def main(cfg):
# TODO: implement device selection
# if config.device == "auto":
# device =
# elif config.device == "cpu":
# net = net.cpu()
# elif config.device == "gpu":
# net = net.cuda()
# else:q
# raise ValueError
if cfg.mode.startswith("train"):
run_train(cfg)
elif cfg.mode.startswith("test"):
# take the best model on validation set
cfg.pretrained_model = r'global_min_acer_model.pth'
run_test(cfg, dir='global_test_36_TTA')
elif cfg.mode == 'realtime':
cfg.pretrained_model = r'global_min_acer_model.pth'
run_realtime(cfg)
return
def test_raw_set(dataset, mode):
for item in dataset:
run_test(item['path'] + item['name'],
item['width'],
item['height'],
item['scale'],
item['radius'],
item['cellsize'],
save_as=item['path'] + item['saveas'],
status=True,
mode=mode)
def handle_argument(argument):
if argument == "-all":
test_all()
elif argument == "-example":
if len(argv) != 3:
print("-example takes as input the name of the example to run. Please run with 'python main.py -example NAME_OF_EXAMPLE.graph'")
return
run_test(argv[2])
elif argument == "-file":
if len(argv) != 3:
print("-file takes as input the name of the example to run. Please run with 'python main.py -file PATH/TO/FILE.graph'")
return
run_file(argv[2])
else:
print("Unrecognized command.")
def launch_closerlook(params):
import sys
sys.path.insert(0, '/private/home/sbaio/aa/dataset_design_few_shot/cl_fsl')
from train import run
from save_features import run_save
from test import run_test
print('Launching Closer Look training with params', params)
# train
run(params)
# save features
run_save(params)
# test
run_test(params)
def body():
res = run_test(*opts.test,
action=opts.action,
where=opts.where)
if type(res) == GeneratorType:
yield from res
return res
def main():
args = command.command()
if args.test:
test.run_test()
return 0
if args.folder is None:
print("folder is a required argument if not using test option ")
return 0
if os.path.exists(args.folder):
if args.duplicates:
command.remove_duplicates_decision(args.folder)
if args.order:
command.file_order_decision(args.folder)
else:
print("The folder does not exist")
def test(self):
# Run test
self.test_result, self.test_result_pp = run_test(
model=self.model,
args=self.args,
device=self.device,
result_dir=os.path.join(self.base_dir, "test"),
)
def old_way():
test.run_test()
G = nx.Graph()
keys = settings.test_set.keys()
for i in range(0, len(keys)):
for j in range(i + 1, len(keys)):
G.add_edge(keys[i], keys[j], weight=settings.links[i][j])
# G.add_edge('A', 'B', weight=0.1)
# G.add_edge('B', 'D', weight=2)
# G.add_edge('A', 'C', weight=3)
# G.add_edge('C', 'D', weight=4)
nx.draw(G, with_labels=True)
plt.show()
def angle_y_test(error_list,path):
error_list.append(Picture(run_test(path + "/vlc/angled/y_axis/-27.jpg",False,'test_rig'),
path + "/vlc/angled/y_axis/-27.jpg",
"-27 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/y_axis/-18.jpg",False,'test_rig'),
path + "/vlc/angled/y_axis/-18.jpg",
"-18 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/y_axis/-9.jpg",False,'test_rig'),
path + "/vlc/angled/y_axis/-9.jpg",
"-9 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/y_axis/0.jpg",False,'test_rig'),
path + "/vlc/angled/y_axis/0.jpg",
"0 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/y_axis/9.jpg",False,'test_rig'),
path + "/vlc/angled/y_axis/9.jpg",
"9 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/y_axis/18.jpg",False,'test_rig'),
path + "/vlc/angled/y_axis/18.jpg",
"18 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/y_axis/27.jpg",False,'test_rig'),
path + "/vlc/angled/y_axis/27.jpg",
"27 deg y axis Rotation, 5 Light Source",
0))
return error_list
def check(challenge_name, submission):
display(Markdown(f'### Testing: {challenge_name}'))
success, passed, failed = run_test(challenge_name, submission)
for args, expected, actual in failed:
print('Input: {}'.format(copy_args))
print('Expected: {!r}' % expected)
print('Got back: {!r}' % actual)
_display('SUCCESS' if success else 'FAIL', success)
def test_formed_set(name,
number,
size,
scale,
radius,
cellsize,
mode,
subfolder=''):
if subfolder:
try:
os.stat('results/' + name + subfolder)
except:
os.mkdir('results/' + name + subfolder)
for i in range(1, number + 1):
run_test('tests/' + name + '/test' + str(i) + '.bmp',
size[0],
size[1],
scale,
radius,
cellsize,
save_as='results/' + name + subfolder + '/test' + str(i) +
mode + '.png',
status=True,
mode=mode)
def angle_z_test(error_list,path):
error_list.append(Picture(run_test(path + "/vlc/angled/z_axis/0.jpg",False,'test_rig'),
path + "/vlc/angled/z_axis/0.jpg",
"0 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/z_axis/45.jpg",False,'test_rig'),
path + "/vlc/angled/z_axis/45.jpg",
"45 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/z_axis/90.jpg",False,'test_rig'),
