def create_threads(detector):
config = configparser.ConfigParser()
config.read(FACE_DETEC_CONF)
video_decoders = []
for item in config['videostream']:
preprocesser = Preprocess(config['videostream'][item],
len(video_decoders), MODEL_WIDTH,
MODEL_HEIGHT)
video_decoders.append(preprocesser)
rtsp_num = len(video_decoders)
if rtsp_num == 0:
log_error("No video stream name or addr configuration in ",
FACE_DETEC_CONF)
return None, None
postprocessor = Postprocess(detector)
display_channel = int(config['display']['channel'])
if (display_channel is None) or (display_channel >= rtsp_num):
log_info("No video to display, display configuration: ",
config['display']['channel'])
else:
video_decoders[display_channel].set_display(True)
ret = postprocessor.create_presenter_channel(FACE_DETEC_CONF)
if ret == False:
log_error("Create presenter channel failed")
return None, None
return video_decoders, postprocessor
def run_single_block(input_list):
baseName,model,out_dir,rst,window,keep_temp,threshold,penality,DB_file,input_file,chromosome,strand,fa_file,depth,start,end = input_list
print("Generating blocks ...%s %d %s"%(baseName,start,end))
####Generate sliding windlows
gw_start_time = datetime.datetime.now()
block = split_chr_bedGraph2(out_dir,input_file,chromosome,strand,window,fa_file,depth,start,end)
#ww = open(baseName,'w')
#for a,b,c in block:
# ww.write('%s\t%s\t%s\n'%(a,b,c))
#ww.close()
gw_end_time = datetime.datetime.now()
print("Generate blocks used time: {}".format(gw_end_time - gw_start_time))
print("Evaluating blocks ...%s %d %s"%(baseName,start,end))
gw_start_time = datetime.datetime.now()
Evaluate(baseName,block,model,out_dir,rst,window,keep_temp)
gw_end_time = datetime.datetime.now()
print("Evaluated blocks used time: {}".format(gw_end_time - gw_start_time))
Scan_Forward(baseName,threshold,penality,out_dir)
Scan_Backward(baseName,threshold,penality,out_dir)
if(keep_temp != 'yes'):
predict_file = out_dir+'/predict/'+baseName+'.txt'
os.system('rm %s'%predict_file)
Postprocess(DB_file,baseName,threshold,penality,out_dir)
if(keep_temp != 'yes'):
forward_file=out_dir+"/maxSum/%s.forward.%d.%d.txt"%(baseName,threshold,penality)
backward_file=out_dir+"/maxSum/%s.backward.%d.%d.txt"%(baseName,threshold,penality)
os.system('rm %s %s'%(forward_file,backward_file))
print('Finished postprocessing...\n')
def run_single_block(input_list):
data = input_list[0]
model = input_list[1]
out_dir = input_list[2]
rst = input_list[3]
window = input_list[4]
keep_temp = input_list[5]
threshold = input_list[6]
penality = input_list[7]
DB_file = input_list[8]
if 'wig' not in data:
baseName = data.split('/')[-1]
Evaluate(model, out_dir, rst, window, baseName, keep_temp)
Scan_Forward(baseName, threshold, penality, out_dir)
Scan_Backward(baseName, threshold, penality, out_dir)
if (keep_temp != 'yes'):
predict_file = out_dir + '/predict/' + baseName + '.txt'
os.system('rm %s' % predict_file)
Postprocess(DB_file, baseName, threshold, penality, out_dir)
if (keep_temp != 'yes'):
forward_file = out_dir + "/maxSum/%s.forward.%d.%d.txt" % (
baseName, threshold, penality)
backward_file = out_dir + "/maxSum/%s.backward.%d.%d.txt" % (
baseName, threshold, penality)
os.system('rm %s %s' % (forward_file, backward_file))
print('Finished postprocessing...\n')
def run_single_block(input_list):
#global ref
#print(ref.keys())
baseName, model, out_dir, rst, window, keep_temp, threshold, penality, DB_file, input_file, fa_file, chromosome, strand, depth, start, end = input_list
#log_dir = out_dir+'/log'
#if not os.path.exists(log_dir):
# os.makedirs(log_dir)
#log.basicConfig(filename='%s/%s.log'%(log_dir,baseName), level=log.INFO)
print("Generating blocks ...%s %d %s" % (baseName, start, end))
