def process_folder(self):
if self.in_folder.endswith(".txt"):
paths_list = read_file(self.in_folder)
else:
paths_list, _ = traverse_dir_files(self.in_folder)
print('[Info] 样本数: {}'.format(len(paths_list)))
random.seed(47)
random.shuffle(paths_list)
if len(paths_list) > self.max_num:
paths_list = paths_list[:self.max_num]
print('[Info] 样本数: {}'.format(len(paths_list)))
time_str = get_current_time_str()
out_file = os.path.join(self.out_folder, "val_{}.txt".format(time_str))
out_html = os.path.join(self.out_folder,
"val_{}.html".format(time_str))
pool = Pool(processes=100)
for img_idx, img_path in enumerate(paths_list):
if "rotation_datasets_hardcase" in img_path: # 过滤hardcase
continue
# ServiceTester.process_img_path(img_idx, img_path, self.service, out_file)
pool.apply_async(ServiceTester.process_img_path,
(img_idx, img_path, self.service, out_file))
pool.close()
pool.join()
print('[Info] 处理完成: {}'.format(out_file))
data_lines = read_file(out_file)
print('[Info] 正确率: {}'.format(
safe_div(len(paths_list) - len(data_lines), len(paths_list))))
out_list = []
for data_line in data_lines:
items = data_line.split("\t")
out_list.append(items)
make_html_page(out_html, out_list)
print('[Info] 处理完成: {}'.format(out_html))
def filter_checked_urls(self):
in_dir = os.path.join(DATA_DIR, 'datasets_v4_checked_r_txt')
out_dir = os.path.join(
DATA_DIR,
'datasets_v4_checked_r_urls_out_{}'.format(get_current_time_str()))
mkdir_if_not_exist(out_dir)
paths_list, names_list = traverse_dir_files(in_dir)
pool = Pool(processes=80)
idx = 0
for in_path, in_name in zip(paths_list, names_list):
out_error_path = os.path.join(out_dir,
'{}.error.txt'.format(in_name))
out_right_path = os.path.join(out_dir,
'{}.right.txt'.format(in_name))
print('[Info] out_file: {} - {}'.format(out_error_path,
out_right_path))
data_lines = read_file(in_path)
print('[Info] 文本数量: {}'.format(len(data_lines)))
for data_line in data_lines:
url = data_line
pool.apply_async(DatasetFilter.check_url,
(idx, url, out_error_path, out_right_path))
idx += 1
if idx % 1000 == 0:
print('[Info] idx: {}'.format(idx))
pool.close()
pool.join()
print('[Info] 处理完成: {}'.format(out_dir))
def run(num, method, args):
p = Pool(num)
for arg in args:
p.apply_async(method, args=(arg, ))
p.close()
p.join()
p.terminate()
class Mobile(Device):
def __init__(self, device_id, address):
super().__init__(device_id)
self.address = address
self.p = Pool(6)
self.capture = None
self.pre_time = time.time() * 1000
def tap_button(self, button):
cmd = "adb -s {:s} shell input tap {:d} {:d}".format(
self.device_id, button[0], button[1])
self.p.apply_async(execute_cmd, args={cmd})
def swipe(self, action):
cmd = "adb -s {:s} shell input swipe {:d} {:d} {:d} {:d} 300".format(
self.device_id, action[0], action[1], action[2], action[3])
self.p.apply_async(execute_cmd, args={cmd})
def get_frame(self):
if self.capture is None:
self.capture = cv2.VideoCapture(self.address)
state, img = self.capture.read()
if state:
time_mill = time.time() * 1000
if time_mill - self.pre_time >= 500:
self.pre_time = time_mill
return [cv2.resize(img, (540, 960)), 0]
else:
return [None, 0]
else:
self.capture.release()
self.capture = None
time.sleep(10)
return [None, -1]
def process(self):
data_dir = os.path.join(DATA_DIR, 'labeled_data')
print('[Info] 数据文件夹: {}'.format(data_dir))
out_dir = os.path.join(
DATA_DIR, 'labeled_data_out_{}'.format(get_current_time_str()))
print('[Info] 输出文件夹: {}'.format(out_dir))
mkdir_if_not_exist(out_dir)
paths_list, names_list = traverse_dir_files(data_dir)
print('[Info] 文件数: {}'.format(len(paths_list)))
out_file_format = os.path.join(out_dir, 'labeled_data_imgs_{}.txt')
pool = Pool(processes=80)
for path, name in zip(paths_list, names_list):
name = name.split(".")[0]
out_file = out_file_format.format(name)
print('[Info] 输出文件: {}'.format(out_file))
data_lines = read_file(path)
for idx, data_line in enumerate(data_lines):
if idx == 0:
continue
# DataPrelabeled.process_line(out_file, idx, data_line)
pool.apply_async(DataPrelabeled.process_line,
(out_file, idx, data_line))
pool.close()
pool.join()
print('[Info] 处理完成: {}'.format(out_dir))
def get_img_href():
"""
与用户交互获取文件保存的路径, 使用多进程-多线程,下载图片
:return: 图片保存情况
"""
url = "http://www.cct58.com/mneinv/1.html"
# head = main_head()
save_path = input("文件保存在: ")
# 根据电脑CPU个数, 创建多进程
os.cpu_count()
# pool = Pool(os.cpu_count())
pool = Pool(10)
# 遍历获取每一页的url链接
for href in get_every_page(url, main_head(), proxy):
try:
for person_mx27, title in get_every_person(href, main_head(),
proxy):
# print(person_mx27, title) http://www.cct58.com/mneinv/19497/mx27/ 夏夏
# downloads(person_mx27, head, proxy,save_path)
pool.apply_async(downloads,
args=(person_mx27, main_head(), proxy,
save_path))
except:
continue
pool.close()
pool.join()
def main(original_dir, output_dir, crf, thread_count=None, overwrite=False):
"""
Compresses every MP4 video in a directory using the H.264 codec.
Args:
original_dir: Directory containing original videos.
output_dir: Directory to output new videos to. Will be created if
it does not exist.
crf: Constant rate factor, i.e. the amount of compressing to perform.
A higher number means higher compression.
thread_count: Amount of threads to use. Set to None automatically
choose a number of threads.