path + "/vlc/angled/z_axis/90.jpg",
"90 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/z_axis/135.jpg",False,'test_rig'),
path + "/vlc/angled/z_axis/135.jpg",
"135 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/z_axis/180.jpg",False,'test_rig'),
path + "/vlc/angled/z_axis/180.jpg",
"180 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/z_axis/225.jpg",False,'test_rig'),
path + "/vlc/angled/z_axis/225.jpg",
"225 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/z_axis/270.jpg",False,'test_rig'),
path + "/vlc/angled/z_axis/270.jpg",
"270 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/z_axis/315.jpg",False,'test_rig'),
path + "/vlc/angled/z_axis/315.jpg",
"315 deg y axis Rotation, 5 Light Source",
0))
error_list.append(Picture(run_test(path + "/vlc/angled/z_axis/0.jpg",False,'test_rig'),
path + "/vlc/angled/z_axis/0.jpg",
"0 deg y axis Rotation, 5 Light Source",
0))
return error_list
def run_all():
start = time.time()
for mode in modes:
for order in orders:
for sort in sorting.__all__:
for n in sizes:
factory = Factory(n, order=order, mode=mode)
data = test.run_test(sort, factory)
try:
test.write_result_line(data)
except Exception:
print('data: \n', data)
print('%s concluido' % str(sort).split(' ')[1])
print('%s concluido' % order)
print('%s concluido' % mode)
end = time.time() - start
print('Concluído')
print("Tempo total: %s segundos ---" % end)
def sample_test(error_list):
error_list.append(Picture(run_test("./samples/x_0_y_1.27.jpg",False,'test_rig'),
"./samples/x_0_y_1.27.jpg",
"Normal 5 Light",
0))
error_list.append(Picture(run_test("./samples/small_blob.jpg",False,'test_rig'),
"./samples/small_blob.jpg",
"Normal 5 Light",
0))
error_list.append(Picture(run_test("./samples/2014-02-27--18-15-03-62.jpg",False,'test_rig'),
"./samples/2014-02-27--18-15-03-62.jpg",
"Normal 5 Light",
0))
error_list.append(Picture(run_test("./samples/2014-02-27--18-16-34-97.jpg",False,'test_rig'),
"./samples/2014-02-27--18-16-34-97.jpg",
"Normal 5 Light",
0))
error_list.append(Picture(run_test("./samples/4908_CBAD.jpg",False,'test_rig'),
"./samples/4908_CBAD.jpg",
"Normal 5 Light",
0))
error_list.append(Picture(run_test("./samples/mirror1.jpg",False,'test_rig'),
"./samples/mirror1.jpg",
"Normal 5 Light",
0))
error_list.append(Picture(run_test("./samples/mirror2.jpg",False,'test_rig'),
"./samples/mirror2.jpg",
"Normal 5 Light",
0))
error_list.append(Picture(run_test("./samples/mirror3.jpg",False,'test_rig'),
"./samples/mirror3.jpg",
"Normal 5 Light",
0))
# freq_list.append(Picture(0,run_test("./samples/4908_IFJE.jpg",True)))
# freq_list.append(Picture(0,run_test("./samples/4908_ADGH.jpg",True)))
# freq_list.append(Picture(error= run_test("./samples/1K_1.jpg",True)))
# freq_list.append(Picture(error= run_test("./samples/4908_SKQP",True)))
# freq_list.append(Picture(0,run_test("./samples/compressed.jpg",True)))
return error_list
overall_accuracy_arr = []
paccuracy_arr = []
naccuracy_arr = []
label=open('label').readline().strip()
net=open('net').readline().strip()
maxiter=0
# read maxiter
for i in open('{}-solver.prototxt'.format(net)):
l = i.split(':')
if l[0].strip() == 'max_iter':
maxiter = int(l[1].split('#')[0].strip())
break
for i in range(3):
netfn = '{}-{}.prototxt'.format(net,i)
weightfn = '{}_{}_iter_{}.caffemodel'.format(net,i,maxiter)
testdbfn = 'data-{}.h5'.format(i)
overall_accuracy,paccuracy,naccuracy = run_test(netfn,weightfn,testdbfn,label)
print 'cross validation:',i,'\toverall accuracy:',overall_accuracy,'\t++ rate:', paccuracy,'\t-- rate:', naccuracy
overall_accuracy_arr.append(overall_accuracy)
paccuracy_arr.append(paccuracy)
naccuracy_arr.append(naccuracy)
print '\nSummary:'
print 'overall accuracy:',sum(overall_accuracy_arr)/len(overall_accuracy_arr)
print '++ rate:',sum(paccuracy_arr)/len(paccuracy_arr)
print '-- rate:',sum(naccuracy_arr)/len(naccuracy_arr)
def dist_TX3K_test(freq_list,path):
init_len = len(freq_list)
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_1.jpg",
"TX3K, .5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_1.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_2.jpg",
"TX3K, .5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_2.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_3.jpg",
"TX3K, .5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_3.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_4.jpg",
"TX3K, .5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_4.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_5.jpg",
"TX3K, .5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_5.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_6.jpg",
"TX3K, .5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_6.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_7.jpg",
