Python mxnet.ndarray.empty() Examples
The following are 30
code examples of mxnet.ndarray.empty().
You can vote up the ones you like or vote down the ones you don't like,
and go to the original project or source file by following the links above each example.
You may also want to check out all available functions/classes of the module
mxnet.ndarray
, or try the search function
.
.
Example #1
| Source File: utils.py From SNIPER-mxnet with Apache License 2.0 | 6 votes |
|
def get_executor(sym, ctx, data_inputs, initializer=None):
data_shapes = {k: v.shape for k, v in data_inputs.items()}
arg_names = sym.list_arguments()
aux_names = sym.list_auxiliary_states()
param_names = list(set(arg_names) - set(data_inputs.keys()))
arg_shapes, output_shapes, aux_shapes = sym.infer_shape(**data_shapes)
arg_name_shape = {k: s for k, s in zip(arg_names, arg_shapes)}
params = {n: nd.empty(arg_name_shape[n], ctx=ctx) for n in param_names}
params_grad = {n: nd.empty(arg_name_shape[n], ctx=ctx) for n in param_names}
aux_states = {k: nd.empty(s, ctx=ctx) for k, s in zip(aux_names, aux_shapes)}
exe = sym.bind(ctx=ctx, args=dict(params, **data_inputs),
args_grad=params_grad,
aux_states=aux_states)
if initializer is not None:
for k, v in params.items():
initializer(k, v)
return exe, params, params_grad, aux_states
Example #2
| Source File: lfw_comparison_and_plot_roc.py From MobileFace with MIT License | 6 votes |
|
def load_dataset_bin(self):
name = 'lfw'
path = os.path.join(self.lfw_dir, name+".bin")
bins, issame_list = pickle.load(open(path, 'rb'))
data_list = []
for flip in [0,1]:
data = nd.empty((len(issame_list)*2, 3, self.image_size[0], self.image_size[1]))
data_list.append(data)
for i in xrange(len(issame_list)*2):
_bin = bins[i]
img = mx.image.imdecode(_bin)
img = nd.transpose(img, axes=(2, 0, 1))
for flip in [0,1]:
if flip==1:
img = mx.ndarray.flip(data=img, axis=2)
data_list[flip][i][:] = img
if i%1000==0:
print('loading bin', i)
print(data_list[0].shape)
return (data_list, issame_list)
Example #3
| Source File: verification.py From 1.FaceRecognition with MIT License | 6 votes |
|
def load_bin(path, image_size):
try:
with open(path, 'rb') as f:
bins, issame_list = pickle.load(f) # py2
except UnicodeDecodeError as e:
with open(path, 'rb') as f:
bins, issame_list = pickle.load(f, encoding='bytes') # py3
data_list = []
for flip in [0, 1]:
data = nd.empty((len(issame_list) * 2, 3, image_size[0], image_size[1]))
data_list.append(data)
for i in range(len(issame_list) * 2):
_bin = bins[i]
img = mx.image.imdecode(_bin)
if img.shape[1] != image_size[0]:
img = mx.image.resize_short(img, image_size[0])
img = nd.transpose(img, axes=(2, 0, 1))
for flip in [0, 1]:
if flip == 1:
img = mx.ndarray.flip(data=img, axis=2)
data_list[flip][i][:] = img
if i % 1000 == 0:
print('loading bin', i)
print(data_list[0].shape)
return (data_list, issame_list)
Example #4
| Source File: lfw.py From 1.FaceRecognition with MIT License | 6 votes |
|
def load_dataset(lfw_dir, image_size):
lfw_pairs = read_pairs(os.path.join(lfw_dir, 'pairs.txt'))
lfw_paths, issame_list = get_paths(lfw_dir, lfw_pairs, 'jpg')
lfw_data_list = []
for flip in [0,1]:
lfw_data = nd.empty((len(lfw_paths), 3, image_size[0], image_size[1]))
lfw_data_list.append(lfw_data)
i = 0
for path in lfw_paths:
with open(path, 'rb') as fin:
_bin = fin.read()
img = mx.image.imdecode(_bin)
img = nd.transpose(img, axes=(2, 0, 1))
for flip in [0,1]:
if flip==1:
img = mx.ndarray.flip(data=img, axis=2)
lfw_data_list[flip][i][:] = img
i+=1
if i%1000==0:
print('loading lfw', i)