"""
if not os.path.exists(output_dir):
os.makedirs(output_dir)
videos = get_videos(original_dir, output_dir, overwrite)
print('Compressing {} videos...'.format(len(videos)))
pool = Pool(thread_count)
for video, output in videos:
pool.apply_async(compress_worker, (video, output, crf))
pool.close()
pool.join()
print('Done')
def check_all(self):
func = self.check
pool = Pool()
for token in self.all():
pool.apply_async(func, args=(token, ))
pool.join()
pool.close()
def process_folder(self, in_dir, out_dir):
"""
处理文件夹
"""
print('[Info] in_dir: {}'.format(in_dir))
print('[Info] out_dir: {}'.format(out_dir))
mkdir_if_not_exist(out_dir)
paths_list, names_list = traverse_dir_files(in_dir)
print('[Info] 待处理文件数量: {}'.format(len(paths_list)))
random.seed(47)
paths_list, names_list = shuffle_two_list(paths_list, names_list)
n_prc = 40
pool = Pool(processes=n_prc) # 多线程下载
for idx, (path, name) in enumerate(zip(paths_list, names_list)):
pool.apply_async(DataProcessor.process_img,
args=(path, name, out_dir))
# DataProcessor.process_img(path, name, out_dir)
if (idx + 1) % 1000 == 0:
print('[Info] num: {}'.format(idx + 1))
# 多进程逻辑
pool.close()
pool.join()
print('[Info] 处理完成! {}'.format(out_dir))
return
class QuantDslApplicationWithMultiprocessing(QuantDslApplication):
def __init__(self, num_workers=None, call_evaluation_queue=None, **kwargs):
if num_workers is not None:
assert call_evaluation_queue is None
# Parent.
self.pool = Pool(processes=num_workers)
self.manager = Manager()
self.call_evaluation_queue = self.manager.Queue()
else:
# Child.
self.pool = None
super(QuantDslApplicationWithMultiprocessing, self).__init__(call_evaluation_queue=call_evaluation_queue, **kwargs)
if self.pool:
# Start worker pool.
app_kwargs = self.get_subprocess_application_args()
args = (self.manager.Lock(), self.__class__, app_kwargs)
for i in range(num_workers):
self.pool.apply_async(loop_on_evaluation_queue, args)
def get_subprocess_application_args(self):
app_kwargs = dict(
call_evaluation_queue=self.call_evaluation_queue,
)
return app_kwargs
def close(self):
super(QuantDslApplicationWithMultiprocessing, self).close()
if self.pool:
self.pool.terminate()
def download_right_angle(self):
files_dir = os.path.join(ROOT_DIR, '..', 'datasets',
'2020_11_26_vpf_right')
paths_list, names_list = traverse_dir_files(files_dir)
pool = Pool(processes=80)
for path, name in zip(paths_list, names_list):
name_x = name.split('.')[0]
urls_file = os.path.join(ROOT_DIR, '..', 'datasets',
'2020_11_26_vpf_right',
'{}.txt'.format(name_x)) # 输入
out_dir = os.path.join(ROOT_DIR, '..', 'datasets',
'datasets_v4_checked', 'vpf_right',
name_x) # 输出
mkdir_if_not_exist(out_dir)
data_lines = read_file(urls_file)
for idx, data_line in enumerate(data_lines):
url, angle = data_line.split(',')
pool.apply_async(DatasetFilter.process_img_angle,
(idx, url, angle, out_dir))
pool.close()
pool.join()
print('[Info] 处理完成: {}'.format(files_dir))
def work(host, port, processes, threads, times):
pool = Pool(processes,
lambda: signal.signal(signal.SIGINT, signal.SIG_IGN))
p = Process(target=progress)
p.daemon = True
start = time.time()
try:
for chunk in divide(times, processes):
pool.apply_async(thread, (host, port, threads, chunk))
p.start()
pool.close()
pool.join()
p.terminate()
p.join()
except KeyboardInterrupt:
pool.terminate()
p.terminate()
p.join()
pool.join()
return time.time() - start
def process(self):
name = "nat_main_all_20211020"
file_path = os.path.join(DATA_DIR, "{}.txt".format(name))
print("[Info] 输入文件: {}".format(file_path))
out_file_path = os.path.join(
DATA_DIR, "{}.out.{}.txt".format(name, get_current_time_str()))
out_html_path = os.path.join(DATA_DIR, "{}.out.html".format(name))
data_lines = read_file(file_path)
random.seed(47)
random.shuffle(data_lines)
# data_lines = data_lines[:1000]
print('[Info] 样本数: {}'.format(len(data_lines)))
pool = Pool(processes=20)
for data_idx, data_line in enumerate(data_lines):
# RoiChecker.process_line_roi(data_idx, data_line, out_file_path)
pool.apply_async(RoiChecker.process_line_roi,
(data_idx, data_line, out_file_path))
pool.close()
pool.join()
print('[Info] 写入完成: {}'.format(out_file_path))
data_lines = read_file(out_file_path)
items_list = []
for data_line in data_lines:
items_list.append(data_line.split("\t"))
make_html_page(out_html_path, items_list)
print('[Info] 写入完成: {}'.format(out_html_path))
class _MultiExecutor(_Executor):
"""Execute functions async in a process pool"""
def __init__(self):
super(_MultiExecutor, self).__init__()
self._children = 0
self.pool = Pool()
def _collector(self, result):
super(_MultiExecutor, self)._collector(result)
self._children -= 1
def execute(self, func, args):
self._children += 1
self.pool.apply_async(func, args, callback=self._collector)
def wait_for_results(self):
self.pool.close()