"TX3K, .5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_7.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_8.jpg",
"TX3K, .5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_8.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_9.jpg",
"TX3K, .5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/0.5m/0.5m_9.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1m/1m_1.jpg",
"TX3K, 1m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1m/1m_1.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1m/1m_2.jpg",
"TX3K, 1m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1m/1m_2.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1m/1m_3.jpg",
"TX3K, 1m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1m/1m_3.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1m/1m_4.jpg",
"TX3K, 1m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1m/1m_4.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1m/1m_5.jpg",
"TX3K, 1m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1m/1m_5.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1m/1m_6.jpg",
"TX3K, 1m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1m/1m_6.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1m/1m_7.jpg",
"TX3K, 1m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1m/1m_7.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1m/1m_8.jpg",
"TX3K, 1m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1m/1m_8.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1m/1m_9.jpg",
"TX3K, 1m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1m/1m_9.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_1.jpg",
"TX3K, 1.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_1.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_2.jpg",
"TX3K, 1.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_2.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_3.jpg",
"TX3K, 1.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_3.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_4.jpg",
"TX3K, 1.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_4.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_5.jpg",
"TX3K, 1.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_5.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_6.jpg",
"TX3K, 1.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_6.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_7.jpg",
"TX3K, 1.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_7.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_8.jpg",
"TX3K, 1.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_8.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_9.jpg",
"TX3K, 1.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/1.5m/1.5m_9.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2m/2m_1.jpg",
"TX3K, 2m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2m/2m_1.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2m/2m_2.jpg",
"TX3K, 2m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2m/2m_2.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2m/2m_3.jpg",
"TX3K, 2m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2m/2m_3.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2m/2m_4.jpg",
"TX3K, 2m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2m/2m_4.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2m/2m_5.jpg",
"TX3K, 2m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2m/2m_5.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2m/2m_6.jpg",
"TX3K, 2m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2m/2m_6.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2m/2m_7.jpg",
"TX3K, 2m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2m/2m_7.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2m/2m_8.jpg",
"TX3K, 2m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2m/2m_8.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2m/2m_9.jpg",
"TX3K, 2m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2m/2m_9.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_1.jpg",
"TX3K, 2.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_1.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_2.jpg",
"TX3K, 2.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_2.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_3.jpg",
"TX3K, 2.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_3.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_4.jpg",
"TX3K, 2.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_4.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_5.jpg",
"TX3K, 2.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_5.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_6.jpg",
"TX3K, 2.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_6.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_7.jpg",