####Generate sliding windlows
gw_start_time = datetime.datetime.now()
block = split_chr_bedGraph2(out_dir, input_file, chromosome, strand,
window, fa_file, depth, start, end)
#ww = open(baseName,'w')
#for a,b,c in block:
# ww.write('%s\t%s\t%s\n'%(a,b,c))
#ww.close()
gw_end_time = datetime.datetime.now()
print("Generate blocks used time: {}\n".format(gw_end_time -
gw_start_time))
print("Evaluating blocks ...%s %d %s" % (baseName, start, end))
ev_start_time = datetime.datetime.now()
Evaluate(baseName, block, model, out_dir, rst, window, keep_temp)
del block #destroyed the block reference
gc.collect() #manually run garbage collection process
ev_end_time = datetime.datetime.now()
print("Evaluated blocks used time: {}\n".format(ev_end_time -
ev_start_time))
print("Postprocessing blocks ...%s %d %s" % (baseName, start, end))
ps_start_time = datetime.datetime.now()
Scan_Forward(baseName, threshold, penality, out_dir)
Scan_Backward(baseName, threshold, penality, out_dir)
if (keep_temp != 'yes'):
predict_file = out_dir + '/predict/' + baseName + '.txt'
os.system('rm %s' % predict_file)
Postprocess(DB_file, baseName, threshold, penality, out_dir)
ps_end_time = datetime.datetime.now()
print("Postprocessed blocks used time: {}\n".format(ps_end_time -
ps_start_time))
if (keep_temp != 'yes'):
forward_file = out_dir + "/maxSum/%s.forward.%d.%d.txt" % (
baseName, threshold, penality)
backward_file = out_dir + "/maxSum/%s.backward.%d.%d.txt" % (
baseName, threshold, penality)
os.system('rm %s %s' % (forward_file, backward_file))
#print('Finished postprocessing...%s\n'%baseName)
return [
gw_end_time - gw_start_time, ev_end_time - ev_start_time,
ps_end_time - ps_start_time
]
def run_single_block(input_list):
baseName, model, out_dir, rst, window, keep_temp, threshold, penality, DB_file, input_file, strand, depth, block_pos = input_list
chromosome = block_pos[3]
start = block_pos[4]
end = block_pos[5]
print("Generating blocks ...%s %d %s" %
(baseName, block_pos[4], block_pos[5]))
####Generate sliding windlows
gw_start_time = datetime.datetime.now()
blocks = Bedgraph_to_blocks(input_file, fa_file, window, depth, block_pos)
#ww = open(baseName,'w')
#for a,b,c in blocks:
# ww.write('%s\t%s\t%s\n'%(a,b,c))
#ww.close()
gw_end_time = datetime.datetime.now()
print("Generate blocks used time: {}\n".format(gw_end_time -
gw_start_time))
print("Evaluating blocks ...%s %d %s" % (baseName, start, end))
ev_start_time = datetime.datetime.now()
Evaluate(baseName, blocks, model, out_dir, rst, window, keep_temp)
del blocks #destroyed the block reference
gc.collect() #manually run garbage collection process
ev_end_time = datetime.datetime.now()
print("Evaluated blocks used time: {}\n".format(ev_end_time -
ev_start_time))
print("Postprocessing blocks ...%s %d %s" % (baseName, start, end))
ps_start_time = datetime.datetime.now()
Scan_Forward(baseName, threshold, penality, out_dir)
Scan_Backward(baseName, threshold, penality, out_dir)
if (keep_temp != 'yes'):
predict_file = out_dir + '/predict/' + baseName + '.txt'
os.system('rm %s' % predict_file)
Postprocess(DB_file, baseName, threshold, penality, out_dir)
ps_end_time = datetime.datetime.now()
print("Postprocessed blocks used time: {}\n".format(ps_end_time -
ps_start_time))
if (keep_temp != 'yes'):
forward_file = out_dir + "/maxSum/%s.forward.%d.%d.txt" % (
baseName, threshold, penality)
backward_file = out_dir + "/maxSum/%s.backward.%d.%d.txt" % (
baseName, threshold, penality)
os.system('rm %s %s' % (forward_file, backward_file))
#print('Finished postprocessing...%s\n'%baseName)
return [
gw_end_time - gw_start_time, ev_end_time - ev_start_time,
ps_end_time - ps_start_time
]
def create_threads(detector):