print(lfw_data_list[0].shape)
print(lfw_data_list[1].shape)
return (lfw_data_list, issame_list)
Example #5
| Source File: utils.py From training_results_v0.6 with Apache License 2.0 | 6 votes |
|
def get_executor(sym, ctx, data_inputs, initializer=None):
data_shapes = {k: v.shape for k, v in data_inputs.items()}
arg_names = sym.list_arguments()
aux_names = sym.list_auxiliary_states()
param_names = list(set(arg_names) - set(data_inputs.keys()))
arg_shapes, output_shapes, aux_shapes = sym.infer_shape(**data_shapes)
arg_name_shape = {k: s for k, s in zip(arg_names, arg_shapes)}
params = {n: nd.empty(arg_name_shape[n], ctx=ctx) for n in param_names}
params_grad = {n: nd.empty(arg_name_shape[n], ctx=ctx) for n in param_names}
aux_states = {k: nd.empty(s, ctx=ctx) for k, s in zip(aux_names, aux_shapes)}
exe = sym.bind(ctx=ctx, args=dict(params, **data_inputs),
args_grad=params_grad,
aux_states=aux_states)
if initializer is not None:
for k, v in params.items():
initializer(k, v)
return exe, params, params_grad, aux_states
Example #6
| Source File: verification.py From 1.FaceRecognition with MIT License | 6 votes |
|
def load_bin(path, image_size):
bins, issame_list = pickle.load(open(path, 'rb'))
data_list = []
for flip in [0,1]:
data = nd.empty((len(issame_list)*2, 3, image_size[0], image_size[1]))
data_list.append(data)
for i in xrange(len(issame_list)*2):
_bin = bins[i]
img = mx.image.imdecode(_bin)
if img.shape[1]!=image_size[0]:
img = mx.image.resize_short(img, image_size[0])
img = nd.transpose(img, axes=(2, 0, 1))
for flip in [0,1]:
if flip==1:
img = mx.ndarray.flip(data=img, axis=2)
data_list[flip][i][:] = img
if i%1000==0:
print('loading bin', i)
print(data_list[0].shape)
return (data_list, issame_list)
Example #7
| Source File: verification.py From MaskInsightface with Apache License 2.0 | 6 votes |
|
def load_bin(path, image_size):
bins, issame_list = pickle.load(open(path, 'rb'), encoding='bytes')
data_list = []
for flip in [0,1]:
data = nd.empty((len(issame_list)*2, 3, image_size[0], image_size[1]))
data_list.append(data)
for i in range(len(issame_list)*2):
_bin = bins[i]
img = mx.image.imdecode(_bin)
img = nd.transpose(img, axes=(2, 0, 1))
for flip in [0,1]:
if flip==1:
img = mx.ndarray.flip(data=img, axis=2)
data_list[flip][i][:] = img
if i%1000==0:
print('loading bin', i)
print(data_list[0].shape)
return (data_list, issame_list)
Example #8
| Source File: verification.py From insightface with MIT License | 6 votes |
|
def load_bin(path, image_size):
bins, issame_list = pickle.load(open(path, 'rb'))
data_list = []
for flip in [0,1]:
data = nd.empty((len(issame_list)*2, 3, image_size[0], image_size[1]))
data_list.append(data)
for i in xrange(len(issame_list)*2):
_bin = bins[i]
img = mx.image.imdecode(_bin)
if img.shape[1]!=image_size[0]:
img = mx.image.resize_short(img, image_size[0])
img = nd.transpose(img, axes=(2, 0, 1))
for flip in [0,1]:
if flip==1:
img = mx.ndarray.flip(data=img, axis=2)
data_list[flip][i][:] = img
if i%1000==0:
print('loading bin', i)
print(data_list[0].shape)
return (data_list, issame_list)
Example #9
| Source File: lfw.py From MaskInsightface with Apache License 2.0 | 6 votes |
|
def load_dataset(lfw_dir, image_size):
lfw_pairs = read_pairs(os.path.join(lfw_dir, 'pairs.txt'))
lfw_paths, issame_list = get_paths(lfw_dir, lfw_pairs, 'jpg')
lfw_data_list = []
for flip in [0,1]:
lfw_data = nd.empty((len(lfw_paths), 3, image_size[0], image_size[1]))
lfw_data_list.append(lfw_data)
i = 0
for path in lfw_paths:
with open(path, 'rb') as fin:
_bin = fin.read()
img = mx.image.imdecode(_bin)
img = nd.transpose(img, axes=(2, 0, 1))
for flip in [0,1]:
if flip==1:
img = mx.ndarray.flip(data=img, axis=2)
lfw_data_list[flip][i][:] = img
i+=1
if i%1000==0:
print('loading lfw', i)
print(lfw_data_list[0].shape)
print(lfw_data_list[1].shape)
return (lfw_data_list, issame_list)
Example #10
| Source File: lfw.py From insightface with MIT License | 6 votes |
|
def load_dataset(lfw_dir, image_size):
lfw_pairs = read_pairs(os.path.join(lfw_dir, 'pairs.txt'))
lfw_paths, issame_list = get_paths(lfw_dir, lfw_pairs, 'jpg')
lfw_data_list = []
for flip in [0,1]:
lfw_data = nd.empty((len(lfw_paths), 3, image_size[0], image_size[1]))
lfw_data_list.append(lfw_data)
i = 0
for path in lfw_paths:
with open(path, 'rb') as fin:
_bin = fin.read()
img = mx.image.imdecode(_bin)
img = nd.transpose(img, axes=(2, 0, 1))
for flip in [0,1]:
if flip==1:
img = mx.ndarray.flip(data=img, axis=2)
lfw_data_list[flip][i][:] = img
i+=1
if i%1000==0:
print('loading lfw', i)
print(lfw_data_list[0].shape)
print(lfw_data_list[1].shape)
return (lfw_data_list, issame_list)
Example #11
| Source File: verification.py From insightface with MIT License | 6 votes |
|
def load_bin(path, image_size):
try:
with open(path, 'rb') as f:
bins, issame_list = pickle.load(f) #py2
except UnicodeDecodeError as e:
with open(path, 'rb') as f:
bins, issame_list = pickle.load(f, encoding='bytes') #py3
data_list = []
for flip in [0,1]:
data = nd.empty((len(issame_list)*2, 3, image_size[0], image_size[1]))
data_list.append(data)
for i in range(len(issame_list)*2):
_bin = bins[i]
img = mx.image.imdecode(_bin)
if img.shape[1]!=image_size[0]:
img = mx.image.resize_short(img, image_size[0])
img = nd.transpose(img, axes=(2, 0, 1))
for flip in [0,1]:
if flip==1:
img = mx.ndarray.flip(data=img, axis=2)
data_list[flip][i][:] = img
if i%1000==0:
print('loading bin', i)
print(data_list[0].shape)
return (data_list, issame_list)
Example #12
| Source File: utils.py From dynamic-training-with-apache-mxnet-on-aws with Apache License 2.0 | 6 votes |
|
def get_executor(sym, ctx, data_inputs, initializer=None):
data_shapes = {k: v.shape for k, v in data_inputs.items()}
arg_names = sym.list_arguments()
aux_names = sym.list_auxiliary_states()
param_names = list(set(arg_names) - set(data_inputs.keys()))
arg_shapes, output_shapes, aux_shapes = sym.infer_shape(**data_shapes)
arg_name_shape = {k: s for k, s in zip(arg_names, arg_shapes)}
params = {n: nd.empty(arg_name_shape[n], ctx=ctx) for n in param_names}
params_grad = {n: nd.empty(arg_name_shape[n], ctx=ctx) for n in param_names}
aux_states = {k: nd.empty(s, ctx=ctx) for k, s in zip(aux_names, aux_shapes)}
exe = sym.bind(ctx=ctx, args=dict(params, **data_inputs),
args_grad=params_grad,
aux_states=aux_states)
if initializer is not None:
for k, v in params.items():
initializer(k, v)
return exe, params, params_grad, aux_states
Example #13
| Source File: bdk_demo.py From SNIPER-mxnet with Apache License 2.0 | 6 votes |
|
def synthetic_grad(X, theta, sigma1, sigma2, sigmax, rescale_grad=1.0, grad=None):
if grad is None:
grad = nd.empty(theta.shape, theta.context)
theta1 = theta.asnumpy()[0]
theta2 = theta.asnumpy()[1]
v1 = sigma1 ** 2
v2 = sigma2 ** 2
vx = sigmax ** 2
denominator = numpy.exp(-(X - theta1) ** 2 / (2 * vx)) + numpy.exp(
-(X - theta1 - theta2) ** 2 / (2 * vx))
grad_npy = numpy.zeros(theta.shape)
grad_npy[0] = -rescale_grad * ((numpy.exp(-(X - theta1) ** 2 / (2 * vx)) * (X - theta1) / vx
+ numpy.exp(-(X - theta1 - theta2) ** 2 / (2 * vx)) * (