"TX3K, 2.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_7.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_8.jpg",
"TX3K, 2.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_8.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_9.jpg",
"TX3K, 2.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/2.5m/2.5m_9.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3m/3m_1.jpg",
"TX3K, 3m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3m/3m_1.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3m/3m_2.jpg",
"TX3K, 3m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3m/3m_2.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3m/3m_3.jpg",
"TX3K, 3m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3m/3m_3.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3m/3m_4.jpg",
"TX3K, 3m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3m/3m_4.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3m/3m_5.jpg",
"TX3K, 3m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3m/3m_5.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3m/3m_6.jpg",
"TX3K, 3m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3m/3m_6.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3m/3m_7.jpg",
"TX3K, 3m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3m/3m_7.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3m/3m_8.jpg",
"TX3K, 3m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3m/3m_8.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3m/3m_9.jpg",
"TX3K, 3m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3m/3m_9.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_1.jpg",
"TX3K, 3.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_1.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_2.jpg",
"TX3K, 3.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_2.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_3.jpg",
"TX3K, 3.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_3.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_4.jpg",
"TX3K, 3.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_4.jpg",True,'test_rig')))
# freq_list.append(Picture(0,run_test(path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_5.jpg",True)))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_6.jpg",
"TX3K, 3.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_6.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_7.jpg",
"TX3K, 3.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_7.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_8.jpg",
"TX3K, 3.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_8.jpg",True,'test_rig')))
freq_list.append(Picture(0,
path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_9.jpg",
"TX3K, 3.5m, 1 Light Source",
run_test(path + "/vlc/back_camera/distance/TX3K/3.5m/3.5m_9.jpg",True,'test_rig')))
i = init_len
while i < len(freq_list):
freq_list[i].freq_diff = freq_list[i].freq_diff - 3000
i = i + 1
return freq_list
import os
from test import clean_output, run_test
if __name__ == "__main__":
output_dir = "output"
clean_output(output_dir)
tests = [f for f in os.listdir("../demo/") if f.endswith(".py")]
print("Found %d tests" % len(tests))
# Step into output directory
os.chdir(output_dir)
failures = []
for test in tests:
run_test(test, generate_reference=1)
elif d is None:
return 'null'
else:
return unicode(d)
def __dump_list(self, l):
'''dump list l into string format of json'''
assert(isinstance(l, list))
sep = u', ' if self.space_between_seq else u','
return '[%s]' % (sep.join((self.__dump_one_object(o) for o in l)))
def __dump_dict(self, d):
'''dump dict d into string format of json'''
assert(isinstance(d, dict))
sep = u', ' if self.space_between_seq else u','
return '{%s}' % (sep.join((self.__dump_str(k) + u':' + self.__dump_one_object(v) for k,v in d.items())))
def __dump_str(self, s):
'''dump unicode s into string format of json'''
assert(isinstance(s, unicode))
res = u''.join((self.dump_back_slash[c] if c in self.dump_back_slash else c for c in s))
return '"%s"' % res
if __name__ == '__main__':
import test
test.run_test()
# if epoch % opt.save_epoch_freq == 0:
# print('saving the model at the end of epoch %d, iters %d' %
# (epoch, total_steps))
# model.save_network('latest')
# model.save_network(epoch)
# writer.plot(epoch, "train", dataset.dataset.dataset.classes)
print('End of epoch %d / %d \t Time Taken: %d sec' %
(epoch, opt.niter + opt.niter_decay,
time.time() - epoch_start_time))
wandb.log({"Epoch elapsed time": time.time() - epoch_start_time},
step=epoch)
model.update_learning_rate()
model.save_network('latest')
# need to re run test on train dataset because otherwise results are weird
run_test(dataset, epoch, "train")
prec = run_test(val_dataset, epoch, "internal_val")