"""create threads"""
config = configparser.ConfigParser()
config.read(COCO_DETEC_CONF)
video_decoders = []
for item in config['videostream']:
preprocesser = Preprocess(config['videostream'][item],
len(video_decoders), MODEL_WIDTH,
MODEL_HEIGHT)
video_decoders.append(preprocesser)
rtsp_num = len(video_decoders)
if rtsp_num == 0:
log_error("No video stream name or addr configuration in ",
COCO_DETEC_CONF)
return None, None
postprocessor = Postprocess(detector)
return video_decoders, postprocessor
def main(out_dir, input_file, input_plus, input_minus, fa_file, keep_temp,
window, name, model, rst, threshold, penality, DB_file):
if not os.path.exists(out_dir):
os.makedirs(out_dir)
out_dir = out_dir + '/' + name
####Generate sliding windlows
Generate_windows(out_dir, input_file, input_plus, input_minus, fa_file,
keep_temp, window, name)
data_dir = out_dir + '/data'
data_files = glob.glob(data_dir + "/*")
for data in data_files:
if 'wig' in data:
continue
baseName = data.split('/')[-1]
Evaluate(model, out_dir, rst, window, baseName, keep_temp)
Scan_Forward(baseName, threshold, penality, out_dir)
Scan_Backward(baseName, threshold, penality, out_dir)
if (keep_temp != 'yes'):
predict_file = out_dir + '/predict/' + baseName + '.txt'
os.system('rm %s' % predict_file)
Postprocess(DB_file, baseName, threshold, penality, out_dir)
if (keep_temp != 'yes'):
forward_file = out_dir + "/maxSum/%s.forward.%d.%d.txt" % (
baseName, threshold, penality)
backward_file = out_dir + "/maxSum/%s.backward.%d.%d.txt" % (
baseName, threshold, penality)
os.system('rm %s %s' % (forward_file, backward_file))
out_file = '%s/%s.predicted.txt' % (out_dir, name)
ww = open(out_file, 'w')
ww.write('predicted_pasid\tdb_diff\tdb_pasid\tscore\n')
ww.close()
os.system('cat %s/maxSum/*bidirection* >>%s' % (out_dir, out_file))
if (keep_temp != 'yes'):
os.system('rm -rf %s/data %s/predict %s/maxSum' %
(out_dir, out_dir, out_dir))
print("Job Done!")
outputFile.write('\t'.join(outputLine) + '\n')
outputFile.write('\n')
sentence = []
tags = []
continue
else:
token, label = line.split('\t')
sentence.append(token)
tags.append(label)
if __name__ == '__main__':
crf_path = 'E:/data/CRF++-0.54/'
crf_data_path = 'crf_input/'
para = 'trigger'
os.system(
'{}crf_learn -a MIRA -f 5 {}template.txt {}{para}_train.txt {para}_model'
.format(crf_path, crf_data_path, crf_data_path, para=para))
if para == 'trigger':
os.system('{}crf_test -m model -v2 {}{para}_test.txt > {para}_raw.txt'.
format(crf_path, crf_data_path, para=para))
Postprocess().processOutput('trigger_raw.txt', 'trigger_result.txt')
else:
os.system(
'{}crf_test -m {para}_model {}{para}_test.txt > {para}_result.txt'.
format(crf_path, crf_data_path, para=para))
img = cv2.imread('./data/VOC2007_trainval/JPEGImages/000017.jpg')
img = cv2.resize(img, (416, 416))
x = torch.from_numpy(img / 255).float().unsqueeze(0).permute(0, -1, 1,
2).cuda()
model = YoloV3(num_classes)
state = torch.load(os.path.join(saved_pth_dir, pth_file))
model.load_state_dict(state['model_state_dict'])
model.eval()
model.cuda()
y_pred = model(x, training=False)
postprocessor = Postprocess(iou_threshold=0.5,
score_threshold=0.5,
max_detection=10).cuda()
boxes, scores, classes, num_detection = postprocessor(y_pred)
num_img = num_detection.shape[0]
for n_img in range(num_img):
h, w, c = img.shape
score = scores.cpu()[n_img]
classs = classes.cpu()[n_img]
classs = torch.argmax(classs, dim=1)
for i in range(num_detection.item()):
box = boxes.cpu()[n_img][i]
score_value = score[i].item()
def VGGish(pump=None,
input_shape=None,
include_top=False,
pooling='avg',
weights='audioset',
name='vggish',
compress=False):
'''A Keras implementation of the VGGish architecture.
