0][-1] + 1, 1]),
list(readnumb_array[xx])
]
else:
tncoordinates_pergene_dict[gene] = [
genecoordinates_dict.get(gene)[0],
genecoordinates_dict.get(gene)[1],
genecoordinates_dict.get(gene)[2], [], []
]
#ONLY ESSENTIAL GENES
tnperessential_dict = {}
readperessential_dict = {}
tncoordinates_peressential_dict = {}
# readperessentialcrude_dict = {}
for gene in essentialcoordinates_dict:
xx = np.where(
np.logical_and(
tncoordinatescopy_array[:, 1] >=
essentialcoordinates_dict.get(gene)[1],
tncoordinatescopy_array[:, 1] <=
essentialcoordinates_dict.get(gene)[2]))
tnperessential_dict[gene] = np.size(xx)
readperessential_dict[gene] = sum(readnumb_array[xx]) - max(
readnumb_array[xx], default=0)
# readperessentialcrude_dict[gene] = sum(readnumb_array[xx])
if np.size(xx) > 0:
tncoordinates_peressential_dict[gene] = [
essentialcoordinates_dict.get(gene)[0],
essentialcoordinates_dict.get(gene)[1],
essentialcoordinates_dict.get(gene)[2],
list(tncoordinatescopy_array[xx[0][0]:xx[0][-1] + 1, 1]),
list(readnumb_array[xx])
]
else:
tncoordinates_peressential_dict[gene] = [
essentialcoordinates_dict.get(gene)[0],
essentialcoordinates_dict.get(gene)[1],
essentialcoordinates_dict.get(gene)[2], [], []
]
del (xx, gene)
#%% CREATE BED FILE
bedfile = bamfile + '.bed'
print('Writing bed file at: ', bedfile)
print('')
with open(bedfile, 'w') as f:
f.write('track name=' + filename + ' useScore=1\n')
coordinates_counter = 0
for tn in tncoordinates_array:
refname = [
key for key, val in ref_tid_dict.items() if val == tn[0] - 1
][0]
if refname == 'Mito':
refname = 'M'
f.write('chr' + refname + ' ' + str(tn[1]) + ' ' + str(tn[1] + 1) +
' . ' +
str(100 + readnumb_array[coordinates_counter] * 20) + '\n')
coordinates_counter += 1
del (bedfile, coordinates_counter, refname)
#%% CREATE TEXT FILE WITH TRANSPOSONS AND READS PER GENE
pergenefile = bamfile + '_pergene.txt'
print('Writing pergene.txt file at: ', pergenefile)
print('')
with open(pergenefile, 'w') as f:
f.write(
'Gene name\tNumber of transposons per gene\tNumber of reads per gene\n'
)
for gene in tnpergene_dict:
tnpergene = tnpergene_dict[gene]
readpergene = readpergene_dict[gene]
if gene in aliases_designation_dict:
gene_alias = aliases_designation_dict.get(gene)[0]
else:
gene_alias = gene
f.write(gene_alias + '\t' + str(tnpergene) + '\t' +
str(readpergene) + '\n')
del (pergenefile, gene, gene_alias, tnpergene, readpergene)
#%% CREATE TEXT FILE TRANSPOSONS AND READS PER ESSENTIAL GENE
peressentialfile = bamfile + '_peressential.txt'
print('Writing peressential.txt file at: ', peressentialfile)
print('')
with open(peressentialfile, 'w') as f:
f.write(
'Gene name\tNumber of transposons per gene\tNumber of reads per gene\n'
)
for essential in tnperessential_dict:
tnperessential = tnperessential_dict[essential]
readperessential = readperessential_dict[essential]
if essential in aliases_designation_dict:
essential_alias = aliases_designation_dict.get(essential)[0]
else:
essential_alias = essential
f.write(essential_alias + '\t' + str(tnperessential) + '\t' +
str(readperessential) + '\n')
del (peressentialfile, essential, essential_alias, tnperessential,
readperessential)
#%% CREATE TEXT FILE WITH LOCATION OF INSERTIONS AND READS PER GENE
pergeneinsertionsfile = bamfile + '_pergene_insertions.txt'
print('Witing pergene_insertions.txt file at: ', pergeneinsertionsfile)
print('')
with open(pergeneinsertionsfile, 'w') as f:
f.write(
'Gene name\tChromosome\tStart location\tEnd location\tInsertion locations\tReads per insertion location\n'
)
for gene in tncoordinates_pergene_dict:
gene_chrom = ref_tid_roman_dict.get(
genecoordinates_dict.get(gene)[0])
tncoordinates = [
ins - chr_summedlength_dict.get(gene_chrom)
for ins in tncoordinates_pergene_dict[gene][3]
]
if gene in aliases_designation_dict:
gene_alias = aliases_designation_dict.get(gene)[0]
else:
gene_alias = gene
f.write(gene_alias + '\t' +
str(tncoordinates_pergene_dict[gene][0]) + '\t' +
str(tncoordinates_pergene_dict[gene][1] -
chr_summedlength_dict.get(gene_chrom)) + '\t' +
str(tncoordinates_pergene_dict[gene][2] -
chr_summedlength_dict.get(gene_chrom)) + '\t' +
str(tncoordinates) + '\t' +
str(tncoordinates_pergene_dict[gene][4]) + '\n')
del (gene, gene_chrom, tncoordinates, gene_alias, pergeneinsertionsfile)
#%% CREATE TEXT FILE WITH LOCATION OF INSERTIONS AND READS PER ESSENTIAL GENE
peressentialinsertionsfile = bamfile + '_peressential_insertions.txt'
print('Writing peressential_insertions.txt file at: ',
peressentialinsertionsfile)
print('')
with open(peressentialinsertionsfile, 'w') as f:
f.write(
'Essential gene name\tChromosome\tStart location\tEnd location\tInsertion locations\tReads per insertion location\n'
)
for essential in tncoordinates_peressential_dict:
gene_chrom = ref_tid_roman_dict.get(
genecoordinates_dict.get(essential)[0])
tncoordinates = [
ins - chr_summedlength_dict.get(gene_chrom)
for ins in tncoordinates_peressential_dict[essential][3]
]
if essential in aliases_designation_dict:
essential_alias = aliases_designation_dict.get(essential)[0]
else:
essential_alias = essential
f.write(essential_alias + '\t' +
str(tncoordinates_peressential_dict[essential][0]) + '\t' +
str(tncoordinates_peressential_dict[essential][1] -
chr_summedlength_dict.get(gene_chrom)) + '\t' +
str(tncoordinates_peressential_dict[essential][2] -
chr_summedlength_dict.get(gene_chrom)) + '\t' +
str(tncoordinates) + '\t' +
str(tncoordinates_peressential_dict[essential][4]) + '\n')
del (essential, gene_chrom, tncoordinates, essential_alias,
peressentialinsertionsfile)
#%% ADD INSERTIONS AT SAME LOCATION BUT WITH DIFFERENT ORIENTATIONS TOGETHER (FOR STORING IN WIG-FILE)
wigfile = bamfile + '.wig'
print('Writing wig file at: ', wigfile)
print('')
readnumbwig_array = readnumb_array.copy()
unique_index_array = np.array([], dtype=int) #=cc
N_uniques_perchr_list = []
ll = 0
for kk in ref_name_list:
index = np.where(tncoordinates_array[:, 0] == int(
ref_tid_dict[kk] + 1)) #get indices for current chromosome.
