| /external/tensorflow/tensorflow/python/keras/layers/ |
| D | wrappers.py | 140 def _get_shape_tuple(self, init_tuple, tensor, start_idx, int_shape=None): argument
160 if int_shape is None:
161 int_shape = K.int_shape(tensor)[start_idx:]
162 if isinstance(int_shape, tensor_shape.TensorShape):
163 int_shape = int_shape.as_list()
164 if not any(not s for s in int_shape):
165 return init_tuple + tuple(int_shape)
167 int_shape = list(int_shape)
168 for i, s in enumerate(int_shape):
170 int_shape[i] = shape[start_idx + i]
[all …]
|
| D | lstm_test.py | 262 np.zeros(keras.backend.int_shape(layer.states[0])),
264 state_shapes = [keras.backend.int_shape(state) for state in layer.states]
271 np.ones(keras.backend.int_shape(layer.states[0])),
|
| D | recurrent_v2.py | 434 input_shape = K.int_shape(inputs)
584 input_shape = K.int_shape(inputs)
1160 input_shape = K.int_shape(inputs)
1371 input_shape = K.int_shape(inputs)
|
| D | lstm_v2_test.py | 210 np.zeros(keras.backend.int_shape(layer.states[0])),
212 state_shapes = [keras.backend.int_shape(state) for state in layer.states]
219 np.ones(keras.backend.int_shape(layer.states[0])),
|
| D | recurrent.py | 683 lambda s: InputSpec(shape=K.int_shape(s)), initial_state)
688 InputSpec(shape=K.int_shape(constant)) for constant in constants
758 input_shape = K.int_shape(nest.flatten(inputs)[0])
760 input_shape = K.int_shape(inputs)
|
| D | convolutional_recurrent.py | 307 timesteps = K.int_shape(inputs)[1]
|
| /external/tensorflow/tensorflow/compiler/xla/tests/ |
| D | outfeed_in_nested_computation_test.cc | 32 Shape int_shape = ShapeUtil::MakeShape(xla::S32, {}); in XLA_TEST_F() local
34 ShapeUtil::MakeTupleShape({int_shape, state_tuple_array_shape}); in XLA_TEST_F()
38 XlaOp num_iter = Infeed(&b, int_shape); in XLA_TEST_F()
46 Outfeed(loop_counter, int_shape, ""); in XLA_TEST_F()
90 local_client_->TransferFromOutfeed(&int_shape)); in XLA_TEST_F()
111 local_client_->TransferFromOutfeed(&int_shape)); in XLA_TEST_F()
|
| /external/tensorflow/tensorflow/python/keras/applications/ |
| D | mobilenet_v2.py | 208 if backend.int_shape(input_tensor)[1] != input_shape[1]:
213 if backend.int_shape(input_tensor)[2] != input_shape[1]:
234 rows = backend.int_shape(input_tensor)[2]
235 cols = backend.int_shape(input_tensor)[3]
237 rows = backend.int_shape(input_tensor)[1]
238 cols = backend.int_shape(input_tensor)[2]
418 in_channels = backend.int_shape(inputs)[channel_axis]
|
| D | mobilenet_v3.py | 187 if backend.int_shape(input_tensor)[1] != input_shape[1]:
192 if backend.int_shape(input_tensor)[2] != input_shape[1]:
211 rows = backend.int_shape(input_tensor)[2]
212 cols = backend.int_shape(input_tensor)[3]
215 rows = backend.int_shape(input_tensor)[1]
216 cols = backend.int_shape(input_tensor)[2]
279 last_conv_ch = _depth(backend.int_shape(x)[channel_axis] * 6)
494 infilters = backend.int_shape(x)[channel_axis]
|
| D | inception_resnet_v2.py | 362 backend.int_shape(x)[channel_axis],
370 output_shape=backend.int_shape(x)[1:],
|
| D | nasnet.py | 553 ip_shape = backend.int_shape(ip)
556 p_shape = backend.int_shape(p)
|
| D | imagenet_utils.py | 399 input_size = backend.int_shape(inputs)[img_dim:(img_dim + 2)]
|
| D | densenet.py | 90 int(backend.int_shape(x)[bn_axis] * reduction),
|
| /external/tensorflow/tensorflow/python/keras/ |
| D | optimizer_v1.py | 191 shapes = [K.int_shape(p) for p in params]
262 accumulators = [K.zeros(K.int_shape(p), dtype=K.dtype(p)) for p in params]
335 shapes = [K.int_shape(p) for p in params]
415 shapes = [K.int_shape(p) for p in params]
501 ms = [K.zeros(K.int_shape(p), dtype=K.dtype(p)) for p in params]
502 vs = [K.zeros(K.int_shape(p), dtype=K.dtype(p)) for p in params]
504 vhats = [K.zeros(K.int_shape(p), dtype=K.dtype(p)) for p in params]
596 shapes = [K.int_shape(p) for p in params]
686 shapes = [K.int_shape(p) for p in params]
|
| D | callbacks_v1.py | 189 shape = K.int_shape(w_img)
193 shape = K.int_shape(w_img)
200 shape = K.int_shape(w_img)
209 shape = K.int_shape(w_img)
|
| D | backend.py | 1096 v._keras_shape = int_shape(value)
1435 def int_shape(x): function
1990 for i, s in zip(int_shape(x), array_ops.unstack(array_ops.shape(x))):
1997 for i, s in zip(int_shape(y), array_ops.unstack(array_ops.shape(y))):
2064 x_shape = int_shape(x)
2065 y_shape = int_shape(y)
3198 original_shape = int_shape(x)
5935 kernel_shape = int_shape(kernel)
6075 bias_shape = int_shape(bias)
|
| D | callbacks.py | 2481 shape = K.int_shape(w_img)
2487 shape = K.int_shape(w_img)
2494 shape = K.int_shape(w_img)
2497 shape = K.int_shape(w_img)
|
| D | metrics.py | 3380 K.int_shape(y_true)) == len(K.int_shape(y_pred))):
|
| D | backend_test.py | 220 self.assertEqual(backend.int_shape(x), (3, 4))
224 self.assertEqual(backend.int_shape(x), (None, 4))
|
| /external/tensorflow/tensorflow/python/keras/engine/ |
| D | node.py | 244 input_shapes = nest.map_structure(backend.int_shape, self.input_tensors)
251 return nest.map_structure(backend.int_shape, self.output_tensors)
|
| D | functional.py | 268 return nest.map_structure(backend.int_shape, self.input)
328 return nest.map_structure(backend.int_shape, self.output)
|
| D | training_v1.py | 2668 self._feed_input_shapes.append(K.int_shape(v))
2932 return K.int_shape(self.output)
|
| /external/tensorflow/tensorflow/python/keras/saving/ |
| D | hdf5_format.py | 408 if K.int_shape(layer.weights[0]) != weights[0].shape:
787 if K.int_shape(symbolic_weights[i]) != weight_values[i].shape:
797 ' has shape {}'.format(K.int_shape(
|
| /external/tensorflow/tensorflow/tools/api/golden/v2/ |
| D | tensorflow.keras.backend.pbtxt | 240 name: "int_shape"
|
| /external/tensorflow/tensorflow/tools/api/golden/v1/ |
| D | tensorflow.keras.backend.pbtxt | 248 name: "int_shape"
|