Skip to content

Fixed bug related to KerasSymbol.tensor #93

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
jricheimer wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Fixed bug related to KerasSymbol.tensor #93

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9d09f0e
Fixed bug related to KerasSymbol.tensor
jricheimer Nov 19, 2017
ce149a3
fixed variable naming issue
jricheimer Nov 22, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 13 additions & 5 deletions keras/backend/mxnet_backend.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -203,7 +203,11 @@ def __init__(self, symbol, name=None, neighbor=None, is_var=False):


def bind(self, data):
self.tensor = data
if not hasattr(self, 'tensor'):
self.tensor = data
else:
self.tensor[:] = data

if self.name in self._bind_values:
assert self._bind_values[self.name].shape == data.shape, \
"Redefinition of variable %s" % self.name
Expand Down Expand Up @@ -884,7 +888,8 @@ def random_uniform_variable(shape, low, high, dtype=None,
value = mx.random.uniform(low=low, high=high, dtype='float32', shape=shape)
if dtype != np.float32:
value = mx.nd.Cast(value, dtype=dtype)
name = _autogen_name("uniform")
if name is None:
name = _autogen_name("uniform")
ret = KerasVariable(name, value.shape, value.dtype)
ret.bind(value)
return ret
Expand Down Expand Up @@ -923,7 +928,8 @@ def random_normal_variable(shape, mean, scale, dtype=None,
value = mx.random.normal(loc=mean, scale=scale, dtype='float32', shape=shape)
if dtype != np.float32:
value = mx.nd.Cast(value, dtype=dtype)
name = _autogen_name("normal")
if name is None:
name = _autogen_name("normal")
ret = KerasVariable(name, value.shape, value.dtype)
ret.bind(value)
return ret
Expand Down Expand Up @@ -2956,7 +2962,8 @@ def random_normal(shape, mean=0.0, std=1.0, dtype=None, seed=None):
if dtype is None:
dtype = floatx()
dtype = np.dtype(dtype)
name = _autogen_name('normal')
if name is None:
name = _autogen_name('normal')
sym = mx.sym.normal(loc=mean, scale=std, shape=shape, dtype='float32', name=name)
if dtype != np.float32:
sym = mx.sym.Cast(data=sym, dtype=dtype)
Expand All @@ -2983,7 +2990,8 @@ def random_uniform(shape, low=0.0, high=1.0, dtype=None, seed=None):
if dtype is None:
dtype = floatx()
dtype = np.dtype(dtype)
name = _autogen_name('uniform')
if name is None:
name = _autogen_name('uniform')
sym = mx.sym.uniform(low=low, high=high, shape=shape, dtype='float32', name=name)
if dtype != np.float32:
sym = mx.sym.Cast(data=sym, dtype=dtype)
Expand Down