unique_index = np.unique(
tncoordinates_array[index][:, 1], return_index=True
)[1] #get all insertion locations (in tncoordinates, all rows, column 1)
unique_index_array = np.append(unique_index_array, (unique_index + ll),
axis=0)
ll += np.count_nonzero(tncoordinates_array[:,
0] == int(ref_tid_dict[kk] +
1))
N_uniques_perchr_list.append(
ll) #total amount unique indices found untill current chromosome
del (ll, kk, unique_index)
duplicate_list = [] #=dd
ll = 0
index_last_unique_previous_chromosome = 0
for ii in N_uniques_perchr_list:
index_last_unique = np.where(unique_index_array <= ii)[0][-1]
for jj in range(ll, ii):
if int(jj) not in unique_index_array[
index_last_unique_previous_chromosome:index_last_unique]:
duplicate_list.append(jj)
index_last_unique_previous_chromosome = index_last_unique
ll = ii
#SUM READNUMB VALUES AT INDEX IN DUPLICATE_LIST AND DUPLICATE_LIST-1
for ii in duplicate_list:
readnumbwig_array[ii -
1] = readnumbwig_array[ii -
1] + readnumbwig_array[ii]
tncoordinateswig_duplicatesremoved_array = np.delete(tncoordinates_array,
duplicate_list,
axis=0)
readnumbwig_duplicatesremoved_array = np.delete(readnumbwig_array,
duplicate_list,
axis=0)
del (ll, ii, jj, N_uniques_perchr_list, index_last_unique, duplicate_list,
readnumbwig_array)
#%% CREATING WIG FILE
with open(wigfile, 'w') as f:
f.write('track type=wiggle_0 ,maxheightPixels=60 name=' + filename +
'\n')
for kk in ref_name_list:
f.write('VariableStep chrom=chr' + kk + '\n')
index = np.where(tncoordinateswig_duplicatesremoved_array[:, 0] ==
int(ref_tid_dict[kk] +
1)) #get indices for current chromosome.
for ii in index[0]:
f.write(
str(tncoordinateswig_duplicatesremoved_array[ii][1]) +
' ' + str(readnumbwig_duplicatesremoved_array[ii]) + '\n')
del (wigfile, kk, ii, index)
# early-stopping parameters
patience = 10000 # look as this many examples regardless
patience_increase = 2 # wait this much longer when a new best is
# found
improvement_threshold = 0.995 # a relative improvement of this much is
# considered significant
validation_frequency = min(n_train_batches, patience / 2)
# go through this many
# minibatche before checking the network
# on the validation set; in this case we
# check every epoch
best_validation_loss = numpy.inf
best_iter = 0
test_score = 0.
start_time = timeit.default_timer()
epoch = 0
done_looping = False
while (epoch < n_epochs) and (not done_looping):
epoch = epoch + 1
for minibatch_index in xrange(n_train_batches):
iter = (epoch - 1) * n_train_batches + minibatch_index
if iter % 100 == 0:
print 'training @ iter = ', iter
cost_ij = train_model(minibatch_index)
if (iter + 1) % validation_frequency == 0:
def main():
time_start = timeit.default_timer()
#global PATH
df = pd.read_csv("/home/dummy/try/SesOutFinalUS.tsv",
dtype=object,
header=None,
delimiter="\t",
error_bad_lines=False)
#df1 = df.head(200)
#df1.to_csv(PATH + "/outputs/sampleOut1.tsv", index = False, header= None, sep='\t')
print 'file read'
#['U','SES', 'DUR', 'SC','C','URL','D','T']
df.columns = [0, 1, 2, 8, 9, 3, 4, 5]
users = {}
curr_user = df.iloc[0, 0]
curr_session = df.iloc[0, 1]
url_list = []
session_list = []
tier1 = 0
tier2 = 0
tier3 = 0
tier4 = 0
url_book = 'https://secure.celebritycruises.com/booking/paymentConfirmation'
url_held = 'https://secure.celebritycruises.com/booking/courtesyHoldConfirmation'
visits = [[0 for i in range(5)] for j in range(4)] #lists of lists
duration_list = [[0 for i in range(5)] for j in range(4)]
day_diff = [[0 for i in range(4)] for j in range(4)]
div_dur = [[0 for i in range(5)] for j in range(4)]
duration = get_sec(df.loc[0, 2])
temp = True
count = 0
c = 0
session_f = [0, 0, 0, 0]
sess_duration = []
for i in range(1, len(df.index)):
if df[0][i] == curr_user:
if df[3][i] not in url_list:
url_list.append(df[3][i])
else:
count += 1
if url_book in url_list or url_held in url_list:
users[curr_user] = Node(3, df[1][i - 1]) #2 is index of tier4
tiers = 3
else:
if int(df[1][i - 1]) == 1 and len(set(url_list)) in [0, 1]:
tiers = 0
c += 1
elif int(df[1][i - 1]) == 1 and len(set(url_list)) >= 2:
tiers = 1 #index for tier2
else:
tiers = 2 #index for tier3
users[curr_user] = Node(tiers, df[1][i - 1])
session_f[tiers] += int(df[1][i - 1])
curr_user = df[0][i]
del url_list[:]
url_list.append(df[3][i])
print 'Total number of Users are ' + str(count)
fivePlus = [0, 0, 0, 0]
foTofi = [0, 0, 0, 0]
for node in users.values():
if node.sessions >= 5:
index = 4
fivePlus[node.tier] += (node.sessions - 4)
foTofi[node.tier] += 1
else:
index = node.sessions - 1
if node.tier == 0:
tier1 += 1
visits[0][index] += 1
elif node.tier == 1:
tier2 += 1
visits[1][index] += 1
elif node.tier == 2:
tier3 += 1
visits[2][index] += 1
else:
tier4 += 1
visits[3][index] += 1
print fivePlus
for i in range(4):
for j in range(5):
div_dur[i][j] = doSum(i, j, visits)
curr_session = df[1][0]
temp = int(curr_session) - 1
curr_user = df[0][0]
usr_page = df[3][0]
r = users[curr_user].tier
duration_list[r][int(curr_session)] += get_sec(df[2][0])
for index, user in enumerate(df[0][1:len(df.index)], start=1):
if user != curr_user:
duration_list[users[curr_user].tier][temp] += get_sec(df[2][index -
1])
curr_user = user
curr_session = df[1][index]
temp = getTemp(int(curr_session))
#usr_page = df[3][index]
elif df[1][index] != curr_session:
if temp != 4:
sec = get_sec(df[5][index].strip(' GMT')) - get_sec(
df[5][index - 1].strip(' GMT'))
l1 = map(int, df.iloc[index, 6].split('-'))
l2 = map(int, df.iloc[index - 1, 6].split('-'))
d1 = date(l1[0], l1[1], l1[2])
d2 = date(l2[0], l2[1], l2[2])
d = (d1 - d2).days
if sec < 0:
d = d - 1
day_diff[users[curr_user].tier][temp] += d
duration_list[users[curr_user].tier][temp] += get_sec(df[2][index -
1])
curr_session = df[1][index]
temp = getTemp(int(curr_session))
else:
pass
for i in range(4):
for j in range(5):
if div_dur[i][j] == 0:
duration_list[i][j] = 0
else:
if j == 4 and fivePlus[i] != 0:
temp = fivePlus[i]
else:
temp = div_dur[i][j]
duration_list[i][j] = str(
datetime.timedelta(seconds=int(duration_list[i][j] /
temp)))
for i in range(4):
for j in range(4):
if div_dur[i][j] == 0:
day_diff[i][j] = 0
else:
if j == 3 and foTofi[i] != 0:
temp = foTofi[i]
else:
temp = div_dur[i][j]
day_diff[i][j] = day_diff[i][j] / temp
print tier1, tier2, tier3, tier4
print '\n'
print visits[0]
print visits[1]
print visits[2]
print visits[3]
print '\n'
print div_dur[0]
print div_dur[1]
print div_dur[2]
print div_dur[3]
print '\n'
print session_f[0]
print session_f[1]
print session_f[2]
print session_f[3]
print '\n'
print duration_list[0]
print duration_list[1]
print duration_list[2]
print duration_list[3]
print '\n'
print day_diff[0]
print day_diff[1]
print day_diff[2]
print day_diff[3]
print '\n'
print tier1 * 100 / (tier1 + tier2 + tier3 + tier4)
time_stop = timeit.default_timer()
print time_stop - time_start
def tearDown(self):
'''after each test function'''
pass
def do(self, func):
'''todo'''
self.assertEqual(func("hello", "ll"), 2)
self.assertEqual(func("aaaaa", "bba"), -1)
self.assertEqual(func("aaaac", "aac"), 2)
pass
def test_func(self):
self.do(s.strStr)
self.assertEqual(s.make_prefix("abc"), [-1, 0, 0])
self.assertEqual(s.make_prefix("aac"), [-1, 0, 1])