Arguments:
pump (pumpp.Pump): The model pump object.
input_shape (tuple): the model input shape. If ``include_top``,
``input_shape`` will be set to ``(params.NUM_FRAMES, params.NUM_BANDS, 1)``,
otherwise it will be ``(None, None, 1)`` to accomodate variable sized
inputs.
include_top (bool): whether to include the fully connected layers. Default is False.
pooling (str): what type of global pooling should be applied if no top? Default is 'avg'
weights (str, None): the weights to use (see WEIGHTS_PATHS). Currently, there is
only 'audioset'. Can also be a path to a keras weights file.
name (str): the name for the model.
Returns:
A Keras model instance.
'''
with tf.name_scope(name):
if input_shape:
pass
elif include_top:
input_shape = params.NUM_FRAMES, params.NUM_BANDS, 1
elif pump:
# print(type(pump))
inputs = pump.layers('tf.keras')[params.PUMP_INPUT]
else:
input_shape = None, None, 1
# use input_shape to make input
if input_shape:
inputs = tfkl.Input(shape=input_shape, name='input_1')
# setup layer params
conv = partial(
tfkl.Conv2D,
kernel_size=(3, 3), strides=(1, 1), activation='relu', padding='same')
maxpool = partial(
tfkl.MaxPooling2D, pool_size=(2, 2), strides=(2, 2), padding='same')
# Block 1
x = conv(64, name='conv1')(inputs)
x = maxpool(name='pool1')(x)
# Block 2
x = conv(128, name='conv2')(x)
x = maxpool(name='pool2')(x)
# Block 3
x = conv(256, name='conv3/conv3_1')(x)
x = conv(256, name='conv3/conv3_2')(x)
x = maxpool(name='pool3')(x)
# Block 4
x = conv(512, name='conv4/conv4_1')(x)
x = conv(512, name='conv4/conv4_2')(x)
x = maxpool(name='pool4')(x)
if include_top:
dense = partial(tfkl.Dense, activation='relu')
# FC block
x = tfkl.Flatten(name='flatten_')(x)
x = dense(4096, name='fc1/fc1_1')(x)
x = dense(4096, name='fc1/fc1_2')(x)
x = dense(params.EMBEDDING_SIZE, name='fc2')(x)
if compress:
x = Postprocess()(x)
else:
globalpool = (
tfkl.GlobalAveragePooling2D() if pooling == 'avg' else
tfkl.GlobalMaxPooling2D() if pooling == 'max' else None)
if globalpool:
x = globalpool(x)
# Create model
model = Model(inputs, x, name='model')
# print(model.summary())
model.load_weights('/scratch/mkolpe2s/CNN_dcase/vggish_audioset_weights_without_fc2.h5')
# load_vggish_weights(model, weights, strict=bool(weights))
# print("Okay")
return model
from crawler import Crawler
from preprocess_purchase import PreprocessPurchase
from inference import Inference
from learning_purchase import Learning
from postprocess import Postprocess, Email
from datetime import datetime
data_version = datetime.now().strftime("%Y%m%d")
load_date = data_version
target = 'purchase'
mode = 'inference'
encoder_version = 'original'
model_version = 'original'
#crawling
crawler = Crawler(data_version, target)
#preprocess
prep = PreprocessPurchase(mode, data_version, encoder_version)
#inference
inf = Inference(data_version, model_version)
#postprocess
post = Postprocess(data_version, model_version)
body = post.urls()
em = Email()
msg = em.create_message(body)
em.send(msg)
def setup(self):
self.worldNP = render.attachNewNode('World')
# World
self.debugNP = self.worldNP.attachNewNode(BulletDebugNode('Debug'))
self.debugNP.hide()
self.world = BulletWorld()
self.world.setGravity(Vec3(0, 0, -9.81))
self.world.setDebugNode(self.debugNP.node())
# Plane
#shape = BulletPlaneShape(Vec3(0, 0, 1), 0)
#mesh = BulletTriangleMesh()
#geomNodes = loader.loadModel('levels/test1/collision').findAllMatches('**/+GeomNode')
#geomNode = geomNodes.getPath(0).node()
#geom = geomNode.getGeom(0)
#mesh.addGeom(geom)
#shape = BulletTriangleMeshShape(mesh, dynamic=False, bvh=True )
#np = self.worldNP.attachNewNode(BulletRigidBodyNode('Ground'))
#np.node().addShape(shape)
#np.setPos(0, 0, 20.0)
#np.setCollideMask(BitMask32.allOn())
#self.world.attachRigidBody(np.node())
#sky dome
self.sun_sky = Sky()
self.sun_sky.setTime(17.0)
#terrain
self.ground = Terrain(self.world, self.worldNP)
self.ground.loadMesh(path + 'levels/gandg2/collision')
self.ground.setMaps(path + 'levels/gandg2/')
self.ground.setTextures((39, 1, 2, 15, 4, 5))
#grass
self.grass = Grass()
self.grass.setMap(path + 'levels/gandg2/grass.png')
self.grass_to_cut = self.grass.getStatus()
# Car
self.car = Car(self.world, self.worldNP)
self.car.setPos(161.0, 160.0, 26)
#camera
self.camera = FlyingCamera()
#car to character scale 0.0128
self.char = Character(self.world, self.worldNP)
self.char.enterCar()
#self.char.setPos(256, 250, 80)
#filter manager, post process
self.filters = Postprocess()
#self.filters.setupFxaa()
#no time to make it work, sorry...