# One would have hoped joining the pool would take care of this, but
# apparently you need to first make sure that all your launched tasks
# has returned their results properly, before calling join, or you
# risk a deadlock.
while self._children > 0:
time.sleep(0.001)
self.pool.join()
def run():
p = Pool(5) # 最多执行5个进程，打印5个数
for i in range(10000):
p.apply_async(fun, args=(i,))
p.close()
p.join() # 等待所有子进程结束，再往后执行
print("end")
def process_v2(self):
val_folder = os.path.join(ROOT_DIR, '..', 'datasets', 'datasets_val')
data_lines, _ = traverse_dir_files(val_folder)
print('[Info] 样本数: {}'.format(len(data_lines)))
type_name = "val"
folder_name = "dataset_{}_{}".format(type_name, len(data_lines))
dataset_folder = os.path.join(self.out_ds_folder, folder_name)
mkdir_if_not_exist(dataset_folder)
print('[Info] 输出文件夹路径: {}'.format(dataset_folder))
mkdir_if_not_exist(self.out_files_folder)
out_path_file = os.path.join(self.out_files_folder, "{}.txt".format(folder_name))
print('[Info] 输出文件路径: {}'.format(out_path_file))
pool = Pool(processes=100)
for data_idx, data_line in enumerate(data_lines):
pool.apply_async(
DatasetReorder.copy_line_mul, (data_idx, data_line, type_name, dataset_folder, out_path_file))
pool.close()
pool.join()
path_list = read_file(out_path_file)
print('[Info] 输出路径: {}, 样本数: {}'.format(len(path_list), len(data_lines)))
print('[Info] 处理完成: {}'.format(out_path_file))
def process(self):
data_dir = os.path.join(DATA_DIR, '2020_12_23')
print('[Info] 数据文件: {}'.format(data_dir))
out_file = os.path.join(DATA_DIR, '2020_12_23.txt')
print('[Info] 写出文件: {}'.format(out_file))
paths_list, names_list = traverse_dir_files(data_dir)
print('[Info] 文件数: {}'.format(len(paths_list)))
data_lines = []
for path, name in zip(paths_list, names_list):
sub_lines = read_file(path)
data_lines += sub_lines
print('[Info] 文本行数: {}'.format(len(data_lines)))
random.seed(47)
random.shuffle(data_lines)
pool = Pool(processes=80)
for idx, data_line in enumerate(data_lines):
# DataPreprocess.process_line(idx, data_line, out_file)
pool.apply_async(DataPreprocess.process_line, (idx, data_line, out_file))
pool.close()
pool.join()
print('[Info] 处理完成: {}'.format(out_file))
def write_sub_bam(chrom_list, used_bam_file_tmp, exclude_bam_file_tmp, out_dir,
total_modify_reads_file, total_delete_reads_file,
total_add_reads_file, process):
write_bam_pool = Pool(int(process))
exclude_bam_list = [exclude_bam_file_tmp]
usedBamList = []
for chrom in chrom_list:
excludeBam_chr = "%s/exclude_%s.bam" % (out_dir, chrom)
exclude_bam_list.append(excludeBam_chr)
usedBam_chr = "%s/used_%s.bam" % (out_dir, chrom)
usedBamList.append(usedBam_chr)
write_bam_pool.apply_async(
write_bam_byChr,
args=(used_bam_file_tmp, chrom, excludeBam_chr, usedBam_chr,
total_modify_reads_file, total_delete_reads_file,
total_add_reads_file))
write_bam_pool.close()
write_bam_pool.join()
exclude_bam_file = os.path.join(out_dir, "exclude.bam")
bamMerge(exclude_bam_list, exclude_bam_file)
used_bam_file = os.path.join(out_dir, "used.bam")
if len(usedBamList) != 1:
bamMerge(usedBamList, used_bam_file)
else:
used_bam_file = usedBamList[0]
bamSort(used_bam_file, os.path.join(out_dir, "used.sort"))
used_sort_bam_file = os.path.join(out_dir, "used.sort.bam")
bamIndex(used_sort_bam_file)
return used_sort_bam_file, exclude_bam_file
def insert_from_log(retry=3, docs=None):
"""
retry: 循环次数，一次循环中如果有失败的任务则会开启下一次循环，循环数-1，直到循环数为0。循环数设为负数表示循环直到所有任务完成。
"""
from multiprocessing.pool import Pool
import multiprocessing
retry -= 1
if docs is None:
docs = find()
pool = Pool(None, initializer, ({name: globals()[name] for name in VARIABLES},))
for doc in docs:
pool.apply_async(handle_doc, kwds=doc, error_callback=on_error)
try:
pool.close()
pool.join()
except KeyboardInterrupt:
pool.terminate()
return
docs = list(find())
logging.warning("Unfilled: %s", len(docs))
if retry:
logging.warning("retry | %s", retry)
if docs:
insert_from_log(retry, docs)
def read_group_posts(history_files, group_dict):
"""
Each user is represented by a list of strings which are their posts.
The posts go through preprocessing.
:param history_files: a list of paths to group Twitter history file
:param group_dict: group dictionary to insert the preprocessed posts to
:return: List of lists representing a list of user posts
"""
posts = manager.list()
def update(*args):
pbar.update()
last_index = 0
for hist_file in history_files:
with open(hist_file) as f:
hist = json.load(f)
pool = Pool(processes=6,
initializer=init,
initargs=(posts, group_dict))
indices = range(last_index, last_index + len(hist))
pbar = tqdm(zip(indices, hist.items()),
total=len(hist.items()),
desc='Reading {}'.format(hist_file))
for item in zip(indices, hist.items()):
pool.apply_async(get_posts, args=(item, ), callback=update)
last_index = item[0] + 1
pool.close()
pool.join()
pbar.close()
return posts
def main():
args = dict(enumerate(sys.argv))
day = args.get(1) or datetime.now().strftime("%Y%m%d")
if len(day) != 8:
raise Exception
print('date={0}'.format(day))
print('Start dumper...')