X - theta1 - theta2) / vx) / denominator).sum() \
+ theta1 / v1
grad_npy[1] = -rescale_grad * ((numpy.exp(-(X - theta1 - theta2) ** 2 / (2 * vx)) * (
X - theta1 - theta2) / vx) / denominator).sum() \
+ theta2 / v2
grad[:] = grad_npy
return grad
Example #14
| Source File: bdk_demo.py From training_results_v0.6 with Apache License 2.0 | 6 votes |
|
def synthetic_grad(X, theta, sigma1, sigma2, sigmax, rescale_grad=1.0, grad=None):
if grad is None:
grad = nd.empty(theta.shape, theta.context)
theta1 = theta.asnumpy()[0]
theta2 = theta.asnumpy()[1]
v1 = sigma1 ** 2
v2 = sigma2 ** 2
vx = sigmax ** 2
denominator = numpy.exp(-(X - theta1) ** 2 / (2 * vx)) + numpy.exp(
-(X - theta1 - theta2) ** 2 / (2 * vx))
grad_npy = numpy.zeros(theta.shape)
grad_npy[0] = -rescale_grad * ((numpy.exp(-(X - theta1) ** 2 / (2 * vx)) * (X - theta1) / vx
+ numpy.exp(-(X - theta1 - theta2) ** 2 / (2 * vx)) * (
X - theta1 - theta2) / vx) / denominator).sum() \
+ theta1 / v1
grad_npy[1] = -rescale_grad * ((numpy.exp(-(X - theta1 - theta2) ** 2 / (2 * vx)) * (
X - theta1 - theta2) / vx) / denominator).sum() \
+ theta2 / v2
grad[:] = grad_npy
return grad
Example #15
| Source File: bdk_demo.py From dynamic-training-with-apache-mxnet-on-aws with Apache License 2.0 | 6 votes |
|
def synthetic_grad(X, theta, sigma1, sigma2, sigmax, rescale_grad=1.0, grad=None):
if grad is None:
grad = nd.empty(theta.shape, theta.context)
theta1 = theta.asnumpy()[0]
theta2 = theta.asnumpy()[1]
v1 = sigma1 ** 2
v2 = sigma2 ** 2
vx = sigmax ** 2
denominator = numpy.exp(-(X - theta1) ** 2 / (2 * vx)) + numpy.exp(
-(X - theta1 - theta2) ** 2 / (2 * vx))
grad_npy = numpy.zeros(theta.shape)
grad_npy[0] = -rescale_grad * ((numpy.exp(-(X - theta1) ** 2 / (2 * vx)) * (X - theta1) / vx
+ numpy.exp(-(X - theta1 - theta2) ** 2 / (2 * vx)) * (
X - theta1 - theta2) / vx) / denominator).sum() \
+ theta1 / v1
grad_npy[1] = -rescale_grad * ((numpy.exp(-(X - theta1 - theta2) ** 2 / (2 * vx)) * (
X - theta1 - theta2) / vx) / denominator).sum() \
+ theta2 / v2
grad[:] = grad_npy
return grad
Example #16
| Source File: lfw_comparison_and_plot_roc.py From MobileFace with MIT License | 5 votes |
|
def load_dataset(self):
lfw_pairs = self.read_pairs(os.path.join(self.lfw_dir, 'pairs.txt'))
lfw_paths, issame_list = self.get_paths(self.lfw_dir, lfw_pairs, 'jpg')
lfw_data_list = []
for flip in [0,1]:
# lfw_data = nd.empty((len(lfw_paths), 3, image_size[0], image_size[1]))
lfw_data = nd.empty((len(lfw_paths), 1, 100, 100))
lfw_data_list.append(lfw_data)
i = 0
for path in lfw_paths:
with open(path, 'rb') as fin:
_bin = fin.read()
img = np.asarray(bytearray(_bin), dtype="uint8")
img = cv2.imdecode(img, 0) # (100, 100)
img = img.reshape((1, img.shape[0], img.shape[1])) # (1, 100, 100)
#img = nd.transpose(img, axes=(2, 0, 1)) # (1L, 100L, 100L)
img = mx.nd.array(img) # (1L, 100L, 100L)
for flip in [0,1]:
if flip==1:
img = mx.ndarray.flip(data=img, axis=2)
lfw_data_list[flip][i][:] = img
i+=1
if i%1000==0:
print('loading lfw', i)
print(lfw_data_list[0].shape)
print(lfw_data_list[1].shape)
return (lfw_data_list, issame_list)
Example #17
| Source File: utils.py From training_results_v0.6 with Apache License 2.0 | 5 votes |
|
def copy_param(exe, new_param=None):
if new_param is None:
new_param = {k: nd.empty(v.shape, ctx=mx.cpu()) for k,v in exe.arg_dict.items()}
for k, v in new_param.items():
exe.arg_dict[k].copyto(v)
return new_param
Example #18
| Source File: data.py From 1.FaceRecognition with MIT License | 5 votes |
|
def next(self):
"""Returns the next batch of data."""
