def __init__(self, nIndex, nOutput, paddingValue=-1, maxNorm=None, normType=None):
super(LookupTable, self).__init__()
self.weight = torch.Tensor(nIndex, nOutput)
self.gradWeight = torch.Tensor(nIndex, nOutput).zero_()
self.paddingValue = paddingValue
self.maxNorm = maxNorm
self.normType = normType
self.shouldScaleGradByFreq = False
self._gradOutput = None
self._sorted = None
self._indices = None
self._count = torch.IntTensor()
self._input = torch.LongTensor()
self.reset()
python类IntTensor()的实例源码
def type(self, type=None, tensorCache=None):
if type is None:
return self._type
super(LookupTable, self).type(type, tensorCache)
if type == 'torch.cuda.FloatTensor':
# CUDA uses _sorted and _indices temporary tensors
self._sorted = torch.cuda.LongTensor()
self._indices = torch.cuda.LongTensor()
self._count = torch.cuda.LongTensor()
self._input = torch.cuda.LongTensor()
else:
# self._count and self._input should only be converted if using Cuda
self._count = torch.IntTensor()
self._input = torch.LongTensor()
return self
def test_serialization_array_with_storage(self):
x = torch.randn(5, 5).cuda()
y = torch.IntTensor(2, 5).fill_(0).cuda()
q = [x, y, x, y.storage()]
with tempfile.NamedTemporaryFile() as f:
torch.save(q, f)
f.seek(0)
q_copy = torch.load(f)
self.assertEqual(q_copy, q, 0)
q_copy[0].fill_(5)
self.assertEqual(q_copy[0], q_copy[2], 0)
self.assertTrue(isinstance(q_copy[0], torch.cuda.DoubleTensor))
self.assertTrue(isinstance(q_copy[1], torch.cuda.IntTensor))
self.assertTrue(isinstance(q_copy[2], torch.cuda.DoubleTensor))
self.assertTrue(isinstance(q_copy[3], torch.cuda.IntStorage))
q_copy[1].fill_(10)
self.assertTrue(q_copy[3], torch.cuda.IntStorage(10).fill_(10))
def _test_neg(self, cast):
float_types = ['torch.DoubleTensor', 'torch.FloatTensor', 'torch.LongTensor']
int_types = ['torch.IntTensor', 'torch.ShortTensor']
for t in float_types + int_types:
if t in float_types:
a = cast(torch.randn(100, 90).type(t))
else:
a = cast(torch.Tensor(100, 90).type(t).random_())
zeros = cast(torch.Tensor().type(t)).resize_as_(a).zero_()
res_add = torch.add(zeros, -1, a)
res_neg = a.clone()
res_neg.neg_()
self.assertEqual(res_neg, res_add)
# test out of place as well
res_neg_out_place = a.clone().neg()
self.assertEqual(res_neg_out_place, res_add)
# test via __neg__ operator
res_neg_op = -a.clone()
self.assertEqual(res_neg_op, res_add)
def _test_btrisolve(self, cast):
a = torch.FloatTensor((((1.3722, -0.9020),
(1.8849, 1.9169)),
((0.7187, -1.1695),
(-0.0139, 1.3572)),
((-1.6181, 0.7148),
(1.3728, 0.1319))))
b = torch.FloatTensor(((4.02, 6.19),
(-1.56, 4.00),
(9.81, -4.09)))
a, b = cast(a), cast(b)
info = cast(torch.IntTensor())
LU_data, pivots = a.btrifact(info=info)
self.assertEqual(info.abs().sum(), 0)
x = torch.btrisolve(b, LU_data, pivots)
b_ = torch.bmm(a, x.unsqueeze(2)).squeeze()
self.assertEqual(b_, b)
def test_abs(self):
size = 1000
max_val = 1000
original = torch.rand(size).mul(max_val)
# Tensor filled with values from {-1, 1}
switch = torch.rand(size).mul(2).floor().mul(2).add(-1)
types = ['torch.DoubleTensor', 'torch.FloatTensor', 'torch.LongTensor', 'torch.IntTensor']
for t in types:
data = original.type(t)