self.assertEqual(s.make_prefix("ababc"), [-1, 0, 0, 1, 2])
if __name__ == "__main__":
count = 100000
t = "ababcabcde"
p = "abcd"
utils.print_func_run_time(count, s.strStr, t = t, p = p)
b = timeit.default_timer()
for i in range(count):
t.index(p)
print(timeit.default_timer() - b)
unittest.main()
_save_log_ = False
if _save_log_:
from datetime import datetime
from std_logger import StdFileLoggerCtrl
# save all console activity to out_log_file
out_log_file = os.path.join(
r'P:\Synchronize\python_script_logs\\%s_log_%s.log' %
(os.path.basename(__file__),
datetime.now().strftime('%Y%m%d%H%M%S')))
log_link = StdFileLoggerCtrl(out_log_file)
print('#### Started on %s ####\n' % time.asctime())
START = timeit.default_timer()
#==========================================================================
# When in post_mortem:
# 1. "where" to show the stack
# 2. "up" move the stack up to an older frame
# 3. "down" move the stack down to a newer frame
# 4. "interact" start an interactive interpreter
#==========================================================================
if DEBUG_FLAG:
try:
main()
except:
import pdb
performance = get_current_performance(
np.zeros(int(num_examples / update_interval)), 0)
# set firing rates to zero initially
for name in input_population_names:
input_groups[name + 'e'].rate = 0
# initialize network
j = 0
num_retries = 0
b.run(0)
weights_name = 'XeAe' + '_' + ending
# start recording time
start_time = timeit.default_timer()
while j < num_examples:
# fetched rates depend on training / test phase, and whether we use the
# testing dataset for the test phase
if test_mode:
if use_testing_set:
rates = testing['x'][j % 10000, :, :] / 8. * input_intensity
else:
rates = training['x'][j % 60000, :, :] / 8. * input_intensity
else:
# ensure weights don't grow without bound
normalize_weights()
# get the firing rates of the next input example
rates = training['x'][j % 60000, :, :] / 8. * input_intensity
from utils import *
cv2.setNumThreads(0)
cv2.ocl.setUseOpenCL(False)
test_dir = 'test/images'
pred_folder = 'pred154_loc'
models_folder = 'weights'
all_files = np.array(get_files())
val_idxs = train_test_split(np.arange(len(all_files)).astype(int), test_size=0.1, random_state=0)[1]
all_files= all_files[val_idxs]
if __name__ == '__main__':
t0 = timeit.default_timer()
makedirs(pred_folder, exist_ok=True)
os.environ['CUDA_DEVICE_ORDER'] = 'PCI_BUS_ID'
os.environ["CUDA_VISIBLE_DEVICES"] = sys.argv[1]
# cudnn.benchmark = True
models = []
for seed in [0]:
snap_to_load = 'se154_loc_{}_1_best'.format(seed)
model = SeNet154_Unet_Loc().cuda()
model = nn.DataParallel(model).cuda()
print("=> loading checkpoint '{}'".format(snap_to_load))
x['id6'] = x['id6'].astype('category')
small = fread(src_jn_y[0]).to_pandas()
small['id4'] = small['id4'].astype('category')
medium = fread(src_jn_y[1]).to_pandas()
medium['id4'] = medium['id4'].astype('category')
medium['id5'] = medium['id5'].astype('category')
big = fread(src_jn_y[2]).to_pandas()
big['id4'] = big['id4'].astype('category')
big['id5'] = big['id5'].astype('category')
big['id6'] = big['id6'].astype('category')
print(len(x.index), flush=True)
print(len(small.index), flush=True)
print(len(medium.index), flush=True)
print(len(big.index), flush=True)
task_init = timeit.default_timer()
print("joining...", flush=True)
question = "small inner on int" # q1
gc.collect()
t_start = timeit.default_timer()
ans = x.merge(small, on='id1')
print(ans.shape, flush=True)
t = timeit.default_timer() - t_start
m = memory_usage()
t_start = timeit.default_timer()
chk = [ans['v1'].sum(), ans['v2'].sum()]
chkt = timeit.default_timer() - t_start
write_log(task=task, data=data_name, in_rows=x.shape[0], question=question, out_rows=ans.shape[0], out_cols=ans.shape[1], solution=solution, version=ver, git=git, fun=fun, run=1, time_sec=t, mem_gb=m, cache=cache, chk=make_chk(chk), chk_time_sec=chkt, on_disk=on_disk)
del ans
gc.collect()
def main(arguments,output_filename):
''' Parse user input, query SQL database, generate pandas dataframes, export JSON for D3 and print HTML code '''
######################################################
### start the alert div that contains any output generated here
######################################################
algorithm_output_str = ''
timing = {}
start_all = timeit.default_timer()
######################################################
# generate an input overview table
######################################################
arg_names = ['Genes','Cluster by','Color by','Interaction type',\
'Minimal number of experiments','Minimal number of publications', 'Minimal number of methods','Method types',\
'Process','Compartment','Expression','Max. number nodes','Filter condition']
input_dict = { arg_names[i]:arguments[i].replace("_"," ") for i in range(len(arg_names)) } # does not include unique_str and excel_flag
input_dict['Expression'] = input_dict['Expression'].replace('G1P','G1(P)') # brackets removed in PHP
df_user_input = pd.DataFrame.from_dict(input_dict,orient='index')
df_user_input = df_user_input.reindex(index = arg_names)
df_user_input.columns = ['user input']
df_user_input_to_print = df_user_input.to_html(classes=['table','table-condensed','table-bordered'])
### process arguments
primary_nodes,cluster_by,color_by,int_type,\
min_exp,min_pub,min_methods,method_types,\
process,compartment,expression,\
max_nodes,filter_condition,\
excel_flag,filter_flag,unique_str = arguments
# make sure types are correct
color_by = color_by.replace('_',' ')
cluster_by = cluster_by.replace('_',' ')
filter_condition = filter_condition.replace('_',' ')
process = process.split(',')
method_types = method_types.split(',')
method_types = [x.replace('_',' ') for x in method_types]
expression = expression.split(',')
if 'G1P' in expression: # brackets removed in php
ind = expression.index('G1P')
expression[ind] = 'G1(P)'
process = [x.replace("_"," ") for x in process]
primary_nodes_str = primary_nodes
if '_' in primary_nodes:
primary_nodes = primary_nodes.split('_')
else:
primary_nodes = [primary_nodes]
min_exp = int(min_exp)
min_pub = int(min_pub)
min_methods = int(min_methods)
max_nodes = int(max_nodes)
excel_flag = bool(int(excel_flag))
filter_flag = bool(int(filter_flag))
split_types = int_type.split(',')
compartment = compartment.replace('_',' ')
timing['input'] = timeit.default_timer() - start_all
if excel_flag:
######################################################
# WRITE TO EXCEL
######################################################
# THIS HAS TO HAPPEN BEFORE HTML REPLACEMENTS
start_excel = timeit.default_timer()
write_excel_file(primary_nodes_str+'_'+unique_str)
timing['excel'] = timeit.default_timer() - start_excel
print(timing)
return
######################################################
### get all interactions related to the input IDs
######################################################
start_initial = timeit.default_timer()
database = SCRIPT_DIR+"/data/DB_genes_and_interactions.db"
conn = create_connection(database)
# get all interactions in which the given genes takes part
placeholders = ', '.join('?' for unused in primary_nodes) # '?, ?, ?, ...'