#print('next')
batch_size = self.batch_size
batch_data = nd.empty((batch_size,)+self.data_shape)
batch_label = nd.empty((batch_size,)+self.label_shape)
i = 0
#self.cutoff = random.randint(800,1280)
try:
while i < batch_size:
#print('N', i)
data, label = self.next_sample()
data = nd.array(data)
data = nd.transpose(data, axes=(2, 0, 1))
label = nd.array(label)
label = nd.transpose(label, axes=(2, 0, 1))
batch_data[i][:] = data
batch_label[i][:] = label
i += 1
except StopIteration:
if i<batch_size:
raise StopIteration
#return {self.data_name : batch_data,
# self.label_name : batch_label}
#print(batch_data.shape, batch_label.shape)
return mx.io.DataBatch([batch_data], [batch_label, self.weight_mask], batch_size - i)
Example #19
| Source File: utils.py From SNIPER-mxnet with Apache License 2.0 | 5 votes |
|
def copy_param(exe, new_param=None):
if new_param is None:
new_param = {k: nd.empty(v.shape, ctx=mx.cpu()) for k,v in exe.arg_dict.items()}
for k, v in new_param.items():
exe.arg_dict[k].copyto(v)
return new_param
Example #20
| Source File: bdk_demo.py From SNIPER-mxnet with Apache License 2.0 | 5 votes |
|
def run_synthetic_SGLD():
theta1 = 0
theta2 = 1
sigma1 = numpy.sqrt(10)
sigma2 = 1
sigmax = numpy.sqrt(2)
X = load_synthetic(theta1=theta1, theta2=theta2, sigmax=sigmax, num=100)
minibatch_size = 1
total_iter_num = 1000000
lr_scheduler = SGLDScheduler(begin_rate=0.01, end_rate=0.0001, total_iter_num=total_iter_num,
factor=0.55)
optimizer = mx.optimizer.create('sgld',
learning_rate=None,
rescale_grad=1.0,
lr_scheduler=lr_scheduler,
wd=0)
updater = mx.optimizer.get_updater(optimizer)
theta = mx.random.normal(0, 1, (2,), mx.cpu())
grad = nd.empty((2,), mx.cpu())
samples = numpy.zeros((2, total_iter_num))
start = time.time()
for i in xrange(total_iter_num):
if (i + 1) % 100000 == 0:
end = time.time()
print("Iter:%d, Time spent: %f" % (i + 1, end - start))
start = time.time()
ind = numpy.random.randint(0, X.shape[0])
synthetic_grad(X[ind], theta, sigma1, sigma2, sigmax, rescale_grad=
X.shape[0] / float(minibatch_size), grad=grad)
updater('theta', grad, theta)
samples[:, i] = theta.asnumpy()
plt.hist2d(samples[0, :], samples[1, :], (200, 200), cmap=plt.cm.jet)
plt.colorbar()
plt.show()
Example #21
| Source File: base.py From SNIPER-mxnet with Apache License 2.0 | 5 votes |
|
def __init__(self, data_shapes, sym_gen, params=None, aux_states=None,
default_bucket_kwargs=None, learn_init_keys=None,
initializer=mx.init.Xavier(factor_type="in", rnd_type="gaussian", magnitude=2),
ctx=mx.gpu(), name='Net'):
self.sym_gen = sym_gen
bucket_kwargs = default_bucket_kwargs.copy() if \
default_bucket_kwargs is not None else dict()
self.curr_bucket_key = None
self.ctx = ctx
self.name = name
self.initializer = initializer
if params is None:
self.params = None
self.params_grad = None
else:
self.params = OrderedDict([(k, v.copyto(ctx)) for k, v in params.items()])
self.params_grad = OrderedDict([(n, nd.empty(v.shape, ctx=ctx))
for n, v in self.params.items()])
if aux_states is not None:
self.aux_states = OrderedDict([(k, v.copyto(ctx)) for k, v in aux_states.items()])
else:
self.aux_states = None