# prec = run_test(None, epoch)
# Save best model and best prediction
if prec > best_prec:
best_prec = prec
model.save_network('best')
epoch_no_improve = 0
else:
# Early stopping
epoch_no_improve += 1
if epoch_no_improve == opt.patience:
model.load_network("best")
model.save_network('latest')
wandb.run.summary[
def body():
res = run_test(*opts.test, action=opts.action, where=opts.where)
if type(res) == GeneratorType:
yield from res
return res
def full_box_test(error_list):#add in testing suite of box test to shed-data
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/1.jpg",False,'box'),
"/home/noah/lab/box_light/full/1.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/2.jpg",False,'box'),
"/home/noah/lab/box_light/full/2.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/3.jpg",False,'box'),
"/home/noah/lab/box_light/full/3.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/4.jpg",False,'box'),
"/home/noah/lab/box_light/full/4.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/5.jpg",False,'box'),
"/home/noah/lab/box_light/full/5.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/6.jpg",False,'box'),
"/home/noah/lab/box_light/full/6.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/7.jpg",False,'box'),
"/home/noah/lab/box_light/full/7.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/8.jpg",False,'box'),
"/home/noah/lab/box_light/full/8.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/9.jpg",False,'box'),
"/home/noah/lab/box_light/full/9.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/10.jpg",False,'box'),
"/home/noah/lab/box_light/full/10.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/11.jpg",False,'box'),
"/home/noah/lab/box_light/full/11.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/12.jpg",False,'box'),
"/home/noah/lab/box_light/full/12.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/13.jpg",False,'box'),
"/home/noah/lab/box_light/full/13.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/14.jpg",False,'box'),
"/home/noah/lab/box_light/full/14.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/15.jpg",False,'box'),
"/home/noah/lab/box_light/full/15.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/16.jpg",False,'box'),
"/home/noah/lab/box_light/full/16.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/17.jpg",False,'box'),
"/home/noah/lab/box_light/full/17.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/18.jpg",False,'box'),
"/home/noah/lab/box_light/full/18.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/19.jpg",False,'box'),
"/home/noah/lab/box_light/full/19.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
"""
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/20.jpg",False,'box'),
"/home/noah/lab/box_light/full/20.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/21.jpg",False,'box'),
"/home/noah/lab/box_light/full/21.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/22.jpg",False,'box'),
"/home/noah/lab/box_light/full/22.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/23.jpg",False,'box'),
"/home/noah/lab/box_light/full/23.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/24.jpg",False,'box'),
"/home/noah/lab/box_light/full/24.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/25.jpg",False,'box'),
"/home/noah/lab/box_light/full/25.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/26.jpg",False,'box'),
"/home/noah/lab/box_light/full/26.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/27.jpg",False,'box'),
"/home/noah/lab/box_light/full/27.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
error_list.append(Picture(run_test("/home/noah/lab/box_light/full/28.jpg",False,'box'),
"/home/noah/lab/box_light/full/28.jpg",
"full 4 light image close up 1,2,2.46,3 k freq",
0))
"""
return error_list
print('saving the model at the end of epoch %d, iters %d' %
(epoch, total_steps))
model.save_network('latest')
model.save_network(epoch)
# Plot training loss per epoch on tensorboard
writer.plot_epoch_loss(train_loss_epoch / len(dataset), epoch)
print('End of epoch %d / %d \t Time Taken: %d sec' %
(epoch, opt.niter + opt.niter_decay + opt.epoch_count,
time.time() - epoch_start_time))
if opt.verbose_plot:
writer.plot_model_wts(model, epoch)
if epoch % opt.run_test_freq == 0:
acc = run_test(epoch)
# Track the best model
if opt.dataset_mode == 'regression' and acc < best_val_reg_acc:
best_val_reg_acc = acc
best_epoch = epoch
elif opt.dataset_mode in ('classification', 'binary_class'
) and acc > best_val_cls_acc:
best_val_cls_acc = acc
best_epoch = epoch
writer.plot_acc(acc, epoch)
lr = model.update_learning_rate(acc, epoch)
writer.plot_lr(lr, epoch)
# At end of training, run the last model on the test set
"""
A definition test case.