switch = switch.type(t)
res = torch.mul(data, switch)
# abs is used in assertEqual so we use the slow version instead
self.assertTensorsSlowEqual(res.abs(), data, 1e-16)
# Checking that the right abs function is called for LongTensor
bignumber = 2 ^ 31 + 1
res = torch.LongTensor((-bignumber,))
self.assertGreater(res.abs()[0], 0)
def forward(self, features, rois):
batch_size, num_channels, data_height, data_width = features.size()
num_rois = rois.size()[0]
output = torch.zeros(num_rois, num_channels, self.pooled_height, self.pooled_width)
argmax = torch.IntTensor(num_rois, num_channels, self.pooled_height, self.pooled_width).zero_()
if not features.is_cuda:
_features = features.permute(0, 2, 3, 1)
roi_pooling.roi_pooling_forward(self.pooled_height, self.pooled_width, self.spatial_scale,
_features, rois, output)
# output = output.cuda()
else:
output = output.cuda()
argmax = argmax.cuda()
roi_pooling.roi_pooling_forward_cuda(self.pooled_height, self.pooled_width, self.spatial_scale,
features, rois, output, argmax)
self.output = output
self.argmax = argmax
self.rois = rois
self.feature_size = features.size()
return output
def __call__(self, img):
assert isinstance(img, np.ndarray)
# handle numpy array
if img.dtype == np.uint16:
img = img.astype(np.int32)
div = 2**16
elif img.dtype == np.uint32:
img = img.astype(np.int32)
div = 2**32
elif img.dtype == np.int32:
div = 2**32
else:
div = 1.
img = torch.from_numpy(img.transpose((2, 0, 1)))
if isinstance(img, torch.ByteTensor):
return img.float().div(255)
elif isinstance(img, torch.IntTensor):
return img.float().div(div)
else:
return img
def __init__(self, cfgfile):
super(Darknet, self).__init__()
self.blocks = parse_cfg(cfgfile)
self.models = self.create_network(self.blocks) # merge conv, bn,leaky
self.loss = self.models[len(self.models)-1]
self.width = int(self.blocks[0]['width'])
self.height = int(self.blocks[0]['height'])
if self.blocks[(len(self.blocks)-1)]['type'] == 'region':
self.anchors = self.loss.anchors
self.num_anchors = self.loss.num_anchors
self.anchor_step = self.loss.anchor_step
self.num_classes = self.loss.num_classes
self.header = torch.IntTensor([0,0,0,0])
self.seen = 0
def process_string(self, sequence, size, remove_repetitions=False):
string = ''
offsets = []
for i in range(size):
char = self.int_to_char[sequence[i]]
if char != self.int_to_char[self.blank_index]:
# if this char is a repetition and remove_repetitions=true, then skip
if remove_repetitions and i != 0 and char == self.int_to_char[sequence[i - 1]]:
pass
elif char == self.labels[self.space_index]:
string += ' '
offsets.append(i)
else:
string = string + char
offsets.append(i)
return string, torch.IntTensor(offsets)
def iteration(input_data):
target = torch.IntTensor(int(batch_size * ((seconds * 100) / 2))).fill_(1) # targets, align half of the audio
target_size = torch.IntTensor(batch_size).fill_(int((seconds * 100) / 2))
input_percentages = torch.IntTensor(batch_size).fill_(1)
inputs = Variable(input_data, requires_grad=False)
target_sizes = Variable(target_size, requires_grad=False)
targets = Variable(target, requires_grad=False)
start = time.time()
out = model(inputs)