self._buckets = dict()
self.learn_init_keys = learn_init_keys if learn_init_keys is not None else []
self.learn_init_key_shapes = {k: data_shapes[k] for k in self.learn_init_keys}
self.switch_bucket(bucket_kwargs=bucket_kwargs, data_shapes=data_shapes)
self.acc_grad = None
Example #22
| Source File: base.py From training_results_v0.6 with Apache License 2.0 | 5 votes |
|
def __init__(self, data_shapes, sym_gen, params=None, aux_states=None,
default_bucket_kwargs=None, learn_init_keys=None,
initializer=mx.init.Xavier(factor_type="in", rnd_type="gaussian", magnitude=2),
ctx=mx.gpu(), name='Net'):
self.sym_gen = sym_gen
bucket_kwargs = default_bucket_kwargs.copy() if \
default_bucket_kwargs is not None else dict()
self.curr_bucket_key = None
self.ctx = ctx
self.name = name
self.initializer = initializer
if params is None:
self.params = None
self.params_grad = None
else:
self.params = OrderedDict([(k, v.copyto(ctx)) for k, v in params.items()])
self.params_grad = OrderedDict([(n, nd.empty(v.shape, ctx=ctx))
for n, v in self.params.items()])
if aux_states is not None:
self.aux_states = OrderedDict([(k, v.copyto(ctx)) for k, v in aux_states.items()])
else:
self.aux_states = None
self._buckets = dict()
self.learn_init_keys = learn_init_keys if learn_init_keys is not None else []
self.learn_init_key_shapes = {k: data_shapes[k] for k in self.learn_init_keys}
self.switch_bucket(bucket_kwargs=bucket_kwargs, data_shapes=data_shapes)
self.acc_grad = None
Example #23
| Source File: bdk_demo.py From training_results_v0.6 with Apache License 2.0 | 5 votes |
|
def run_synthetic_SGLD():
theta1 = 0
theta2 = 1
sigma1 = numpy.sqrt(10)
sigma2 = 1
sigmax = numpy.sqrt(2)
X = load_synthetic(theta1=theta1, theta2=theta2, sigmax=sigmax, num=100)
minibatch_size = 1
total_iter_num = 1000000
lr_scheduler = SGLDScheduler(begin_rate=0.01, end_rate=0.0001, total_iter_num=total_iter_num,
factor=0.55)
optimizer = mx.optimizer.create('sgld',
learning_rate=None,
rescale_grad=1.0,
lr_scheduler=lr_scheduler,
wd=0)
updater = mx.optimizer.get_updater(optimizer)
theta = mx.random.normal(0, 1, (2,), mx.cpu())
grad = nd.empty((2,), mx.cpu())
samples = numpy.zeros((2, total_iter_num))
start = time.time()
for i in xrange(total_iter_num):
if (i + 1) % 100000 == 0:
end = time.time()
print("Iter:%d, Time spent: %f" % (i + 1, end - start))
start = time.time()
ind = numpy.random.randint(0, X.shape[0])
synthetic_grad(X[ind], theta, sigma1, sigma2, sigmax, rescale_grad=
X.shape[0] / float(minibatch_size), grad=grad)
updater('theta', grad, theta)
samples[:, i] = theta.asnumpy()
plt.hist2d(samples[0, :], samples[1, :], (200, 200), cmap=plt.cm.jet)
plt.colorbar()
plt.show()
Example #24
| Source File: tensor.py From dgl with Apache License 2.0 | 5 votes |
|
def backward(self, grad_out):
in_data_nd, out_data_nd, degs = self.saved_tensors
grad_in = nd.empty(in_data_nd.shape, ctx=grad_out.context,
dtype=grad_out.dtype)
if self.reducer == 'mean':
grad_out = grad_out / degs
grad_out_nd = zerocopy_to_dgl_ndarray(grad_out)
K.backward_copy_reduce(
self.reducer if self.reducer != 'mean' else 'sum',