"""
def test_create_network(self):
"""
Tests creating a network.
"""
network = vertigo.create_network("test")
self.assert_equals("test", network.address)
network.address = "foo"
self.assert_equals("foo", network.address)
network.enable_acking()
self.assert_true(network.acking_enabled())
network.disable_acking()
self.assert_false(network.acking_enabled())
network.num_ackers = 10
self.assert_equals(10, network.num_ackers)
network.ack_expire = 50000
self.assert_equals(50000, network.ack_expire)
component = network.from_verticle('test_feeder_verticle', main='test_feeder_verticle.py')
self.assert_equals('test_feeder_verticle', component.name)
self.assert_equals('test_feeder_verticle.py', component.main)
component.workers = 4
self.assert_equals(4, component.workers)
component2 = component.to_verticle('test_worker_verticle')
component2.main = 'test_worker_verticle.py'
self.assert_equals('test_worker_verticle.py', component2.main)
self.complete()
run_test(DefinitionTestCase())
print('TRAIN ACC [%.3f]'%(writer.acc))
writer.plot_train_acc(writer.acc, epoch)
if epoch % opt.save_epoch_freq == 0:
print('saving the model at the end of epoch %d, iters %d' %
(epoch, total_steps))
model.save_network('latest_net')
# model.save_network('epoch_%d' % (epoch))
model.update_learning_rate()
if opt.verbose_plot:
writer.plot_model_wts(model, epoch)
if epoch % opt.run_test_freq == 0:
acc = run_test(opt.test_data, epoch)
writer.plot_acc(acc, epoch)
logger.info('Loss: %f, Acc: %f', loss, acc)
logger.info('IndexError count - train: %d', heappop_error_train)
print('End of epoch %d / %d \t Time Taken: %d sec' %
(epoch, opt.niter + opt.niter_decay, time.time() - epoch_start_time))
logger.info('End of epoch %d / %d \t Time Taken: %d sec',
epoch, opt.niter + opt.niter_decay, time.time() - epoch_start_time)
if (acc >= best_tst_acc) and epoch > 5:
best_tst_acc = acc
model.save_network('%.6f-%04d' % (acc, epoch))
print('Saving model....')
