def main():
print "------------------"
iops = Iops(True)
iops.start()
if DEL_CONTAINER_NAME:
for _del_str in DEL_CONTAINER_NAME:
del iops.hosts_container[_del_str]
#print iops.hosts_container
iops.linq_to_dict()
print "-------------------"
conn = iops.get_mysql_connection(DES_STORE_IP, DES_STORE_USER,
DES_STORE_PASSWORD, PORT)
if conn == None:
raise TypeErro(r'src_store_addr is not connect!')
pool = ThreadPool(processes=len(iops.req_data))
pool.daemon = True
for _info in iops.req_data:
_path = []
_path = [
'/container/stat/{0}/diskiops'.format(i).encode("utf-8")
for i in iops.req_data[_info].values()[0]
]
pool.apply_async(iops.request,
(iops.req_data[_info].keys()[0], _path, _info, conn))
pool.close()
pool.join()
if conn is not None:
conn.close()
def process_page(self, page_number):
results = self.github_search.page_result(page_number)
if not results:
return None
pool = Pool(MAX_THREADS)
for result in results:
pool.apply_async(self.search_content, (result, ))
pool.daemon = True
pool.close()
pool.join()
def load_batch_images(params, direction, x, patch_loc,map_loc):
#We should modify this function to load images with different number of channels
size=params["size"]
im_type=params["im_type"]
patch_use=params['patch_use']
sbt=1
if(im_type=="depth"):
normalizer=52492
sbt=params["depth_mean"]
if(im_type=="pre_depth"):
normalizer=52492
sbt=params["pre_depth_mean"]
if(im_type=="gray"):
normalizer=255
sbt=params["gray_mean"]
if(im_type=="rgb"):
normalizer=255
sbt=params["rgb_mean"]
if(im_type=="hha_depth_fc6"):
normalizer=47.2940864563
sbt=params["hha_depth_fc6_mean"]
if(im_type=="rgb_conv4_2"):
normalizer=8010.15771484
sbt=params["rgb_conv4_2_mean"]
if(im_type=="rgb_conv5_2"):
normalizer=1042.07519531
sbt=params["rgb_conv5_2_mean"]
im_order=0
if(direction=="S"):
im_order=1
map_arg=[(direction, im_type, normalizer, patch_loc,map_loc, patch_use, sbt, size, im[im_order]) for im in x]
pool_img = ThreadPool(params["n_procc"])
pool_img.daemon=True
results = pool_img.map(load_image_wrapper,map_arg)
pool_img.close()
pool_img.join()
batch_l=convert_set(results,im_type)
return numpy.array(batch_l)