# The query differs based on whether we need to subselect on the 'type' of interaction
if len(split_types) == 3:
query = "SELECT source,target FROM interactions WHERE ( (source IN (%s) or target IN (%s)) and num_experiments >= (%s) \
and num_publications >= (%s) and num_methods >= (%s))" % (placeholders,placeholders,min_exp,min_pub,min_methods)
cursor = conn.execute(query,primary_nodes+primary_nodes)
else:
placeholders_type = ', '.join('?' for unused in split_types)
query = "SELECT source,target FROM interactions WHERE ( (source IN (%s) or target IN (%s)) AND type IN (%s) \
AND num_experiments >= (%s) and num_publications >= (%s) and num_methods >= (%s))" % (placeholders,placeholders, \
placeholders_type,min_exp,min_pub,min_methods)
cursor = conn.execute(query,primary_nodes+primary_nodes+split_types)
# construct dataframe of interacting genes: the nodes
node_list = list(set([x for y in cursor for x in y])) # get rid of duplicates of which there will be many
if len(node_list) == 0:
raise ValueError('No interactions matching these conditions.')
# get the info from the database for each node to make the 'nodes' dataframe
if 'No_data' in expression:
query = """SELECT standard_name,systematic_name,name_desc,desc,go_term_1,go_term_2,\
GFP_abundance,GFP_localization,CYCLoPs_Excel_string,CYCLoPs_html,expression_peak_phase,\
expression_peak_time,CYCLoPs_dict FROM genes \
WHERE standard_name in (%s) AND (standard_name in (%s) OR expression_peak_phase in (%s) OR expression_peak_phase is NULL) AND (standard_name in (%s) OR go_term_1 in (%s) OR go_term_2 in (%s))""" \
% (', '.join('?' for _ in node_list), ', '.join('?' for _ in primary_nodes), ', '.join('?' for _ in expression), ', '.join('?' for _ in primary_nodes),', '.join('?' for _ in process),', '.join('?' for _ in process))
else:
query = """SELECT standard_name,systematic_name,name_desc,desc,go_term_1,go_term_2,\
GFP_abundance,GFP_localization,CYCLoPs_Excel_string,CYCLoPs_html,expression_peak_phase,\
expression_peak_time,CYCLoPs_dict FROM genes \
WHERE standard_name in (%s) AND (standard_name in (%s) OR expression_peak_phase in (%s)) AND (standard_name in (%s) OR go_term_1 in (%s) OR go_term_2 in (%s))""" \
% (', '.join('?' for _ in node_list), ', '.join('?' for _ in primary_nodes), ', '.join('?' for _ in expression), ', '.join('?' for _ in primary_nodes), ', '.join('?' for _ in process),', '.join('?' for _ in process))
cursor = conn.execute(query,node_list+primary_nodes+expression+primary_nodes+process+process)
data = [list(l) for l in cursor] # cursor itself is a generator, this is a list of lists
nodes = pd.DataFrame(data,columns=['Standard name','Systematic name','Name description','Description',
'GO term 1','GO term 2','GFP abundance','GFP localization','CYCLoPs_Excel_string',
'CYCLoPs_html','Expression peak phase','Expression peak time','CYCLoPs_dict'])
timing['Get node information from database'] = timeit.default_timer() - start_initial
### make actual dictionaries out of CYCLoPs_dict column
nodes['CYCLoPs_dict'] = nodes['CYCLoPs_dict'].apply(ast.literal_eval)
len_nodes_query = len(nodes)
######################################################
### BASED ON THE COMPARTMENT FILTER: DROP NODES
######################################################
start_node_drop = timeit.default_timer()
if 'GFP:' in compartment:
comp_to_check = compartment.replace('GFP:','')
print('Prior to compartment filtering:', len(nodes), 'nodes. Filtering on', comp_to_check)
s = pd.Series([comp_to_check in x for x in nodes['GFP localization'].str.split(', ')])
nodes = nodes[s.values]
nodes = nodes.reset_index(drop=True)
print('After compartment filtering:', len(nodes), 'nodes.')
elif 'CYCLoPs:' in compartment:
comp_to_check = compartment.replace('CYCLoPs:','')
print('Prior to compartment filtering:', len(nodes), 'nodes. Filtering on', comp_to_check)
l_o_l = [[list(nodes.iloc[i]['CYCLoPs_dict'][x].keys()) for x in list(nodes.iloc[i]['CYCLoPs_dict'].keys()) ] for i in range(len(nodes)) ]
s = pd.Series([comp_to_check in [v for WT in l_o_l[i] for v in WT] for i in range(len(l_o_l))])
nodes = nodes[s.values]
nodes = nodes.reset_index(drop=True)
print('After compartment filtering:', len(nodes), 'nodes.')
else: #it is 'Any'
pass
### Combine the expression columns
nodes['Expression peak'] = nodes['Expression peak phase'] + " (" + nodes['Expression peak time'].map(str) + " min)"
nodes['Expression peak'] = nodes['Expression peak'].mask(nodes['Expression peak'].isnull(), "No data")
# alphabetize
nodes = nodes.sort_values(by='Standard name',ascending=True)
nodes = nodes.reset_index(drop=True)
node_list = list(nodes['Standard name'].values)
nodes['primary node'] = [x in primary_nodes for x in nodes['Standard name']]
if len(nodes) == 0:
raise ValueError("Filtering left no nodes.")
timing['Node filter: compartment'] = timeit.default_timer() - start_node_drop
######################################################
# Clustering and coloring
######################################################
start = timeit.default_timer()
### Clustering part
if cluster_by in ['GO term 1','GO term 2']:
nodes['cluster'] = nodes[cluster_by]
elif 'CYCLoPs WT' in cluster_by:
WT_string = 'WT' + cluster_by[-1]
# loop over all nodes find their highest expression compartment for the WT given by WT_string
# NOTE: SOMETIMES A DICTIONARY WITH EXPRESSION DATA FOR A GIVEN WT IS EMPTY WE NEED TO CHECK FOR THIS
# Example: GET1 in WT1
l = nodes['CYCLoPs_dict'].values
l_max_comps = [ max(l[i][WT_string], key=lambda key: l[i][WT_string][key]) if (type(l[i]) != str and len(l[i][WT_string]) > 0) else 'No data' for i in range(len(nodes))]
nodes['cluster'] = pd.Series(l_max_comps).values
elif cluster_by == 'No clustering':
nodes['cluster'] = ['No clustering' for i in range(len(nodes))]
else:
raise SystemExit(cluster_by,f"Unexpected value for clustering variable: {cluster_by}.")