self.graph, self.target, in_data_nd, out_data_nd,
grad_out_nd, zerocopy_to_dgl_ndarray_for_write(grad_in),
self.in_map[1], self.out_map[1])
# clear saved tensors explicitly
self.saved_tensors = None
return grad_in
Example #25
| Source File: tensor.py From dgl with Apache License 2.0 | 5 votes |
|
def forward(self, in_data):
feat_shape = in_data.shape[1:]
out_data = nd.empty((self.out_size,) + feat_shape,
ctx=in_data.context, dtype=in_data.dtype)
in_data_nd = zerocopy_to_dgl_ndarray(in_data)
out_data_nd = zerocopy_to_dgl_ndarray_for_write(out_data)
K.copy_reduce(
self.reducer if self.reducer != 'mean' else 'sum',
self.graph, self.target, in_data_nd, out_data_nd,
self.in_map[0], self.out_map[0])
# normalize if mean reducer
# NOTE(zihao): this is a temporary hack and we should have better solution in the future.
if self.reducer == 'mean':
in_ones = nd.ones((in_data.shape[0],),
ctx=in_data.context, dtype=in_data.dtype)
degs = nd.empty((out_data.shape[0],),
ctx=out_data.context, dtype=out_data.dtype)
in_ones_nd = zerocopy_to_dgl_ndarray(in_ones)
degs_nd = zerocopy_to_dgl_ndarray(degs)
K.copy_reduce(
'sum', self.graph, self.target, in_ones_nd, degs_nd,
self.in_map[0], self.out_map[0])
# reshape
degs = degs.reshape((out_data.shape[0],) + (1,) * (out_data.ndim - 1)).clip(1, float('inf'))
out_data = out_data / degs
else:
degs = None
self.save_for_backward(in_data_nd, out_data_nd, degs)
return out_data
Example #26
| Source File: tensor.py From dgl with Apache License 2.0 | 5 votes |
|
def backward(self, grad_out):
lhs_data_nd, rhs_data_nd, out_data_nd, feat_shape, degs = self.saved_tensors
if self.reducer == 'mean':
grad_out = grad_out / degs
grad_out_nd = zerocopy_to_dgl_ndarray(grad_out)
grad_lhs = nd.empty((lhs_data_nd.shape[0],) + feat_shape,
ctx=grad_out.context, dtype=grad_out.dtype)
K.backward_lhs_binary_op_reduce(
self.reducer if self.reducer != 'mean' else 'sum',
self.binary_op, self.graph, self.lhs, self.rhs,
lhs_data_nd, rhs_data_nd, out_data_nd, grad_out_nd,
zerocopy_to_dgl_ndarray_for_write(grad_lhs), self.lhs_map[1],
self.rhs_map[1], self.out_map[1])
grad_lhs = _reduce_grad(grad_lhs, lhs_data_nd.shape)
grad_rhs = nd.empty((rhs_data_nd.shape[0],) + feat_shape,
ctx=grad_out.context, dtype=grad_out.dtype)
K.backward_rhs_binary_op_reduce(
self.reducer if self.reducer != 'mean' else 'sum',
self.binary_op, self.graph, self.lhs, self.rhs,
lhs_data_nd, rhs_data_nd, out_data_nd, grad_out_nd,
zerocopy_to_dgl_ndarray_for_write(grad_rhs), self.lhs_map[1],
self.rhs_map[1], self.out_map[1])
grad_rhs = _reduce_grad(grad_rhs, rhs_data_nd.shape)
# clear saved tensors explicitly
self.saved_tensors = None
return grad_lhs, grad_rhs
Example #27
| Source File: tensor.py From dgl with Apache License 2.0 | 5 votes |
|
def gather_row(data, row_index):
# MXNet workaround for empty row index
if len(row_index) == 0:
if data.shape[0] == 0:
return data
else:
return data[0:0]
if isinstance(row_index, nd.NDArray):
return nd.take(data, row_index)
else:
return data[row_index,]
Example #28
| Source File: data.py From insightface with MIT License | 5 votes |
|
def next(self):
"""Returns the next batch of data."""
