EasyCV/tests/models/backbones/test_genet.py

68 lines
2.1 KiB
Python

# Copyright (c) Alibaba, Inc. and its affiliates.
import copy
import random
import unittest
import numpy as np
import torch
from easycv.models import modelzoo
from easycv.models.backbones import PlainNet
class PlainNetTest(unittest.TestCase):
def setUp(self):
print(('Testing %s.%s' % (type(self).__name__, self._testMethodName)))
def test_genet_withfc(self):
for struct in ['normal', 'large']:
with torch.no_grad():
# input data
batch_size = random.randint(10, 30)
a = torch.rand(batch_size, 3, 224, 224).to('cuda')
num_classes = random.randint(10, 1000)
net = PlainNet(struct, num_classes=num_classes).to('cuda')
net.init_weights()
net.eval()
self.assertTrue(len(list(net(a)[-1].shape)) == 2)
self.assertTrue(net(a)[-1].size(1) == num_classes)
self.assertTrue(net(a)[-1].size(0) == batch_size)
def test_genet_withoutfc(self):
for struct in ['normal', 'large']:
with torch.no_grad():
# input data
batch_size = random.randint(10, 30)
a = torch.rand(batch_size, 3, 256, 256).to('cuda')
net = PlainNet(struct, num_classes=0).to('cuda')
net.init_weights()
net.eval()
self.assertTrue(net(a)[-1].size(1) == 2560)
self.assertTrue(net(a)[-1].size(0) == batch_size)
def test_genet_load_modelzoo(self):
for struct in ['normal', 'large']:
with torch.no_grad():
net = PlainNet(struct, num_classes=1000).to('cuda')
original_weight = net.module_list[0].netblock.weight
original_weight = copy.deepcopy(
original_weight.cpu().data.numpy())
net.init_weights()
load_weight = net.module_list[0].netblock.weight.cpu(
).data.numpy()
self.assertFalse(np.allclose(original_weight, load_weight))
if __name__ == '__main__':
unittest.main()