out = out.transpose(0, 1) # TxNxH
seq_length = out.size(0)
sizes = Variable(input_percentages.mul_(int(seq_length)).int(), requires_grad=False)
loss = criterion(out, targets, sizes, target_sizes)
loss = loss / inputs.size(0) # average the loss by minibatch
# compute gradient
optimizer.zero_grad()
loss.backward()
optimizer.step()
torch.cuda.synchronize()
end = time.time()
del loss
del out
return start, end
def _collate_fn(batch):
def func(p):
return p[0].size(1)
longest_sample = max(batch, key=func)[0]
freq_size = longest_sample.size(0)
minibatch_size = len(batch)
max_seqlength = longest_sample.size(1)
inputs = torch.zeros(minibatch_size, 1, freq_size, max_seqlength)
input_percentages = torch.FloatTensor(minibatch_size)
target_sizes = torch.IntTensor(minibatch_size)
targets = []
for x in range(minibatch_size):
sample = batch[x]
tensor = sample[0]
target = sample[1]
seq_length = tensor.size(1)
inputs[x][0].narrow(1, 0, seq_length).copy_(tensor)
input_percentages[x] = seq_length / float(max_seqlength)
target_sizes[x] = len(target)
targets.extend(target)
targets = torch.IntTensor(targets)
return inputs, targets, input_percentages, target_sizes
def __init__(self, nIndex, nOutput, paddingValue=-1, maxNorm=None, normType=None):
super(LookupTable, self).__init__()
self.weight = torch.Tensor(nIndex, nOutput)
self.gradWeight = torch.Tensor(nIndex, nOutput).zero_()
self.paddingValue = paddingValue
self.maxNorm = maxNorm
self.normType = normType
self.shouldScaleGradByFreq = False
self._gradOutput = None
self._sorted = None
self._indices = None
self._count = torch.IntTensor()
self._input = torch.LongTensor()
self.reset()
def type(self, type=None, tensorCache=None):
if type is None:
return self._type
super(LookupTable, self).type(type, tensorCache)
if type == 'torch.cuda.FloatTensor':
# CUDA uses _sorted and _indices temporary tensors
self._sorted = torch.cuda.LongTensor()
self._indices = torch.cuda.LongTensor()
self._count = torch.cuda.LongTensor()
self._input = torch.cuda.LongTensor()
else:
# self._count and self._input should only be converted if using Cuda
self._count = torch.IntTensor()
self._input = torch.LongTensor()
return self
def test_serialization_array_with_storage(self):
x = torch.randn(5, 5).cuda()
y = torch.IntTensor(2, 5).fill_(0).cuda()
q = [x, y, x, y.storage()]
with tempfile.NamedTemporaryFile() as f:
torch.save(q, f)
f.seek(0)
q_copy = torch.load(f)
self.assertEqual(q_copy, q, 0)
q_copy[0].fill_(5)
self.assertEqual(q_copy[0], q_copy[2], 0)
self.assertTrue(isinstance(q_copy[0], torch.cuda.DoubleTensor))
self.assertTrue(isinstance(q_copy[1], torch.cuda.IntTensor))
self.assertTrue(isinstance(q_copy[2], torch.cuda.DoubleTensor))
self.assertTrue(isinstance(q_copy[3], torch.cuda.IntStorage))
q_copy[1].fill_(10)
self.assertTrue(q_copy[3], torch.cuda.IntStorage(10).fill_(10))
def _test_neg(self, cast):
float_types = ['torch.DoubleTensor', 'torch.FloatTensor', 'torch.LongTensor']
int_types = ['torch.IntTensor', 'torch.ShortTensor', 'torch.ByteTensor',
'torch.CharTensor']
for t in float_types + int_types:
if t in float_types:
a = cast(torch.randn(100, 90).type(t))
else:
a = cast(torch.Tensor(100, 90).type(t).random_())