model.set_input_data(data)
model.optimize()
running_loss += model.loss_val
if total_steps % opt.frequency == 0:
loss_val = running_loss / opt.frequency
writer.print_loss(epoch, count, loss_val)
writer.plot_loss(epoch, count, loss_val, len(dataset))
running_loss = 0
if i % opt.loop_frequency == 0:
print('saving the latest model (epoch %d, total_steps %d)' %
(epoch, total_steps))
model.save_network('latest')
# break
if epoch % opt.epoch_frequency == 0:
print('saving the model at the end of epoch %d, iters %d' %
(epoch, total_steps))
if (epoch - 1) % 20 == 0:
model.log_history_and_plot(writer, epoch, count)
model.log_features_and_plot(epoch, count)
model.save_network('latest')
model.save_network(epoch)
if epoch % opt.test_frequency == 0:
acc = run_test(epoch)
writer.plot_acc(acc, epoch)
# break
wait = input("input")
writer.close()
for i in range(size):
settings.clusters[i].level = labels[i]
nodes.append({
"id": settings.clusters[i].get_name(),
"group": labels[i],
"size": len(settings.clusters[i].funcs) + 2
})
links = []
for i in range(size):
for j in range(i + 1, size):
if settings.links[i][j] != 0:
links.append({
"source": settings.clusters[i].get_name(),
"target": settings.clusters[j].get_name(),
"value": settings.links[i][j] * 5
})
data = {"nodes": nodes, "links": links}
store2json('miserables2.json', data)
if __name__ == '__main__':
test.run_test()
test.print_links()
test.print_clusters()
labels = get_random_labels()
generate_force_layout(labels)
def main():
opt = TrainOptions().parse()
if opt == None:
return
dataset = DataLoader(opt)
dataset_size = len(dataset) * opt.num_grasps_per_object
model = create_model(opt)
writer = Writer(opt)
total_steps = 0
for epoch in range(opt.epoch_count, opt.niter + opt.niter_decay + 1):
epoch_start_time = time.time()
iter_data_time = time.time()
epoch_iter = 0
for i, data in enumerate(dataset):
iter_start_time = time.time()
if total_steps % opt.print_freq == 0:
t_data = iter_start_time - iter_data_time
total_steps += opt.batch_size
epoch_iter += opt.batch_size
model.set_input(data)
model.optimize_parameters()
if total_steps % opt.print_freq == 0:
loss_types = []
if opt.arch == "vae":
loss = [
model.loss, model.kl_loss, model.reconstruction_loss,
model.confidence_loss, model.l2_loss
]
loss_types = [
"total_loss", "kl_loss", "reconstruction_loss",
"confidence loss", "l2_loss"
]
elif opt.arch == "gan":
loss = [
model.loss, model.reconstruction_loss,
model.confidence_loss
]
loss_types = [
"total_loss", "reconstruction_loss", "confidence_loss"
]
else:
loss = [
model.loss, model.classification_loss,
model.confidence_loss
]
loss_types = [
"total_loss", "classification_loss", "confidence_loss"
]
t = (time.time() - iter_start_time) / opt.batch_size
writer.print_current_losses(epoch, epoch_iter, loss, t, t_data,
loss_types)
writer.plot_loss(loss, epoch, epoch_iter, dataset_size,
loss_types)
if i % opt.save_latest_freq == 0:
print('saving the latest model (epoch %d, total_steps %d)' %
(epoch, total_steps))
model.save_network('latest', epoch)
iter_data_time = time.time()
if epoch % opt.save_epoch_freq == 0:
print('saving the model at the end of epoch %d, iters %d' %
(epoch, total_steps))
model.save_network('latest', epoch)
model.save_network(str(epoch), epoch)
print('End of epoch %d / %d \t Time Taken: %d sec' %
(epoch, opt.niter + opt.niter_decay,
time.time() - epoch_start_time))
model.update_learning_rate()
if opt.verbose_plot:
writer.plot_model_wts(model, epoch)
if epoch % opt.run_test_freq == 0:
acc = run_test(epoch, name=opt.name)
writer.plot_acc(acc, epoch)
writer.close()
network = self._create_fail_network('test_failing_polling_feeder.py')
cluster = LocalCluster()
def deploy_handler(error, context):
self.assert_null(error)
self.assert_not_null(context)
cluster.deploy(network, deploy_handler)
def test_stream_feeder_ack(self):
"""
Tests the stream feeder acking support.
"""
network = self._create_ack_network('test_acking_stream_feeder.py')
cluster = LocalCluster()
def deploy_handler(error, context):
self.assert_null(error)
self.assert_not_null(context)
cluster.deploy(network, deploy_handler)
def test_stream_feeder_fail(self):
"""
Tests the stream feeder fail support.