goods = 'goods_{0}'.format(day)
new = 'new_goods_{0}'.format(day)
hot = 'hot_goods_{0}'.format(day)
shop = 'shop_statistics_{0}'.format(day)
detail = 'd_{0}'.format(day)
monitor = 'm_{0}'.format(day)
# todo 保存目录为可选项
# todo 保存目录为 linux windows 兼容
pool = Pool()
pool.apply_async(func=dump_mongodb, args=(detail, ))
pool.apply_async(func=dump_mongodb, args=(monitor, ))
pool.apply_async(func=dump_mysql, args=(goods, ))
pool.apply_async(func=dump_mysql, args=(new, ))
pool.apply_async(func=dump_mysql, args=(hot, ))
pool.apply_async(func=dump_mysql, args=(shop, ))
pool.close()
pool.join()
class RVWorkerPool(object):
def __init__(self, processes=None):
self.pool = Pool(processes)
self.pipe_target = os.cpu_count()
self.in_pipe = 0
self.jobs = []
def registerJob(self, func, args, ref=None, priority=1):
job = JobRef(func, args, ref, priority)
# Replace duplicate-source unstarted jobs
if ref is not None:
self.jobs = [j for j in self.jobs if j.ref != ref or j.started()]
self.jobs.append(job)
if self.in_pipe < self.pipe_target:
job.jobStarted(self.pool.apply_async(func, args))
self.in_pipe += 1
return job
def poll(self):
done = []
sorted_jobs = sorted(self.jobs, key=lambda j: j.priority)
for job in sorted_jobs:
if job.started():
# Check if it's finished
if job.ready():
self.jobs.remove(job)
self.in_pipe -= 1
else:
if self.in_pipe < self.pipe_target:
job.jobStarted(self.pool.apply_async(job.func, job.args))
self.in_pipe += 1
def terminate(self):
self.pool.terminate()
def diff_pro(dbname, tablename):
"difference between slave and master "
if is_sigint_up:
os._exit(0)
pool = Pool(processes=2)
result = {}
result['slave'] = pool.apply_async(table_count,
args=(myhost, myport, myuser,
mypassword, dbname, tablename))
result['master'] = pool.apply_async(table_count,
args=(mahost, maport, mauser,
mapassword, dbname, tablename))
pool.close()
pool.join()
slcount = result['slave'].get()['count']
slstart = result['slave'].get()['start']
slend = result['slave'].get()['end']
slengine = result['slave'].get()['engine']
sldelay = result['slave'].get()['delay']
macount = result['master'].get()['count']
mastart = result['master'].get()['start']
maend = result['master'].get()['end']
maengine = result['master'].get()['engine']
dbtable = '%s.%s' % (dbname, tablename)
if slcount == -2 or macount == -2:
mark = 'error'
elif slcount == -1 or macount == -1:
mark = '*'
else:
mark = macount - slcount
msg = "%-40s |%-9s %-9s|%-8s %-8s |%-8s %-8s |%-7s|%-15s %-15s |%-10s" % (
dbtable, maengine, slengine, mastart, slstart, maend, slend, sldelay,
macount, slcount, mark)
print msg
result.clear()
def compress_batch(input_folder, output_folder, target_compression, recursive=True, threads=1, follow_symlinks=True):
# We require an absolute input path to we can replicate the data structure relative to it later on
input_folder = os.path.abspath(input_folder)
file_list = get_fast5_file_list(input_folder, recursive, follow_symlinks=follow_symlinks)
if len(file_list) == 0:
raise ValueError("No input fast5 files found in '{}'. Recursive={}".format(input_folder, recursive))
# Set up the process pool and the progressbar
pool = Pool(min(threads, len(file_list)))
pbar = get_progress_bar(len(file_list))
def update(result):
pbar.update(pbar.currval + 1)
for input_file in file_list:
input_path = os.path.join(input_folder, input_file)
output_path = os.path.join(output_folder, os.path.relpath(input_path, input_folder))
pool.apply_async(func=compress_file,
args=(input_path, output_path, target_compression),
callback=update)
# Tear down the process pool and pbar. We can't use contextmanagers since we need to close() then join()
pool.close()
pool.join()
pbar.finish()
def test_queue_and_thread():
strategy_name = parse_args()
manager = Manager()
in_queue = manager.Queue()
out_queue = manager.Queue()
# monitor([in_queue, out_queue])
pool = Pool(processes=3)
results = [
pool.apply_async(start_engine, (in_queue, out_queue, strategy_name)),
pool.apply_async(start_feed, (in_queue, )),
pool.apply_async(start_executor, (out_queue, ))
]
try:
for r in results:
r.get()
except KeyboardInterrupt:
print("Keybord interruption in main thread")
finally:
print("Cleaning main thread")
def fill_views(self):
client = Client('hscic')
pool = Pool(processes=len(self.view_paths))
tables = []
prescribing_date = ImportLog.objects.latest_in_category(
'prescribing').current_at.strftime('%Y-%m-%d')
for path in self.view_paths:
table_name = "vw__%s" % os.path.basename(path).replace('.sql', '')
table = client.get_table(table_name)
tables.append(table)
with open(path) as f:
sql = f.read()
substitutions = {'this_month': prescribing_date}
args = [table.name, sql, substitutions]
pool.apply_async(query_and_export, args)
pool.close()
pool.join() # wait for all worker processes to exit
for table in tables:
self.download_and_import(table)
self.log("-------------")
def main():
parser = argparse.ArgumentParser()
parser.add_argument('--boosted', action='store_true')
args = parser.parse_args()
boosted = args.boosted
if boosted:
pool = Pool(16)
session, old_ip = renew_connection('')
old_ip = ''
else:
session = new_session()
req = Request('GET', CURLIE)
prepared_req = session.prepare_request(req)
resp = session.send(prepared_req, timeout=12)
soup = BeautifulSoup(resp.text, 'html.parser')
print('id,parent_id,name')
cats = soup.select('.top-cat > a')
for cat in cats:
print('"' + cat['href'] + '","","' + cat.text + '"')
if boosted:
session, old_ip = renew_connection(old_ip)
pool.apply_async(get_nodes, args=[cat['href'], '- ', session])
else:
get_nodes(cat['href'], session)
time.sleep(1)
if boosted:
pool.close()
pool.join()
def download_imgs_for_mp(img_file, out_folder, prefix=None, n_prc=10):
"""
多线程下载
:param img_file: 图片文件
:param out_folder: 输出文件夹
:param prefix: 图片前缀
:param n_prc: 进程数, 默认40个
:return: None
"""
print_info('进程总数: %s' % n_prc)
pool = Pool(processes=n_prc) # 多线程下载
paths_list = read_file(img_file)
print_info('文件数: %s' % len(paths_list))
_, imgs_names = traverse_dir_files(out_folder)
for (index, path) in enumerate(paths_list):
if prefix:
pool.apply_async(download_img, (path, out_folder, imgs_names, prefix + '_' + str(index) + '.jpg'))
else:
pool.apply_async(download_img, (path, out_folder, imgs_names))
pool.close()
pool.join()
# _, imgs_names = traverse_dir_files(out_folder)
# print_info('图片总数: %s' % len(imgs_names))
print_info('全部下载完成')
def merge_hardcase(self, file_list, type_name):
data_lines = []
for file_name in file_list:
file_path = os.path.join(self.folder, file_name)
sub_lines = read_file(file_path)
data_lines += sub_lines
print('[Info] 样本行数: {}'.format(len(data_lines)))
folder_name = "dataset_{}_{}".format(type_name, len(data_lines))
dataset_folder = os.path.join(self.out_ds_folder, folder_name)
mkdir_if_not_exist(dataset_folder)
print('[Info] 输出文件夹路径: {}'.format(dataset_folder))
mkdir_if_not_exist(self.out_files_folder)
out_path_file = os.path.join(self.out_files_folder, "{}.txt".format(folder_name))
print('[Info] 输出文件路径: {}'.format(out_path_file))
pool = Pool(processes=100)
for data_idx, data_line in enumerate(data_lines):
pool.apply_async(
DatasetReorder.copy_line_mul, (data_idx, data_line, type_name, dataset_folder, out_path_file))
pool.close()
pool.join()
path_list = read_file(out_path_file)
print('[Info] 输出路径: {}, 样本数: {}'.format(len(path_list), len(data_lines)))
print('[Info] 处理完成: {}'.format(out_path_file))
def process_pool():
pool = Pool(processes=3)
for i in range(6):
pool.apply_async(func)
pool.close()
pool.join()
def fill_views(self):
client = Client('hscic')
pool = Pool(processes=len(self.view_paths))
tables = []
prescribing_date = ImportLog.objects.latest_in_category(
'prescribing').current_at.strftime('%Y-%m-%d')
for path in self.view_paths:
table_name = "vw__%s" % os.path.basename(path).replace('.sql', '')
table = client.get_table(table_name)
tables.append(table)
with open(path) as f:
sql = f.read()
substitutions = {'this_month': prescribing_date}
args = [table.table_id, sql, substitutions]
pool.apply_async(query_and_export, args)
pool.close()
pool.join() # wait for all worker processes to exit
for table in tables:
self.download_and_import(table)
self.log("-------------")
with connection.cursor() as cursor:
for view_id in self.materialized_views:
# This is quite slow! up to 10 mins.
cursor.execute("REFRESH MATERIALIZED VIEW %s" % view_id)
def work(host, port, processes, threads, times):
pool = Pool(processes,
lambda: signal.signal(signal.SIGINT, signal.SIG_IGN))
p = Process(target=progress)
p.daemon = True
start = time.time()
try:
for chunk in divide(times, processes):
pool.apply_async(thread, (host, port, threads, chunk))
p.start()
pool.close()
pool.join()
p.terminate()
p.join()
except KeyboardInterrupt:
pool.terminate()
p.terminate()
p.join()
pool.join()
return time.time() - start
def main():
print('Process (%s) start...' % os.getpid())
p = Pool()
for i in range(4):
p.apply_async(long_time_task, args=(i,))
print('Waiting for all subprocesses done...')
p.close()
p.join()
print('All subprocesses done.')
class TcpController(object):
def __init__(self,handlers):
self.handlers=handlers
self.workers=Pool(MAX_PROCESS_POOL_SIZE)
def process(self,header,body):
self.workers.apply_async(wrap,(self.handlers[header](),body,))
def destroy(self):
self.handlers=None
self.workers.close()
def _get(self, args):
draft_id = args[0]
id = args[1] if len(args) > 1 else None
q = self.db.query(Player)
if id is not None:
player = q.filter(Player.id == int(id)).first()
team = self.db.query(Team).filter(and_(Team.is_owner == True,
Team.draft_id == draft_id)).first()
available_players = self.db.query(Player).join(Player.core).filter(and_(PlayerCore.rank != None,
PlayerCore.target_price != None,
PlayerCore.points > 0,
Player.draft_id == draft_id,
Player.team_id == None,
Player.id != player.id)).order_by(PlayerCore.rank).all()
min_price = 1
max_price = min(player.core.target_price + 21, team.money)
manager = Manager()
max_starters_points = manager.dict()
max_bench_points = manager.dict()
pool = Pool(processes=8)
starters, bench = get_starters_and_bench(self.db, team.id)
max_starters_points[0] = optimizer.optimize_roster(starters, available_players, team.money - (constants.BENCH_SIZE - len(bench)))[1]
for m in range(min_price, 10):
pool.apply_async(wrap_optimizer, args=(starters, available_players, team.money - m - (constants.BENCH_SIZE - len(bench)) + 1, max_bench_points, m))
full_starters = True
for s in starters:
if s is None:
full_starters = False
if not full_starters:
starters_clone = list(starters)
bench_clone = list(bench)
place_player(player, starters_clone, bench_clone)
for m in range(min_price, max_price):
pool.apply_async(wrap_optimizer, args=(starters_clone, available_players, team.money - m - (constants.BENCH_SIZE - len(bench_clone)), max_starters_points, m))
pool.close()
pool.join()
ret = player.to_dict(['core'])
ret['max_starters_points'] = dict(max_starters_points)
ret['max_bench_points'] = dict(max_bench_points)
return ret
else:
players = q.join(PlayerCore).filter(and_(Player.draft_id == int(draft_id),
PlayerCore.rank != None,
PlayerCore.target_price != None)).all()
return {'players': [p.to_dict(['core']) for p in players]}
def main():
"""
Build all the models. Spin off a new process for each participant
because the ANN library is not multithreaded. Process is used instead
of thread to leverage multiple cores.
"""
parser = ArgumentParser()
parser.add_argument("inputFilename")
parser.add_argument("outputDirectory")
args = parser.parse_args()
inputFilename = args.inputFilename
outputDirectory = args.outputDirectory
data = pickle.load( open(inputFilename, 'rb') )
tasks = [ 'matb', 'rantask' ]
participantIds = [ '001', '002', '003', '004', '005', '006', '007' ]
# Cut off first row header for each data set
for task in tasks:
for participantId in participantIds:
data[participantId][task] = data[participantId][task][1:]
splits = performSplit( data )
# Record start time so that the elapsed time can be determined
start_time = time.time()
# Create a multicore processing pool with 7 processes ( 7 so that one core stays free
# for system processes )
pool = Pool( processes = 7 )
# Build models for participants in a task
for task in tasks:
for participantId in participantIds:
outputFilename = path.join( outputDirectory, 'testingOn-' + participantId + '-' + task + '.txt' )
# Spin off a process for the building
pool.apply_async( tuneANN, ( splits[participantId][task], outputFilename ) )
# Close down the pool so that we can wait on all the processes
pool.close()
pool.join()
# Calculate and print the elapsed time
elapsed_time = time.time() - start_time
print( "Elapsed time: " + str(elapsed_time) )
def stat_volume(stime,etime):
tgsinfo = read_tgs_info()
# from multiprocessing.dummy import Pool as ThreadPool
from multiprocessing.pool import Pool
pool = Pool()
volume = [pool.apply_async(stat_tgs_volume,args=(stime,etime,int(cid))) for cid in tgsinfo.keys()]
pool.close()
print 'waiting to join....'
pool.join()
print 'start to writing to file...'
volume0 = []
for i,elem in enumerate(volume):
volume0.append((tgsinfo.keys()[i], elem.get()))
volume0.sort(key=lambda x:x[1], reverse=True)
total = 0
with open(os.path.join(root_dir, "result", "volume.txt"),"w") as f:
for i,elem in enumerate(volume0):
# cid = tgsinfo.keys()[i]
# vol = elem.get()
total += elem[1]
line = "%5s,%s: %d\n" % (elem[0], tgsinfo[elem[0]]['kkmc'], elem[1])
f.write(line)
print 'totally %d records.' % (total)
def manager_process(dir_queue, file_queue, out_queue):
"""Dispatches and manages path and scanning workers.