if color_by in ['GO term 1','GO term 2']:
# set the color based on the color_by variable in a new column of 'nodes' DF
nodes['color'] = nodes[color_by]
elif 'CYCLoPs WT' in color_by:
WT_string = 'WT' + color_by[-1]
# loop over all nodes find their highest expression compartment for the WT given by WT_string
# NOTE: SOMETIMES A DICTIONARY WITH EXPRESSION DATA FOR A GIVEN WT IS EMPTY WE NEED TO CHECK FOR THIS
# Example: GET1 in WT1
l = nodes['CYCLoPs_dict'].values
l_max_comps = [ max(l[i][WT_string], key=lambda key: l[i][WT_string][key]) if \
(type(l[i]) != str and len(l[i][WT_string]) > 0) else 'No data' for i in range(len(nodes))]
# set the color based on the maximum compartment found above in a new column in the nodes DF
nodes['color'] = pd.Series(l_max_comps).values
elif color_by == "Peak expression phase":
nodes['color'] = nodes['Expression peak phase']
elif color_by == 'No coloring':
nodes['color'] = ["No data" for i in range(len(nodes))]
else:
raise SystemExit(color_by, f'Unexpected value for coloring variable: {color_by}')
# now we can drop expression peak phase/time as separate fields
nodes = nodes.drop('Expression peak phase',1)
nodes = nodes.drop('Expression peak time',1)
timing['Setting node cluster and color attributes'] = timeit.default_timer() - start
len_nodes_filtered_comp = len(nodes)
######################################################
### GET ALL INTERACTIONS BETWEEN ALL NODES
######################################################
start_final_sql = timeit.default_timer()
max_interactions = 10000 # a too high value here seems to make the server run out of memory and this is the most time-expensive step on the server
placeholders = ', '.join('?' for unused in node_list) # '?, ?, ?, ...'
placeholders_primary_nodes = ', '.join('?' for unused in primary_nodes)
# Multiple query options
# if there are more than max_interactions satisfying the criteria then ORDEr BY:
# - Pick interactions with primary_nodes first
# - pick regulations/physical over genetic
# - pick more over less: exp, pubs, methods
# - Pick regulation over physical when equal in exp/pubs/methods because regulatory interactions are often singular in these.
if len(split_types) == 3:
query = "SELECT * FROM interactions \
WHERE ( (source IN (%s) AND target IN (%s)) \
AND num_experiments >= (%s) AND num_publications >= (%s) AND num_methods >= (%s)) \
ORDER BY \
CASE WHEN ((source IN (%s)) OR (target IN (%s))) THEN 1 ELSE 2 END ASC, \
CASE type WHEN 'physical' OR 'regulation' THEN 1 WHEN 'genetic' THEN 2 END ASC, \
num_experiments DESC, num_publications DESC, num_methods DESC, \
CASE type WHEN 'regulation' THEN 1 WHEN 'physical' THEN 2 WHEN 'genetic' THEN 3 END ASC \
limit (%s)" \
% (placeholders,placeholders,min_exp,min_pub,min_methods,placeholders_primary_nodes,placeholders_primary_nodes,max_interactions)
interactome = pd.read_sql_query(query, conn, params=node_list+node_list+primary_nodes+primary_nodes)
else:
placeholders_type = ', '.join('?' for unused in split_types)
query = "SELECT * FROM interactions \
WHERE ( (source IN (%s) AND target IN (%s)) AND type IN (%s) \
AND num_experiments >= (%s) and num_publications >= (%s) and num_methods >= (%s)) \
ORDER BY \
CASE WHEN ((source IN (%s)) OR (target IN (%s))) THEN 1 ELSE 2 END ASC, \
CASE type WHEN 'physical' OR 'regulation' THEN 1 WHEN 'genetic' THEN 2 END ASC, \
num_experiments DESC, num_publications DESC, num_methods DESC, \
CASE type WHEN 'regulation' THEN 1 WHEN 'physical' THEN 2 WHEN 'genetic' THEN 3 END ASC \
limit (%s)" \
% (placeholders, placeholders,placeholders_type,min_exp,min_pub,min_methods,placeholders_primary_nodes,placeholders_primary_nodes,max_interactions)
interactome = pd.read_sql_query(query, conn, params=node_list+node_list+split_types+primary_nodes+primary_nodes)
interactome.columns = ['source','target','type','Evidence','Evidence HTML','#Experiments',\
'#Publications','#Methods']
timing['Interactome SQL + dataframe + processing'] = timeit.default_timer() - start_final_sql
######################################################
### BASED ON THE METHOD TYPES FILTER: DROP INTERACTIONS
######################################################
start = timeit.default_timer()
to_drop = []
with open(SCRIPT_DIR+'/data/unique_experimental_methods.txt') as f:
read_methods = f.read().splitlines()
total_methods = len(read_methods)
if len(method_types) < total_methods: # some have been deselected
algorithm_output_str += '<p>' + 'We have on file: ' + str(total_methods) + ' methods. User queried for: ' + str(len(method_types)) + '</p>'
len_before = len(interactome)
interactome = interactome[interactome.apply(lambda x: find_methods_in_evidence(x['Evidence'],method_types),1)]
algorithm_output_str += '<p>' + 'We dropped: ' + str(len_before - len(interactome)) + ' interactions based on the methods.' + '</p>'
if len(interactome) == 0:
raise ValueError('No interactions matching these conditions.')