"""
network = self._create_fail_network('test_failing_stream_feeder.py')
cluster = LocalCluster()
def deploy_handler(error, context):
self.assert_null(error)
self.assert_not_null(context)
cluster.deploy(network, deploy_handler)
run_test(FeederTestCase())
amount=pruning_percentage)
# Control Seed
# torch.manual_seed(args.seed)
# Select Device
use_cuda = not args.no_cuda and torch.cuda.is_available()
device = torch.device("cuda" if use_cuda else 'cpu')
if use_cuda:
print("Using CUDA!")
torch.cuda.manual_seed(args.seed)
else:
print('Not using CUDA!!!')
model = test.run_test(args.model, 1)
# NOTE : `weight_decay` term denotes L2 regularization loss term
# optimizer = optim.Adam(model.parameters(), lr=args.lr, weight_decay=0.0001)
# initial_optimizer_state_dict = optimizer.state_dict()
# model.prune_by_std(args.sensitivity)
# Pruning
# print("--- Before pruning ---")
# prune_darts(model, var1)
for i2 in [0.25, 0.30, 0.40, 0.5, 0.6, 0.7, 0.8, 0.9]:
model = test.run_test(args.model, 1)
prune_darts(model, i2)
print("--- Quantization --- ")
cluster.deploy_network(network, handler=deploy_handler)
vertigo.deploy_cluster('test_basic_send', handler=cluster_handler)
def test_group_send(self):
"""Test sending group messages between two components."""
network = vertigo.create_network('test-group')
network.add_verticle('sender', main='test_group_sender.py')
network.add_verticle('receiver', main='test_group_receiver.py')
network.create_connection(('sender', 'out'), ('receiver', 'in'))
def cluster_handler(error, cluster):
self.assert_null(error)
def deploy_handler(error, network):
self.assert_null(error)
cluster.deploy_network(network, handler=deploy_handler)
vertigo.deploy_cluster('test_group_send', handler=cluster_handler)
def test_batch_send(self):
"""Test sending batch messages between two components."""
network = vertigo.create_network('test-batch')
network.add_verticle('sender', main='test_batch_sender.py')
network.add_verticle('receiver', main='test_batch_receiver.py')
network.create_connection(('sender', 'out'), ('receiver', 'in'))
def cluster_handler(error, cluster):
self.assert_null(error)
def deploy_handler(error, network):
self.assert_null(error)
cluster.deploy_network(network, handler=deploy_handler)
vertigo.deploy_cluster('test_batch_send', handler=cluster_handler)
run_test(NetworkTestCase())
val_vec.append(running_loss_val)
if ckpt and np.sum(
np.asarray(val_vec) > running_loss_val) == len(val_vec) - 1:
print('Making check point - Best')
torch.save(model.state_dict(),
os.path.join(save_dir, dir_name, 'best.pkl'))
print('Epoch:', epoch + 1, '|| Total loss validation:',
running_loss_val)
try:
if epoch % 1 == 0:
j = run_test(model=model,
dataloader=val_dataloader,
path=join(save_dir, dir_name),
device=device,
save=False)
J.append(j)
except:
J.append([0 for i in range(25)])
except KeyboardInterrupt:
print('Training stopped by KeyboardInterrupt')
except:
1 + 1
print('=======================================')
print(' Error found')
print('=======================================')
raise
def runme():
if (len(sys.argv) > 1 and sys.argv[1] == "fill"):
run_model()
run_test()
port = int(os.environ.get('PORT', 5000))
app.run(debug=True, port=port, host='0.0.0.0')
from net.huffmancoding import huffman_encode_model
import test
import util
parser = argparse.ArgumentParser(
description='Huffman encode a quantized model')
parser.add_argument(
'--model',
type=str,
default=
'../cnn/eval-EXP-20200415-092238/model_after_weight_sharing.ptmodel',
help='saved quantized model')
parser.add_argument('--no-cuda',
action='store_true',
default=False,
help='disables CUDA')
parser.add_argument(
'--output',
default='saves/model_after_pruning_and_quantization.ptmodel',
type=str,
help='path to model output')
args = parser.parse_args()
use_cuda = not args.no_cuda and torch.cuda.is_available()
device = torch.device("cuda" if use_cuda else 'cpu')
model = test.run_test(args.output, 2)
# model = torch.load(args.model)
huffman_encode_model(model)