"""
pool = Pool(options.num_threads)
atexit.register(at_exit_manager, pool)
logging.info('Gathering Files...')
pool.apply(explore_path, (dir_queue, file_queue))
logging.info('Files gathered. Scanning %s files...', file_queue.qsize())
logging.info('Starting %s scan processes...', options.num_threads)
print '~' * 80
thread.start_new_thread(print_status, (file_queue,))
for _ in range(options.num_threads):
pool.apply_async(parallel_scan, (file_queue, out_queue))
pool.close()
pool.join()
out_queue.put(StopIteration)
def run_jar_new_thread(config_path, **kwargs):
"""
kwargs - arguments dictionary of run_jar()
"""
pool = Pool(processes=1)
__import__("ipdb").set_trace()
result = pool.apply_async(_run_jar_with_config, [config_path], kwargs['kwargs'])
job_id = result.get()
return job_id
def run(self):
if(self.fileName[0] != None and self.fileName[0] !=""):
# p = Process(target=self.creationPDF, args=(self.fileName[0],))
# self.creationPDF(self.fileName[0])
# p.start()
pool = Pool(processes=4) # start 4 worker processes
result = pool.apply_async(self.creationPDF, [self.fileName[0]])
else:
print("Sauvegarde annule")
def multiprocess_all_chromosomes(func, cls, *args, **kwargs):
'''
Convenience method for splitting up queries based on tag id.
'''
processes = current_settings.ALLOWED_PROCESSES
set_chromosome_lists(cls, use_table=kwargs.get('use_table', None))
p = Pool(processes)
try:
for chr_list in current_settings.CHR_LISTS:
p.apply_async(func, args=[cls, chr_list, ] + list(args))
p.close()
p.join()
except Exception as e:
print('Terminating pool.')
p.terminate()
raise e
def create_execution_pool():
global execution_pool
pool_size = engine.app.config['POOL_SIZE']
execution_pool = Pool(pool_size, initializer=initialize_worker)
futures = []
for i in xrange(pool_size * 2):
futures.append(execution_pool.apply_async(connect_worker))
[f.get() for f in futures]
def main():
queue_logger = setup_redirection()
queue_logger.write("ABCDEF\n")
try:
p = Pool(10)
results = [p.apply_async(some_process_body) for i in xrange(20)]
[result.get() for result in results]
p.close()
finally:
queue_logger.stop()
def _transport_backup_parallel(data, filename, aws_key, aws_secret, bucketname):
"""
Parallel multipart upload.
"""
headers = {}
_logger.info('Backing up via S3 parallel multipart upload agent')
keyname = filename
tempInFile = NamedTemporaryFile(suffix='.zip', prefix='db-backup-', delete=False)
tempInFile.write(data)
tempInFile.close()
source_path = tempInFile.name
source_size = os.stat(source_path).st_size
parallel_processes = (multiprocessing.cpu_count() * 2) + 1
conn = boto.connect_s3(aws_key, aws_secret)
bucket = conn.get_bucket(bucketname)
mtype = 'application/zip, application/octet-stream'
headers.update({'Content-Type': mtype})
mp = bucket.initiate_multipart_upload(keyname, headers=headers)
bytes_per_chunk = max(int(math.sqrt(5242880) * math.sqrt(source_size)),
5242880)
chunk_amount = int(math.ceil(source_size / float(bytes_per_chunk)))
pool = Pool(processes=parallel_processes)
for i in range(chunk_amount):
offset = i * bytes_per_chunk
remaining_bytes = source_size - offset
bytes = min([bytes_per_chunk, remaining_bytes])
part_num = i + 1
pool.apply_async(_upload_part, [bucketname, aws_key, aws_secret, mp.id,
part_num, source_path, offset, bytes])
pool.close()
pool.join()
if len(mp.get_all_parts()) == chunk_amount:
mp.complete_upload()
else:
mp.cancel_upload()
os.unlink(tempInFile.name)
_logger.info('Data successfully backed up to s3')
def process(configuration: LogParserConfig, db: DatabaseConnection, process_count: int):
influxdb_client = get_client(configuration)
pool = Pool(process_count)
currently_processing = defaultdict(list)
dl_dir = os.path.join(db.root_dir, 'data')
while True:
logging.info('Checking for new files to process')
settings = db.get_settings()
new_files = get_new_files_to_process(configuration.buckets, settings)
settings.files.extend(new_files)
timestamp = int(time.time())
completed_files = defaultdict(list)
# Empty queue of finished work and create a lists of all completed files per bucket
while True:
if finished_queue.empty():
break
success, bucket, filename = finished_queue.get()
currently_processing[bucket].remove(filename)
if success:
completed_files[bucket].append(filename)
# Set processed timestamp on processed files
for file in settings.files:
if file.name in completed_files[file.bucket]:
file.processed_timestamp = timestamp
logging.info('%d files completed processing since last loop', sum(len(l) for l in completed_files.values()))
completed_files.clear()
db.save_settings(settings)
added = 0
for file in settings.files:
if file.processed_timestamp is None and file.name not in currently_processing[file.bucket]:
currently_processing[file.bucket].append(file.name)
pool.apply_async(process_file, (file.bucket, file.name, dl_dir, influxdb_client), {}, after_processed,
after_error)
added += 1
processing_count = sum(len(l) for l in currently_processing.values())
if added:
logging.info('Added %s files to pool, %s files currently processing.', added, processing_count)
else:
logging.info('Nothing new to process, sleeping %s seconds. %s files currently in queue to be processed.',
configuration.interval, processing_count)
time.sleep(configuration.interval)
continue
def start(self):
print 'Start Speech Download Service...'
language = self._language
download_dir = self.setup_download_dir()
with open(self._text_file_path, 'r') as f:
# python 2.7 not supporting context manager with Pool()
pool = Pool(processes=8)
progress = Progress(len(f.readlines()))
f.seek(0)
results = []
def update_progress(result):
results.append(result)
progress.update(len(results))
for text in f:
text = text.rstrip('\n')
encoded_args = urlencode({
'hl': language,
'src': text,
'key': config.API_KEY
})
url = config.SPEECH_URL + encoded_args
download_path = os.path.join(download_dir, text + config.DOWNLOAD_FILE_TYPE)
downloader = SpeechDownloader(url, download_path)
# Pool().apply() blocks until the process is finished
# pool.apply(downloader.download)
pool.apply_async(downloader.download, callback=update_progress)
# Prevents any more tasks from being submitted to the pool
pool.close()
# Wait for the worker process to exit
pool.join()
print 'Done, Downloaded %d Speeches' % len(results)
def create_dictionary(filelist, output, tagmap, delimeter = '/'):
"""
Create a dictionary out of slashtag-based files.