timing['Filter based on methods'] = timeit.default_timer() - start
######################################################
# Network properties with networkx: 1
######################################################
start = timeit.default_timer()
df_network = pd.Series()
df_network['Number of nodes'] = len(nodes)
df_network['Number of edges'] = len(interactome)
# use networkx
nodes, interactome, df_network, G = calc_network_props(primary_nodes, nodes, interactome, df_network, filter_condition)
df_network = df_network.to_frame()
df_network = df_network.transpose()
timing['networkx properties calculation'] = timeit.default_timer() - start
######################################################
# Export visualized networkx graph to graph formats (GEFX)
######################################################
start = timeit.default_timer()
nx.write_gexf(G, SCRIPT_DIR+'/../output/networkx/' + primary_nodes_str + "_" + unique_str + "_full.gexf")
timing['networkx export'] = timeit.default_timer() - start
######################################################
# Save the full network information
######################################################
start = timeit.default_timer()
nodes_full = nodes.copy()
interactome_full = interactome.copy()
timing['Save full network'] = timeit.default_timer() - start
######################################################
# Pickle the dataframes
######################################################
start = timeit.default_timer()
filename_base = os.path.abspath(SCRIPT_DIR+'/../output/excel_files/')
file_id = primary_nodes_str+'_'+unique_str
df_user_input.to_pickle(filename_base+'/user_input_'+file_id)
nodes_full.to_pickle(filename_base+'/nodes_'+file_id)
interactome_full.to_pickle(filename_base+'/interactome_'+file_id)
timing['Pickle full network'] = timeit.default_timer() - start
# ######################################################
# # WRITE "FULL" NETWORK TO JSON
# # this will include a filtering step for really big networks
# ######################################################
start_json = timeit.default_timer()
write_network_to_json(nodes_full,interactome_full,filter_condition,output_filename,G,'full',primary_nodes)
timing['json_full'] = timeit.default_timer() - start_json
######################################################
# FILTER NODES TO MANAGEABLE VISUALIZATION
if (filter_flag):
start_filter = timeit.default_timer()
len_interactome = len(interactome)
# reduce nodes
nodes = nodes.sort_values(by=['primary node',filter_condition],ascending=False)
nodes = nodes.iloc[:max_nodes]
nodes.reset_index(drop=True,inplace=True)
# reduce interactions
n = nodes['Standard name'].values # list of remaining node IDs
interactome = interactome[ (interactome['source'].isin(n)) & (interactome['target'].isin(n)) ]
interactome.reset_index(drop=True,inplace=True)
# SHOW WARNING MESSAGE ABOUT FILTER STEP
filter_message = "Note: this query returned {} nodes and {} interactions. We reduced the network to {} nodes based on {} resulting in {} interactions. \
All interactions and nodes are contained in the <i>full</i> Excel file. ".format(len_nodes_filtered_comp,len_interactome,max_nodes,filter_condition,len(interactome))
s = filter_message
print("<script>create_alert(\""+s+"\",\"alert-warning\");</script>")
timing['filter'] = timeit.default_timer() - start_filter
######################################################
# Network properties with networkx: 2
######################################################
start = timeit.default_timer()
# df_network = pd.Series()
df_network['Number of nodes'] = len(nodes)
df_network['Number of edges'] = len(interactome)
# use networkx
nodes, interactome, df_network, G = calc_network_props(primary_nodes, nodes, interactome, df_network, filter_condition)
timing['networkx properties calculation'] += timeit.default_timer() - start
######################################################
# Export full networkx graph to graph formats (GEFX)
######################################################
start = timeit.default_timer()
nx.write_gexf(G, SCRIPT_DIR+'/../output/networkx/' + primary_nodes_str + "_" + unique_str + ".gexf")
timing['networkx export'] += timeit.default_timer() - start
######################################################
# Nxviz image generation: matrixplot
######################################################
start = timeit.default_timer()
c = nv.MatrixPlot(G)
c.draw()
plt.savefig(SCRIPT_DIR+'/../output/nxviz/matrix_' + unique_str + '.png')
timing['nxviz matrix plot'] = timeit.default_timer() - start
######################################################
### Write the network to json
######################################################
start_json = timeit.default_timer()
write_network_to_json(nodes,interactome,filter_condition,output_filename,G)
timing['json'] = timeit.default_timer() - start_json
# remove the Evidence HTML column
interactome = interactome.drop('Evidence',1)
interactome = interactome.rename(columns={'Evidence HTML':'Evidence'})
if not excel_flag:
######################################################
### End output text alert div
######################################################
print("</div>")
######################################################
# Generate strings for the nodes and interactome dataframes to print
######################################################
start_print = timeit.default_timer()
# drop columns
nodes = nodes.drop(['Description','CYCLoPs_Excel_string','CYCLoPs_dict','cluster','color'],1)
# Add HTML links to database/SGD to symbols
nodes['Standard name'] = nodes['Standard name'].apply(lambda x: "<a href='index.php?id=database&gene=" + x + "' target='blank'>" + x + "</a>")
# change CYCLoPs column name and export html
# escape makes the HTML links work
nodes = nodes.rename(columns={'CYCLoPs_html':'CYCLoPs'})
nodes = nodes.to_html(escape=False,index=False,classes=['table','table-condensed','table-bordered'])
nodes = nodes.replace('<table','<table id=\"proteins_table\"',1)
interactome['source'] = interactome['source'].apply(lambda x: "<a href='index.php?id=database&gene=" + x + "' target='blank'>" + x + "</a>" )
interactome['target'] = interactome['target'].apply(lambda x: "<a href='index.php?id=database&gene=" + x + "' target='blank'>" + x + "</a>")
# escape makes the HTML links work
interactome = interactome.to_html(escape=False,index=False,classes=['table','table-condensed','table-bordered'])
interactome = interactome.replace('<table','<table id=\"interactions_table\"',1)
######################################################
# PRINT COLLAPSABLE BOOTSTRAP HTML CODE WITH THE DATAFRAMES
######################################################
# the 'in' class makes the collapse open by default: the interactions here
print("""
<div class="panel-group" id="accordion">
<div class="panel panel-default">
<div class="panel-heading">
<h4 class="panel-title">
<a data-toggle="collapse" data-parent="#accordion" href="#collapse1">
User input</a>
</h4>
</div>
<div id="collapse1" class="panel-collapse collapse">
<div class="panel-body">
<div class="table-responsive">
""")
print(df_user_input_to_print)
print("""
</div>
</div>
</div>
</div>
<div class="panel panel-default">
<div class="panel-heading">
<h4 class="panel-title">
<a data-toggle="collapse" data-parent="#accordion" href="#collapse2">
Network properties</a>
</h4>
</div>
<div id="collapse2" class="panel-collapse collapse">
<div class="panel-body">
<div class="table-responsive">
""")
print(df_network.to_html(classes=['table','table-condensed','table-bordered'],index=False))
print("""
</div>
</div>
</div>
</div>
<div class="panel panel-default">
<div class="panel-heading">
<h4 class="panel-title">
<a data-toggle="collapse" data-parent="#accordion" href="#collapse3">
Network nodes (proteins)</a>
</h4>
</div>
<div id="collapse3" class="panel-collapse collapse">
<div class="panel-body">
Use the search utility to find the gene you are looking for. The table scrolls horizontally and vertically.
By clicking the column headers the table will be sorted on that column. Use shift+click to sort on multiple columns.
Default sorting is on number of experiments, number of publications, number of methods and alphabetical on standard name, in that order.
<div class="table-responsive">
""")
print(nodes)
print("""
</div>
</div>
</div>
</div>
<div class="panel panel-default">
<div class="panel-heading">
<h4 class="panel-title">
<a data-toggle="collapse" data-parent="#accordion" href="#collapse4">
Interactions</a>
</h4>
</div>
<div id="collapse4" class="panel-collapse collapse">
<div class="panel-body">
Use the search utility to find the gene you are looking for.
By clicking the column headers the table will be sorted on that column. Use shift+click to sort on multiple columns.
Default sorting is on number of experiments, number of publications, number of methods and alphabetical on standard name, in that order.