self.filters.setupFilters()
#map objects .. hardcoded because of time
self.object_root = render.attachNewNode('object_root')
obj = [(path + 'models/pyweek_wall1', 0.0, (303.0, 405.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (301.0, 405.0,
25.0980434417725),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (299.0, 405.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (297.0, 405.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (295.0, 405.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (293.0, 405.0,
25.0980434417725),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (291.0, 405.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (289.0, 405.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (287.0, 405.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (285.0, 405.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (283.0, 405.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (281.0, 405.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (281.0, 385.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (283.0, 385.0,
25.0980453491211),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (285.0, 385.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (304.0, 404.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (304.0, 402.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (304.0, 400.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (304.0, 398.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (304.0, 396.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (304.0, 394.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (304.0, 392.0,
25.237850189209),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (280.0, 404.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (280.0, 398.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (280.0, 396.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (280.0, 394.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (280.0, 392.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (280.0, 390.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (280.0, 388.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (280.0, 386.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (286.0, 386.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (286.0, 388.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (286.0, 390.0,
25.0980434417725),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (287.0, 391.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (289.0, 391.0,
25.1190624237061),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (291.0, 391.0,
25.1960334777832),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (293.0, 391.0,
25.1596641540527),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (295.0, 391.0,
25.2697868347168),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (297.0, 391.0,
25.3282146453857),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (299.0, 391.0,
25.3496627807617),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (301.0, 391.0,
25.2688617706299),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 0.0, (303.0, 391.0,
25.2534332275391),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_wall1', 90.0, (280.0, 402.0,
25.0980415344238),
path + 'models/pyweek_wall1_collision'),
(path + 'models/pyweek_box', 0.0,
(279.600006103516, 401.700012207031, 25.0980415344238), None),
(path + 'models/pyweek_box', 0.0,
(279.399993896484, 402.200012207031, 25.0980415344238), None),
(path + 'models/pyweek_box', 0.0,
(279.600006103516, 402.700012207031, 25.0980415344238), None),
(path + 'models/pyweek_box', 0.0,
(279.399993896484, 403.399993896484, 25.0980415344238), None),
(path + 'models/pyweek_box', 0.0,
(278.799987792969, 402.799987792969, 25.0980415344238), None),