#print('next')
batch_size = self.batch_size
batch_data = nd.empty((batch_size,)+self.data_shape)
batch_label = nd.empty((batch_size,)+self.label_shape)
i = 0
#self.cutoff = random.randint(800,1280)
try:
while i < batch_size:
#print('N', i)
data, label = self.next_sample()
data = nd.array(data)
data = nd.transpose(data, axes=(2, 0, 1))
label = nd.array(label)
label = nd.transpose(label, axes=(2, 0, 1))
batch_data[i][:] = data
batch_label[i][:] = label
i += 1
except StopIteration:
if i<batch_size:
raise StopIteration
#return {self.data_name : batch_data,
# self.label_name : batch_label}
#print(batch_data.shape, batch_label.shape)
return mx.io.DataBatch([batch_data], [batch_label, self.weight_mask], batch_size - i)
Example #29
| Source File: base.py From dynamic-training-with-apache-mxnet-on-aws with Apache License 2.0 | 5 votes |
|
def __init__(self, data_shapes, sym_gen, params=None, aux_states=None,
default_bucket_kwargs=None, learn_init_keys=None,
initializer=mx.init.Xavier(factor_type="in", rnd_type="gaussian", magnitude=2),
ctx=mx.gpu(), name='Net'):
self.sym_gen = sym_gen
bucket_kwargs = default_bucket_kwargs.copy() if \
default_bucket_kwargs is not None else dict()
self.curr_bucket_key = None
self.ctx = ctx
self.name = name
self.initializer = initializer
if params is None:
self.params = None
self.params_grad = None
else:
self.params = OrderedDict([(k, v.copyto(ctx)) for k, v in params.items()])
self.params_grad = OrderedDict([(n, nd.empty(v.shape, ctx=ctx))
for n, v in self.params.items()])
if aux_states is not None:
self.aux_states = OrderedDict([(k, v.copyto(ctx)) for k, v in aux_states.items()])
else:
self.aux_states = None
self._buckets = dict()
self.learn_init_keys = learn_init_keys if learn_init_keys is not None else []
self.learn_init_key_shapes = {k: data_shapes[k] for k in self.learn_init_keys}
self.switch_bucket(bucket_kwargs=bucket_kwargs, data_shapes=data_shapes)
self.acc_grad = None
Example #30
| Source File: bdk_demo.py From dynamic-training-with-apache-mxnet-on-aws with Apache License 2.0 | 5 votes |
|
def run_synthetic_SGLD():
theta1 = 0
theta2 = 1
sigma1 = numpy.sqrt(10)
sigma2 = 1
sigmax = numpy.sqrt(2)
X = load_synthetic(theta1=theta1, theta2=theta2, sigmax=sigmax, num=100)
minibatch_size = 1
total_iter_num = 1000000
lr_scheduler = SGLDScheduler(begin_rate=0.01, end_rate=0.0001, total_iter_num=total_iter_num,
factor=0.55)
optimizer = mx.optimizer.create('sgld',
learning_rate=None,
rescale_grad=1.0,
lr_scheduler=lr_scheduler,
wd=0)
updater = mx.optimizer.get_updater(optimizer)
theta = mx.random.normal(0, 1, (2,), mx.cpu())
grad = nd.empty((2,), mx.cpu())
samples = numpy.zeros((2, total_iter_num))
start = time.time()
for i in xrange(total_iter_num):
if (i + 1) % 100000 == 0:
end = time.time()
print("Iter:%d, Time spent: %f" % (i + 1, end - start))
start = time.time()
ind = numpy.random.randint(0, X.shape[0])
synthetic_grad(X[ind], theta, sigma1, sigma2, sigmax, rescale_grad=
X.shape[0] / float(minibatch_size), grad=grad)
updater('theta', grad, theta)
samples[:, i] = theta.asnumpy()
plt.hist2d(samples[0, :], samples[1, :], (200, 200), cmap=plt.cm.jet)
plt.colorbar()
plt.show()