:param filelist: List of file paths
:type filelist: list[str]
:param output: output file path
:type output: str
:param tagmap: Optional
:type tagmap: TagMap
"""
c = POSEvalDict()
counts = {'tokens':0, 'lines':0}
def merge_counts(result):
d, cur_tokencount, cur_linecount = result
c.combine(d)
counts['tokens'] += cur_tokencount
counts['lines'] += cur_linecount
tm = None
if tagmap:
tm = TagMap(tagmap)
# Initialize multithreading...
p = Pool(cpu_count())
for path in filelist:
p.apply_async(process_file, args=[path, tm, delimeter], callback=merge_counts)
# result = p.apply(process_file, args=[path, tm, delimeter])
# merge_counts(result)
p.close()
p.join()
# Now, dump the pickled POSEvalDict.
print("Writing out dictionary...", end=' ')
pickle.dump(c, open(output, 'wb'))
print("Done.")
print("{} tokens processed, {} sentences.".format(counts['tokens'], counts['lines']))
class MultiProcPlugin(DistributedPluginBase):
"""Execute workflow with multiprocessing
The plugin_args input to run can be used to control the multiprocessing
execution. Currently supported options are:
- n_procs : number of processes to use
- non_daemon : boolean flag to execute as non-daemon processes
"""
def __init__(self, plugin_args=None):
super(MultiProcPlugin, self).__init__(plugin_args=plugin_args)
self._taskresult = {}
self._taskid = 0
n_procs = 1
non_daemon = False
if plugin_args:
if 'n_procs' in plugin_args:
n_procs = plugin_args['n_procs']
if 'non_daemon' in plugin_args:
non_daemon = plugin_args['non_daemon']
if non_daemon:
# run the execution using the non-daemon pool subclass
self.pool = NonDaemonPool(processes=n_procs)
else:
self.pool = Pool(processes=n_procs)
def _get_result(self, taskid):
if taskid not in self._taskresult:
raise RuntimeError('Multiproc task %d not found'%taskid)
if not self._taskresult[taskid].ready():
return None
return self._taskresult[taskid].get()
def _submit_job(self, node, updatehash=False):
self._taskid += 1
self._taskresult[self._taskid] = self.pool.apply_async(run_node,
(node, updatehash,))
return self._taskid
def _report_crash(self, node, result=None):
if result and result['traceback']:
node._result = result['result']
node._traceback = result['traceback']
return report_crash(node,
traceback=result['traceback'])
else:
return report_crash(node)
def _clear_task(self, taskid):
del self._taskresult[taskid]
def fetch_all(page_type, ids, output):
#Python multithreding mess incoming
start = time()
pages = []
links = build(page_type, ids)
#Number of worker processes to start
num_of_proc = 8
pool = Pool(processes=num_of_proc)
#Builds the process pool
results = [pool.apply_async(fetch, (url, page_type[1], output,)) for url in links]
#Proceeds to run the processes
for result in results:
result.get()
print("Fetched all in %s" %(time() - start))
def fetch_all(page_type, ids, output):
#Python multithreding mess incoming
start = time()
pages = []
links = build(page_type, ids)
#Number of worker processes to start
num_of_proc = 8
pool = Pool(processes=num_of_proc)
#Fetches pages asynchronously
results = [pool.apply_async(fetch, (url, page_type[1], output,)) for url in links]
#Appends the fetched pages into the pages list
for result in results:
result.get()
print("Fetched all in %s" %(time() - start))
def _save_content(self, key, content, headers):
print "key.name", key.name
source_size = os.stat(content.file.name).st_size
bytes_per_chunk = max(int(math.sqrt(5242880) * math.sqrt(source_size)), 5242880)
chunk_amount = int(math.ceil(source_size / float(bytes_per_chunk)))
mp = self.bucket.initiate_multipart_upload(key.name, headers=headers)
pool = Pool(processes=1)
for i in range(chunk_amount):
offset = i * bytes_per_chunk
remaining_bytes = source_size - offset
bytes = min([bytes_per_chunk, remaining_bytes])
part_num = i + 1
pool.apply_async(_upload_part, [self.bucket_name, self.access_key,
self.secret_key, mp.id, part_num,
content.file.name, offset, bytes])
pool.close()
pool.join()
if len(mp.get_all_parts()) == chunk_amount:
mp.complete_upload()
key = self.bucket.get_key(key.name)
else:
mp.cancel_upload()
def load_from_web():
print "Loading from Web"
movies = common.read_json(JSON_IN_FILE)
pool = Pool(5)
worker = [pool.apply_async(process_movie, [m]) for m in movies]
imdb_movies = []
for w in worker:
w.wait()
result = w.get()
if result is not None:
imdb_movies.append(w.get())
common.write_json(JSON_OUT_FILE, imdb_movies)
def multi_proc5(self, batch):
start_time = datetime.datetime.now()
sql = "select count(id) from records"
count_result = db_connection.execute(sql)
for row in count_result:
count = row[0]
break
sql = "select id from records"
result = db_connection.execute(sql)
record_ids = []
for idx, row in enumerate(result):
if (idx % int(count/4) == 0) or (idx == count - 1): #4 because that is how many workers we have
if idx == 0:
some_records = []
else:
record_ids.append(some_records)
some_records = []
some_records.append(row[0])
input_pool = Pool(4)
#Add id messages to input queue
input_pool.map(partial(add_batch_ids_to_queue, batch_size=int(batch)), record_ids)
input_pool.close()
input_pool.join()
output_pool = Pool(4)
#Read ids from input_queue, read message from DB and write it to output_queue
worker_results = []
for i in range(4):
worker_results.append(output_pool.apply_async(read_id_from_queue, ()))
output_pool.close()
for r in worker_results:
r.get() # This reports results, including errors, of workers
output_pool.join() # This blocks until all the processes have finished
end_time = datetime.datetime.now()
time_taken = (end_time - start_time).total_seconds()
return time_taken
def create_training_parallel(count):
pool_size = 8
batch_count = pool_size * 5
pool = Pool(pool_size)
print("generating")
results = []
for i in range(batch_count):
results.append(pool.apply_async(create_training_data, (count/batch_count,)))
pool.close()
pool.join()
print("concatenating")
output = []
for r in results:
output.extend(r.get(1000))
return output
def multi_proc3(self, batch):
start_time = datetime.datetime.now()
sql = "select count(id) from records"