(path + 'models/pyweek_box', 0.0,
(278.799987792969, 402.100006103516, 25.0980415344238), None),
(path + 'models/pyweek_box', 0.0, (279.0, 401.5,
25.0980415344238), None),
(path + 'models/pyweek_box', 0.0, (278.5, 401.600006103516,
25.0980415344238), None),
(path + 'models/pyweek_box', 0.0,
(278.799987792969, 401.899993896484, 25.5980415344238), None),
(path + 'models/pyweek_box', 90.0, (279.5, 402.600006103516,
25.5980415344238), None),
(path + 'models/pyweek_box', 90.0, (279.0, 402.5,
25.5980415344238), None),
(path + 'models/pyweek_box', 90.0, (279.399993896484, 402.0,
25.5980415344238), None),
(path + 'models/pyweek_box', 90.0,
(278.100006103516, 402.299987792969, 25.0980415344238), None),
(path + 'models/pyweek_box', 90.0,
(277.799987792969, 401.700012207031, 25.0980415344238), None),
(path + 'models/pyweek_box', 90.0,
(278.200012207031, 401.899993896484, 25.5980415344238), None),
(path + 'models/pyweek_box', 90.0,
(279.399993896484, 402.399993896484, 26.0980415344238), None),
(path + 'models/pyweek_box', 90.0, (279.0, 401.899993896484,
26.0980415344238), None),
(path + 'models/pyweek_box', 90.0,
(278.799987792969, 402.399993896484, 26.0980415344238), None)]
for i in obj:
loadObject(model=i[0],
H=i[1],
pos=i[2],
world=self.world,
worldNP=self.worldNP,
root=self.object_root,
collision_solid=i[3])
self.object_root.flattenStrong()
#models/pyweek_gate,90.0,(280.0,399.0,25.0980415344238))
#gui
self.hud = HUD()
#volume
sfxMgr = base.sfxManagerList[0]
sfxMgr.setVolume(cfg['sound-volume'] * 0.01)
musicMgr = base.musicManager
musicMgr.setVolume(cfg['music-volume'] * 0.01)
#music
self.driving_music = loader.loadMusic(path + 'music/driving.ogg')
self.driving_music.setLoop(True)
self.driving_music.play()
self.walking_music = loader.loadMusic(path + 'music/walking.ogg')
self.walking_music.setLoop(True)
print self.char.actor.getScale(render)
Generate_windows(out_dir,input_file,input_plus,input_minus,fa_file,keep_temp,window,name,depth)
>>>>>>> e6e9d5b305368adbd0eec6e055e70e82e740a154
data_dir = out_dir+'/data'
data_files = glob.glob(data_dir+"/*")
for data in data_files:
if 'wig' in data:
continue
baseName = data.split('/')[-1]
Evaluate(model,out_dir,rst,window,baseName,keep_temp)
Scan_Forward(baseName,threshold,penality,out_dir)
Scan_Backward(baseName,threshold,penality,out_dir)
if(keep_temp != 'yes'):
predict_file = out_dir+'/predict/'+baseName+'.txt'
os.system('rm %s'%predict_file)
Postprocess(DB_file,baseName,threshold,penality,out_dir)
if(keep_temp != 'yes'):
forward_file=out_dir+"/maxSum/%s.forward.%d.%d.txt"%(baseName,threshold,penality)
backward_file=out_dir+"/maxSum/%s.backward.%d.%d.txt"%(baseName,threshold,penality)
os.system('rm %s %s'%(forward_file,backward_file))
out_file = '%s/%s.predicted.txt' %(out_dir,name)
ww = open(out_file,'w')
<<<<<<< HEAD
ww.write('predicted_pasid\tdb_diff\tdb_pasid\tscore\n')
=======
if(DB_file is not None):
ww.write('predicted_pasid\tdb_pasid\tdb_diff\tscore\n')
else:
ww.write('predicted_pasid\tscore\n')
>>>>>>> e6e9d5b305368adbd0eec6e055e70e82e740a154
def VGGish(pump=None,
input_shape=None,
include_top=False,
pooling='avg',
weights='audioset',
name='vggish',
compress=False):
with tf.name_scope(name):
if input_shape:
pass
elif pump:
inputs = pump.layers('tf.keras')[params.PUMP_INPUT]
elif include_top:
input_shape = params.NUM_FRAMES, params.NUM_BANDS, 1
else:
input_shape = None, None, 1
# use input_shape to make input
if input_shape:
inputs = kl.Input(shape=input_shape, name='input_1')
# setup layer params
conv = partial(
kl.Conv2D,
kernel_size=(3, 3), strides=(1, 1), activation='relu', padding='same')
maxpool = partial(
kl.MaxPooling2D, pool_size=(2, 2), strides=(2, 2), padding='same')
# Block 1
x = conv(64, name='conv1')(inputs)