zeros = cast(torch.Tensor().type(t)).resize_as_(a).zero_()
res_add = torch.add(zeros, -1, a)
res_neg = a.clone()
res_neg.neg_()
self.assertEqual(res_neg, res_add)
# test out of place as well
res_neg_out_place = a.clone().neg()
self.assertEqual(res_neg_out_place, res_add)
# test via __neg__ operator
res_neg_op = -a.clone()
self.assertEqual(res_neg_op, res_add)
def _test_btrisolve(self, cast):
a = torch.FloatTensor((((1.3722, -0.9020),
(1.8849, 1.9169)),
((0.7187, -1.1695),
(-0.0139, 1.3572)),
((-1.6181, 0.7148),
(1.3728, 0.1319))))
b = torch.FloatTensor(((4.02, 6.19),
(-1.56, 4.00),
(9.81, -4.09)))
a, b = cast(a), cast(b)
info = cast(torch.IntTensor())
LU_data, pivots = a.btrifact(info=info)
self.assertEqual(info.abs().sum(), 0)
x = torch.btrisolve(b, LU_data, pivots)
b_ = torch.bmm(a, x.unsqueeze(2)).squeeze()
self.assertEqual(b_, b)
def test_abs(self):
size = 1000
max_val = 1000
original = torch.rand(size).mul(max_val)
# Tensor filled with values from {-1, 1}
switch = torch.rand(size).mul(2).floor().mul(2).add(-1)
types = ['torch.DoubleTensor', 'torch.FloatTensor', 'torch.LongTensor', 'torch.IntTensor']
for t in types:
data = original.type(t)
switch = switch.type(t)
res = torch.mul(data, switch)
# abs is used in assertEqual so we use the slow version instead
self.assertTensorsSlowEqual(res.abs(), data, 1e-16)
# Checking that the right abs function is called for LongTensor
bignumber = 2 ^ 31 + 1
res = torch.LongTensor((-bignumber,))
self.assertGreater(res.abs()[0], 0)
def test_new(self):
x = torch.autograd.Variable(torch.Tensor())
y = torch.autograd.Variable(torch.randn(4, 4))
z = torch.autograd.Variable(torch.IntTensor([1, 2, 3]))
self.assertEqual(x.new().shape, [0])
self.assertEqual(x.new(), x)
self.assertEqual(x.new(1, 2).shape, [1, 2])
self.assertEqual(x.new(torch.Size([3, 4])).shape, [3, 4])
self.assertEqual(x.new([3, 4]).shape, [2])
self.assertEqual(x.new([3, 4]).tolist(), [3, 4])
self.assertEqual(x.new((3, 4)).tolist(), [3, 4])
self.assertEqual(x.new(size=(3, 4)).shape, [3, 4])
self.assertEqual(x.new(tuple()).shape, [0])
self.assertEqual(x.new(y.storage()).data_ptr(), y.data_ptr())
self.assertEqual(x.new(y).data_ptr(), y.data_ptr())
self.assertIsNot(x.new(y), y)
self.assertRaises(TypeError, lambda: x.new(z))
# TypeError would be better
self.assertRaises(RuntimeError, lambda: x.new(z.storage()))
def forward(self, features, rois):
batch_size, num_channels, data_height, data_width = features.size()
num_rois = rois.size()[0]
output = torch.zeros(num_rois, num_channels, self.pooled_height, self.pooled_width)
argmax = torch.IntTensor(num_rois, num_channels, self.pooled_height, self.pooled_width).zero_()
if not features.is_cuda:
_features = features.permute(0, 2, 3, 1)
roi_pooling.roi_pooling_forward(self.pooled_height, self.pooled_width, self.spatial_scale,
_features, rois, output)
# output = output.cuda()
else:
output = output.cuda()
argmax = argmax.cuda()
roi_pooling.roi_pooling_forward_cuda(self.pooled_height, self.pooled_width, self.spatial_scale,
features, rois, output, argmax)
self.output = output
self.argmax = argmax
self.rois = rois
self.feature_size = features.size()
return output