<div class="table-responsive">
""")
print(interactome)
print("""
</div>
</div>
</div>
</div>
""")
######################################################
# Optional diagnostics
######################################################
print("""
<div class="panel panel-default">
<div class="panel-heading">
<h4 class="panel-title">
<a data-toggle="collapse" data-parent="#accordion" href="#collapse5">
Diagnostics: calculation time</a>
</h4>
</div>
<div id="collapse5" class="panel-collapse collapse">
<div class="panel-body">
<div class="table-responsive">
""")
timing['print frames'] = timeit.default_timer() - start_print
timing['all'] = timeit.default_timer() - start_all
df_timing = pd.Series(timing)
df_timing = df_timing.to_frame()
df_timing.columns = ['Time']
df_timing['Percentage'] = [v/timing['all']*100 for v in df_timing['Time'] ]
print(df_timing.sort_values('Percentage').to_html(classes=['table','table-condensed','table-bordered']))
print("Accounted for:", sum([timing[k] for k in timing if k != 'all' ])/timing['all'] * 100, "percent of the time spent in Python.")
print("""
</div>
</div>
</div>
</div>
</div>
""")
######################################################
# Show algorithm output in an alert at the bottom of the page
######################################################
if algorithm_output_str != '':
print("<div class=\"alert alert-dismissable alert-info\">")
print(algorithm_output_str)
print("</div>")
"""Step3. Save all the results"""
if not os.path.exists(args.outputDir):
os.makedirs(args.outputDir)
pvalues = {}
pvalues['Gpvals'] = gpvals.tolist()
pvalues['clu_pvals'] = clu_pvals.tolist()
pvalues['Lpvals_fdr'] = lpvals_fdr.tolist()
with open(os.path.join(args.outputDir, 'pvalues.json'), 'w') as outfile:
json.dump(pvalues, outfile)
efit = {}
efit['efitBetas'] = efit_beta.tolist()
efit['efitYdesign'] = efity_design.tolist()
efit['efitEtas'] = efit_eta.tolist()
with open(os.path.join(args.outputDir, 'efit.json'), 'w') as outfile:
json.dump(efit, outfile)
if __name__ == '__main__':
args = parser.parse_args()
start_all = timeit.default_timer()
run_script(args)
stop_all = timeit.default_timer()
delta_time_all = str(stop_all - start_all)
print("The total elapsed time is " + delta_time_all)
def multiply_by_Z_viaMKL( self, x ):
'''Multiplies the vector passed as argument by the matrix Z'''
code = 'multiply_by_Z_viaMKL'
start = default_timer()
# Dissecting the "cspblas_dcsrgemv" name:
# "c" - for "c-blas" like interface (as opposed to fortran)
# Also means expects sparse arrays to use 0-based indexing, which python does
# "sp" for sparse
# "d" for double-precision
# "csr" for compressed row format
# "ge" for "general", e.g., the matrix has no special structure such as symmetry
# "mv" for "matrix-vector" multiply
A = self.data_obj.Z
if not sparse.isspmatrix_csr(A):
raise Exception("Matrix must be in csr format")
(m,n) = A.shape
# # The data of the matrix
# data = A.data.ctypes.data_as(POINTER(c_double))
# indptr = A.indptr.ctypes.data_as(POINTER(c_int))
# indices = A.indices.ctypes.data_as(POINTER(c_int))
# Allocate output, using same conventions as input
nVectors = 1
if x.ndim is 1:
y = np.empty(m,dtype=np.double,order='F')
if x.size != n:
raise Exception("x must have n entries. x.size is %d, n is %d" % (x.size,n))
elif x.shape[1] is 1:
y = np.empty((m,1),dtype=np.double,order='F')
if x.shape[0] != n:
raise Exception("x must have n entries. x.size is %d, n is %d" % (x.size,n))
else:
nVectors = x.shape[1]
y = np.empty((m,nVectors),dtype=np.double,order='F')
if x.shape[0] != n:
raise Exception("x must have n entries. x.size is %d, n is %d" % (x.size,n))
# Check input
if x.dtype.type is not np.double:
x = x.astype(np.double,copy=True)
# Put it in column-major order, otherwise for nVectors > 1 this FAILS completely
if x.flags['F_CONTIGUOUS'] is not True:
x = x.copy(order='F')
if nVectors == 1:
np_x = x.ctypes.data_as(POINTER(c_double))
np_y = y.ctypes.data_as(POINTER(c_double))
# now call MKL. This returns the answer in np_y, which links to y
self.SpMV(byref(c_char(b"N")), byref(c_int(m)), self.Z_data , self.Z_indptr, self.Z_indices, np_x, np_y)
else:
for columns in xrange(nVectors):
xx = x[:,columns]
yy = y[:,columns]
np_x = xx.ctypes.data_as(POINTER(c_double))
np_y = yy.ctypes.data_as(POINTER(c_double))
self.SpMV(byref(c_char("N")), byref(c_int(m)),data,indptr, indices, np_x, np_y)
end = default_timer()
time_elapsed = end - start
self.update_time(code, time_elapsed)
return y
def main():
bstTimeStart = timeit.default_timer()
bst(A)
bstTimeEnd = timeit.default_timer()
print("The BST comparison count is:", bstCount)
print("The BST completion time is:", (bstTimeEnd - bstTimeStart) * 1000)
from pyspark import SparkConf
from pyspark.ml import Pipeline
from pyspark.ml.classification import DecisionTreeClassifier
from pyspark.ml.feature import StringIndexer, VectorIndexer, VectorAssembler
from pyspark.ml.evaluation import MulticlassClassificationEvaluator
from pyspark.ml.feature import PCA
from pyspark.ml.linalg import Vectors
from pyspark.sql import SparkSession
from pyspark.sql.types import StructType, StructField, StringType, IntegerType
if __name__ == "__main__":
timerstart = timeit.default_timer()
#Check arguments
if len(sys.argv) != 3:
print("Usage: train.py <csv dataset> <model output path>",
file=sys.stderr)
sys.exit(-1)
# Create spark session
spark = SparkSession\
.builder\
.appName("TrainDecisionTreeIDS-Python")\
.getOrCreate()
# Define dataset schema, dataset csv generated by flowtbag https://github.com/DanielArndt/flowtbag
schema = StructType([
def main(opts):
adj2_ = torch.from_numpy(graph.cihp2pascal_nlp_adj).float()
adj2_test = adj2_.unsqueeze(0).unsqueeze(0).expand(1, 1, 7,
20).cuda().transpose(
2, 3)
adj1_ = Variable(
torch.from_numpy(graph.preprocess_adj(graph.pascal_graph)).float())
adj3_test = adj1_.unsqueeze(0).unsqueeze(0).expand(1, 1, 7, 7).cuda()
cihp_adj = graph.preprocess_adj(graph.cihp_graph)
adj3_ = Variable(torch.from_numpy(cihp_adj).float())
adj1_test = adj3_.unsqueeze(0).unsqueeze(0).expand(1, 1, 20, 20).cuda()
p = OrderedDict() # Parameters to include in report
p['trainBatch'] = opts.batch # Training batch size
p['nAveGrad'] = 1 # Average the gradient of several iterations
p['lr'] = opts.lr # Learning rate
p['lrFtr'] = 1e-5
p['lraspp'] = 1e-5
p['lrpro'] = 1e-5
p['lrdecoder'] = 1e-5
p['lrother'] = 1e-5
p['wd'] = 5e-4 # Weight decay
p['momentum'] = 0.9 # Momentum
p['epoch_size'] = 10 # How many epochs to change learning rate
p['num_workers'] = opts.numworker
backbone = 'xception' # Use xception or resnet as feature extractor,
with open(opts.txt_file, 'r') as f:
img_list = f.readlines()
max_id = 0
save_dir_root = os.path.join(os.path.dirname(os.path.abspath(__file__)))
exp_name = os.path.dirname(os.path.abspath(__file__)).split('/')[-1]
runs = glob.glob(os.path.join(save_dir_root, 'run', 'run_*'))
for r in runs:
run_id = int(r.split('_')[-1])
if run_id >= max_id:
max_id = run_id + 1
# run_id = int(runs[-1].split('_')[-1]) + 1 if runs else 0
# Network definition
if backbone == 'xception':
net = deeplab_xception_transfer.deeplab_xception_transfer_projection_savemem(
n_classes=opts.classes,
os=16,
hidden_layers=opts.hidden_layers,
source_classes=7,
)
elif backbone == 'resnet':
# net = deeplab_resnet.DeepLabv3_plus(nInputChannels=3, n_classes=7, os=16, pretrained=True)
raise NotImplementedError
else:
raise NotImplementedError
if gpu_id >= 0:
net.cuda()
# net load weights
if not opts.loadmodel == '':
x = torch.load(opts.loadmodel)
net.load_source_model(x)
print('load model:', opts.loadmodel)
else:
print('no model load !!!!!!!!')