x = maxpool(name='pool1')(x)
# Block 2
x = conv(128, name='conv2')(x)
x = maxpool(name='pool2')(x)
# Block 3
x = conv(256, name='conv3/conv3_1')(x)
x = conv(256, name='conv3/conv3_2')(x)
x = maxpool(name='pool3')(x)
# Block 4
x = conv(512, name='conv4/conv4_1')(x)
x = conv(512, name='conv4/conv4_2')(x)
x = maxpool(name='pool4')(x)
if include_top:
dense = partial(kl.Dense, activation='relu')
# FC block
x = kl.Flatten(name='flatten_')(x)
x = dense(4096, name='fc1/fc1_1')(x)
x = dense(4096, name='fc1/fc1_2')(x)
x = dense(params.EMBEDDING_SIZE, name='fc2')(x)
if compress:
x = Postprocess()(x)
else:
globalpool = (
kl.GlobalAveragePooling2D() if pooling == 'avg' else
kl.GlobalMaxPooling2D() if pooling == 'max' else None)
if globalpool:
x = globalpool(x)
# Create model
model = Model(inputs, x, name='model')
load_vggish_weights(model, weights, strict=bool(weights))
return model
def predict(self):
time_preprocess = TimeCalculation(name='Preprocess')
time_predict = TimeCalculation(name='Predict')
time_postprocess = TimeCalculation(name='Postprocess')
time_total = TimeCalculation(name='Total')
# 总计时开始
time_total.begin()
# 预处理计时开始
time_preprocess.begin()
# dicom 转换成 nii, 如果 nii_path 有值, 则默认不需要转换，直接去 nii_path 作为 origin_nii_path
print('------------>> step1: dicom2nii...')
origin_nii_path = self.nii_path if self.nii_path else dicom2nii(
dicom_path=self.dicom_path, save_path=self.output_path)
origin_nii = image.load_img(origin_nii_path)
print('------> origin nii shape: ' + str(origin_nii.get_data().shape))
# 预处理
print('------------>> step2: preprocrss...')
r_nii_data, main_axis, padding = PreProcess4MSpacing.preprocess(
origin_nii, self.standard_shape, normalize='standardiztion')
x_padding = padding[0]
y_padding = padding[1]
z_padding = padding[2]
# 预处理计时结束
time_preprocess.end()
# 预测计时开始
time_predict.begin()
# 载入模型
print('------------>> step3: load model...')
model = self.load_model()
# 开始预测
print('------------>> step4: start to predict...')
# 计算分块取数据需要的步骤数
step = int(math.ceil(r_nii_data.shape[0] / self.standard_shape[0]))
print('------> generator steps: ' + str(step))
pre_imgs = model.predict_generator(generator=GeneratorMSpacing(
normalize_func=PreProcess4MSpacing.standardiztion).pre_generator(
r_nii_data),
steps=step)
# 预测计时结束
time_predict.end()
# 后处理计时开始
time_postprocess.begin()
output_imgs = np.argmax(pre_imgs, axis=4)
# 巨坑：输出图像前把类型转换成float32,不然MRIcorN显示的时候会出现问题
output_imgs = output_imgs.astype(np.float32)
print('------> output image shape: ' + str(output_imgs.shape))
print('------> unique of output image: ' + str(np.unique(output_imgs)))
# 输出图像并保存
print('------------>> step5: stitching_predicted_data...')
stitching_data = self.stitching_predicted_data(output_imgs,
self.standard_shape)
# 还原 padding 的数据
print(stitching_data.shape)
stitching_data = stitching_data[:(
stitching_data.shape[0] -
x_padding[1]), :(stitching_data.shape[1] -
y_padding[1]), :(stitching_data.shape[2] -
z_padding[1])]
print('------> not padding output shape: ' + str(stitching_data.shape))
# 主轴还原
transpose_shape_dict = {0: (0, 1, 2), 1: (1, 0, 2), 2: (1, 2, 0)}
stitching_data = np.transpose(stitching_data,
transpose_shape_dict[main_axis])
print(np.unique(stitching_data))
predict_nii = np_2_nii(origin_nii_path,
stitching_data,
need_save=False)
print('------------>> step6: postprocess...')
Postprocess(predict_nii, origin_nii, origin_nii_path,
self.output_path).postprocess()
# 输出 nii 图像
print('------> output nii file...')
self.output_nii(predict_nii, origin_nii)
self.output_nii(predict_nii, origin_nii, 1)
self.output_nii(predict_nii, origin_nii, 2)
self.output_nii(predict_nii, origin_nii, 3)
# 计算 dice 值
self.calculate_dice(predict_nii)
# 后处理计时结束
time_postprocess.end()
# 总计时结束
time_total.end()
print('------------>> Done!')