count_result = db_connection.execute(sql)
for row in count_result:
count = row[0]
break
sql = "select id from records"
result = db_connection.execute(sql)
record_ids = []
for idx, row in enumerate(result):
if (idx % int(batch) == 0) or (idx == count - 1):
if idx == 0:
some_records = []
else:
record_ids.append(some_records)
some_records = []
some_records.append(row[0])
#Add id messages to input queue
msg_handler = MessageHandler()
for records in record_ids:
msg_handler.add_message(json.dumps({"ids":records}), "input_queue")
worker_results = []
p = Pool(4)
for i in range(4):
worker_results.append(p.apply_async(read_id_from_queue, ()))
p.close()
for r in worker_results:
r.get()
p.join() # This blocks until all the processes have finished
end_time = datetime.datetime.now()
time_taken = (end_time - start_time).total_seconds()
return time_taken
def multi_proc4(self, batch):
start_time = datetime.datetime.now()
sql = "select count(id) from records"
count_result = db_connection.execute(sql)
for row in count_result:
count = row[0]
break
sql = "select id from records"
result = db_connection.execute(sql)
record_ids = []
for idx, row in enumerate(result):
if (idx % int(batch) == 0) or (idx == count - 1):
if idx == 0:
some_records = []
else:
record_ids.append(some_records)
some_records = []
some_records.append(row[0])
p = Pool(4)
#Add id messages to input queue
p.map(add_ids_to_queue, record_ids)
#Read ids from input_queue, read message from DB and write it to output_queue
worker_results = []
p = Pool(4)
for i in range(4):
worker_results.append(p.apply_async(read_id_from_queue, ()))
p.close()
for r in worker_results:
r.get()
p.join() # This blocks until all the processes have finished
end_time = datetime.datetime.now()
time_taken = (end_time - start_time).total_seconds()
return time_taken
def render(self, scene, bucket_order_info: BucketOrderInfo, multiThread: bool=True):
self.scene = scene
if self.surface_integrator is not None:
self.surface_integrator.Preprocess(scene, self.camera, self)
if self.volume_integrator is not None:
self.volume_integrator.Preprocess(scene, self.camera, self)
sample = Sample(self.main_sampler, self.surface_integrator, self.volume_integrator, scene)
if multiThread:
my_bucket_orders = BucketOrder.create(bucket_order_info.width, bucket_order_info.height,
bucket_order_info.bucket_order_type)
pool = Pool(processes=multiprocessing.cpu_count())
# pool = Pool(processes=1)
pool._wrap_exception = False
results = []
for i in range(bucket_order_info.width * bucket_order_info.height):
# for i in range(55, 56):
a = pool.apply_async(self.render_task, args=(
i, my_bucket_orders.buckets_orders[i], bucket_order_info, sample, self),
callback=self.draw)
results.append(a)
for r in results:
r.wait()
else:
bucketOrderInfo = BucketOrderInfo(BucketOrderSortType.Random, 1, 1)
self.render_task(0, 0, bucketOrderInfo, sample, self)
print("Render end")
data = write_png(self.camera.film.data, self.camera.film.width, self.camera.film.height)
with open("my_image.png", 'wb') as fd:
fd.write(data)
def fill_views(self):
paths = []
if self.view:
for path in self.view_paths:
if self.view in path:
paths.append(path)
break
else:
paths = self.view_paths
pool = Pool(processes=len(paths))
pool_results = []
prescribing_date = ImportLog.objects.latest_in_category(
'prescribing').current_at.strftime('%Y-%m-%d')
for view in paths:
if self.view and self.view not in view:
continue
# Perform bigquery parts of operation in parallel
result = pool.apply_async(
query_and_export, [self.dataset, view, prescribing_date])
pool_results.append(result)
pool.close()
pool.join() # wait for all worker processes to exit
for result in pool_results:
tablename, gcs_uri = result.get()
f = download_and_unzip(gcs_uri)
copy_str = "COPY %s(%s) FROM STDIN "
copy_str += "WITH (FORMAT CSV)"
fieldnames = f.readline().split(',')
with connection.cursor() as cursor:
with utils.constraint_and_index_reconstructor(tablename):
self.log("Deleting from table...")
cursor.execute("DELETE FROM %s" % tablename)
self.log("Copying CSV to postgres...")
try:
cursor.copy_expert(copy_str % (
tablename, ','.join(fieldnames)), f)
except Exception:
import shutil
shutil.copyfile(f.name, "/tmp/error")
raise
f.close()
self.log("-------------")
def stat_first_tgs(stime,etime):
begtime = long(time.mktime(stime.timetuple())*1000)
endtime = long(time.mktime(etime.timetuple())*1000)
tgsinfo = read_tgs_info()
vehicles = {}
test = random.sample(tgsinfo.keys(), 100)
from multiprocessing.pool import Pool
pool = Pool()
result = [pool.apply_async(_stat_first_tgs_single, args=(int(cid),begtime,endtime)) for cid in tgsinfo.keys()]
pool.close()
pool.join()
result1 = [elem.get() for elem in result]
print 'joining....'
result2 = reduce(_combine, result1)
print 'totally %d vehicles. ' % (len(result2))
# print type(result2)
c = Counter()
for veh, info in result2.iteritems():
c[info[1]] += 1
print 'writing result into file...'
with open(os.path.join(root_dir,"result","first_tgs.txt"),"w") as f:
c1 = c.most_common()
# print c1[0]
# for cid,count in c1.iteritems():
for elem in c1:
line = "%5d,%6d\n" % (elem[0],elem[1])
f.write(line)
print 'finished.'
def __test_multi_processed(self, test_set, method, model, threshold, trees, all_node_ids):
"""
Create a process pool to distribute the prediction.
"""
#process_count = multiprocessing.cpu_count()
process_count = 4
pool = Pool(processes=process_count)
step = int(math.ceil(float(len(test_set)) / process_count))
results = []
for j in range(0, len(test_set), step):
meme_ids = test_set[j: j + step]
res = pool.apply_async(test_meme,
(meme_ids, method, model, threshold, trees, all_node_ids, self.user_ids,
self.users_map, self.verbosity))
results.append(res)
pool.close()
pool.join()
prp1_list = []
prp2_list = []
precisions = []
recalls = []
fprs = []
f1s = []
# Collect results of the processes.
for res in results:
r1, r2, r3, r4, r5, r6 = res.get()
precisions.extend(r1)
recalls.extend(r2)
fprs.extend(r3)
recalls.extend(r4)
prp1_list.extend(r5)
prp2_list.extend(r6)
return precisions, recalls, f1s, fprs, prp1_list, prp2_list