## multi scale
scale_list = [1, 0.5, 0.75, 1.25, 1.5, 1.75]
testloader_list = []
testloader_flip_list = []
for pv in scale_list:
composed_transforms_ts = transforms.Compose(
[tr.Scale_(pv),
tr.Normalize_xception_tf(),
tr.ToTensor_()])
composed_transforms_ts_flip = transforms.Compose([
tr.Scale_(pv),
tr.HorizontalFlip(),
tr.Normalize_xception_tf(),
tr.ToTensor_()
])
voc_val = cihp.VOCSegmentation(split='test',
transform=composed_transforms_ts)
voc_val_f = cihp.VOCSegmentation(split='test',
transform=composed_transforms_ts_flip)
testloader = DataLoader(voc_val,
batch_size=1,
shuffle=False,
num_workers=p['num_workers'])
testloader_flip = DataLoader(voc_val_f,
batch_size=1,
shuffle=False,
num_workers=p['num_workers'])
testloader_list.append(copy.deepcopy(testloader))
testloader_flip_list.append(copy.deepcopy(testloader_flip))
print("Eval Network")
if not os.path.exists(opts.output_path + 'cihp_output_vis/'):
os.makedirs(opts.output_path + 'cihp_output_vis/')
if not os.path.exists(opts.output_path + 'cihp_output/'):
os.makedirs(opts.output_path + 'cihp_output/')
start_time = timeit.default_timer()
# One testing epoch
total_iou = 0.0
net.eval()
for ii, large_sample_batched in enumerate(
zip(*testloader_list, *testloader_flip_list)):
print(ii)
#1 0.5 0.75 1.25 1.5 1.75 ; flip:
sample1 = large_sample_batched[:6]
sample2 = large_sample_batched[6:]
for iii, sample_batched in enumerate(zip(sample1, sample2)):
inputs, labels = sample_batched[0]['image'], sample_batched[0][
'label']
inputs_f, _ = sample_batched[1]['image'], sample_batched[1][
'label']
inputs = torch.cat((inputs, inputs_f), dim=0)
if iii == 0:
_, _, h, w = inputs.size()
# assert inputs.size() == inputs_f.size()
# Forward pass of the mini-batch
inputs, labels = Variable(inputs,
requires_grad=False), Variable(labels)
with torch.no_grad():
if gpu_id >= 0:
inputs, labels = inputs.cuda(), labels.cuda()
# outputs = net.forward(inputs)
# pdb.set_trace()
outputs = net.forward(inputs, adj1_test.cuda(),
adj3_test.cuda(), adj2_test.cuda())
outputs = (outputs[0] +
flip(flip_cihp(outputs[1]), dim=-1)) / 2
outputs = outputs.unsqueeze(0)
if iii > 0:
outputs = F.upsample(outputs,
size=(h, w),
mode='bilinear',
align_corners=True)
outputs_final = outputs_final + outputs
else:
outputs_final = outputs.clone()
################ plot pic
predictions = torch.max(outputs_final, 1)[1]
prob_predictions = torch.max(outputs_final, 1)[0]
results = predictions.cpu().numpy()
prob_results = prob_predictions.cpu().numpy()
vis_res = decode_labels(results)
parsing_im = Image.fromarray(vis_res[0])
parsing_im.save(opts.output_path +
'cihp_output_vis/{}.png'.format(img_list[ii][:-1]))
cv2.imwrite(
opts.output_path + 'cihp_output/{}.png'.format(img_list[ii][:-1]),
results[0, :, :])
# np.save('../../cihp_prob_output/{}.npy'.format(img_list[ii][:-1]), prob_results[0, :, :])
# pred_list.append(predictions.cpu())
# label_list.append(labels.squeeze(1).cpu())
# loss = criterion(outputs, labels, batch_average=True)
# running_loss_ts += loss.item()
# total_iou += utils.get_iou(predictions, labels)
end_time = timeit.default_timer()
print('time use for ' + str(ii) + ' is :' + str(end_time - start_time))
# Eval
pred_path = opts.output_path + 'cihp_output/'
eval_(pred_path=pred_path,
gt_path=opts.gt_path,
classes=opts.classes,
txt_file=opts.txt_file)
matrix[k - 1][j - 1] = 1 + lcs(X, Y, k - 1, j - 1, matrix)
return matrix[k - 1][j - 1]
else: # 최적 부분 구조 성질 2번, 3번
# store it in arr to avoid further repetitive
# work in future function calls
matrix[k - 1][j - 1] = max(lcs(X, Y, k, j - 1, matrix), lcs(X, Y, k - 1, j, matrix))
return matrix[k - 1][j - 1]
sys.setrecursionlimit(10000)
n = int(input("Size: ")) # 생성할 문자열의 길이
string_pool = string.ascii_uppercase # 대문자만 이용하여 문자열 생성
X = ""
Y = ""
for i in range(n): # 랜덤한 문자열 생성
X += random.choice(string_pool) # 랜덤한 문자열 하나 선택
Y += random.choice(string_pool) # 랜덤한 문자열 하나 선택
print("X: ", X) # 입력 수열1 출력
print("Y: ", Y) # 입력 수열2 출력
n = len(Y) # 입력 수열의 길이 n
dp = [[-1] * (n+1) for _ in range(n+1)] # 이전에 계산한 값을 저장할 2차원 리스트 선언
t1 = timeit.default_timer() # LCS 알고리즘 시작시간
print("Length of LCS:", lcs(X, Y, n, n, dp)) # LCS 함수 호출 및 반환값 출력
t2 = timeit.default_timer() # LCS 알고리즘 종료시간
print("Running time: ", (t2 - t1) * 1000) # 삽입 정렬 실행시간
def __enter__(self):
self._start = default_timer()
conn_spec=conn_dict)
A = mmread('../ii.wmat')
#rows, cols = A.nonzero()
nest.Connect(A.row + NE + 1,
A.col + NE + 1,
syn_spec=inh_syn_dict,
conn_spec=conn_dict)
if (not fast):
spikes = nest.Create("spike_detector", 1, [{
"label": "va-py-ex",
"to_file": True
}])
spikes_E = spikes[:1]
nest.Connect(nodes_E[:N_rec], spikes_E)
starttime = timeit.default_timer()
nest.Simulate(simtime)
totaltime = timeit.default_timer() - starttime
print("Real Time Sim: " + str(totaltime) + "s")
if (fast):
f = open("timefile.dat", "w")
f.write("%f" % totaltime)
f.close()
if (not fast):
rate_iaf = nest.GetStatus(spikes)[0]["n_events"] / (
(simtime / 1000.0) * N_rec)
print("Average Rate of recorded electrodes: " + str(rate_iaf) + "Hz")
def __exit__(self, typ, value, traceback):
# Time can go backwards.
self._gauge.set(max(default_timer() - self._start, 0))
def __exit__(self, typ, value, traceback):
# Time can go backwards.
self._histogram.observe(max(default_timer() - self._start, 0))