| /external/tensorflow/tensorflow/contrib/nn/python/ops/ |
| D | sampling_ops.py | 31 def _rank_resample(weights, biases, inputs, sampled_values, num_resampled, argument
99 embedding_ops.embedding_lookup(biases, sampled, partition_strategy), [-1])
111 biases, argument
219 weights, biases, labels, inputs, sampled_values, resampling_temperature
230 resampled_values = _rank_resample(weights, biases, inputs, sampled_values,
235 biases=biases,
248 biases, argument
323 biases=biases,
|
| D | sampling_ops_test.py | 155 biases=self._biases(),
174 biases=self._biases(),
192 biases=self._biases(),
204 def _testCompareWithNN(self, weights, biases, partition_strategy): argument
208 biases=biases(),
221 biases=biases(),
275 biases = constant_op.constant([0., 0.])
280 biases=biases,
293 biases=biases,
|
| /external/tensorflow/tensorflow/examples/tutorials/mnist/ |
| D | mnist.py | 62 biases = tf.Variable(tf.zeros([hidden1_units]),
64 hidden1 = tf.nn.relu(tf.matmul(images, weights) + biases)
71 biases = tf.Variable(tf.zeros([hidden2_units]),
73 hidden2 = tf.nn.relu(tf.matmul(hidden1, weights) + biases)
80 biases = tf.Variable(tf.zeros([NUM_CLASSES]),
82 logits = tf.matmul(hidden2, weights) + biases
|
| D | mnist_with_summaries.py | 92 biases = bias_variable([output_dim])
93 variable_summaries(biases)
95 preactivate = tf.matmul(input_tensor, weights) + biases
|
| /external/tensorflow/tensorflow/lite/kernels/ |
| D | bidirectional_sequence_rnn_test.cc | 631 const std::initializer_list<float> biases = { variable
793 rnn.SetFwBias(biases); in TEST()
794 rnn.SetBwBias(biases); in TEST()
832 rnn.SetFwBias(biases); in TEST()
833 rnn.SetBwBias(biases); in TEST()
869 rnn.SetFwBias(biases); in TEST()
870 rnn.SetBwBias(biases); in TEST()
908 rnn.SetFwBias(biases); in TEST()
909 rnn.SetBwBias(biases); in TEST()
953 rnn.SetFwBias(biases); in TEST()
[all …]
|
| /external/tensorflow/tensorflow/contrib/model_pruning/examples/cifar10/ |
| D | cifar10_pruning.py | 194 biases = _variable_on_cpu('biases', [64], tf.constant_initializer(0.0))
195 pre_activation = tf.nn.bias_add(conv, biases)
216 biases = _variable_on_cpu('biases', [64], tf.constant_initializer(0.1))
217 pre_activation = tf.nn.bias_add(conv, biases)
239 biases = _variable_on_cpu('biases', [384], tf.constant_initializer(0.1))
241 tf.matmul(reshape, pruning.apply_mask(weights, scope)) + biases,
249 biases = _variable_on_cpu('biases', [192], tf.constant_initializer(0.1))
251 tf.matmul(local3, pruning.apply_mask(weights, scope)) + biases,
262 biases = _variable_on_cpu('biases', [NUM_CLASSES],
266 biases,
|
| /external/skqp/src/gpu/gradients/ |
| D | GrUnrolledBinaryGradientColorizer.fp | 137 // The raster implementation also uses scales and biases, but since they must be calculated
140 SkPMColor4f biases[kMaxIntervals];
169 bias.store(biases + intervalCount);
177 biases[i] = SK_PMColor4fTRANSPARENT;
183 scales[6], scales[7], biases[0], biases[1], biases[2], biases[3], biases[4],
184 biases[5], biases[6], biases[7],
|
| D | GrUnrolledBinaryGradientColorizer.cpp | 333 SkPMColor4f biases[kMaxIntervals]; in Make() local
362 bias.store(biases + intervalCount); in Make()
370 biases[i] = SK_PMColor4fTRANSPARENT; in Make()
376 scales[6], scales[7], biases[0], biases[1], biases[2], biases[3], biases[4], biases[5], in Make()
377 biases[6], biases[7], in Make()
|
| /external/skia/src/gpu/gradients/ |
| D | GrUnrolledBinaryGradientColorizer.fp | 137 // The raster implementation also uses scales and biases, but since they must be calculated
140 SkPMColor4f biases[kMaxIntervals];
169 bias.store(biases + intervalCount);
177 biases[i] = SK_PMColor4fTRANSPARENT;
183 scales[6], scales[7], biases[0], biases[1], biases[2], biases[3], biases[4],
184 biases[5], biases[6], biases[7],
|
| D | GrUnrolledBinaryGradientColorizer.cpp | 333 SkPMColor4f biases[kMaxIntervals]; in Make() local
362 bias.store(biases + intervalCount); in Make()
370 biases[i] = SK_PMColor4fTRANSPARENT; in Make()
376 scales[6], scales[7], biases[0], biases[1], biases[2], biases[3], biases[4], biases[5], in Make()
377 biases[6], biases[7], in Make()
|
| /external/tensorflow/tensorflow/contrib/cudnn_rnn/python/ops/ |
| D | cudnn_rnn_ops.py | 225 weights, biases = self._cu_canonical_to_tf_canonical(cu_weights, cu_biases)
226 return weights, biases
238 weights, biases = gen_cudnn_rnn_ops.cudnn_rnn_params_to_canonical(
247 return (weights, biases)
264 biases=cu_biases,
363 biases = tf_canonicals[len(tf_canonicals) // 2:]
367 layer_biases_num = len(biases) // self._num_layers
370 layer_biases = biases[i * layer_biases_num:(i + 1) * layer_biases_num]
394 def _cudnn_to_tf_biases(self, *biases): argument
534 def _cudnn_to_tf_biases(self, *biases): argument
[all …]
|
| /external/tensorflow/tensorflow/contrib/learn/python/learn/ops/ |
| D | losses_ops.py | 37 def mean_squared_error_regressor(tensor_in, labels, weights, biases, name=None): argument
41 predictions = nn.xw_plus_b(tensor_in, weights, biases)
52 biases, argument
77 logits = nn.xw_plus_b(tensor_in, weights, biases)
|
| D | ops_test.py | 40 biases = constant_op.constant([0.2, 0.3])
43 biases, class_weight)
|
| /external/tensorflow/tensorflow/contrib/learn/python/learn/estimators/ |
| D | nonlinear_test.py | 84 biases = ([regressor.get_variable_value("dnn/hiddenlayer_0/biases")] +
88 self.assertEqual(biases[0].shape, (10,))
89 self.assertEqual(biases[1].shape, (20,))
90 self.assertEqual(biases[2].shape, (10,))
91 self.assertEqual(biases[3].shape, (1,))
|
| /external/tensorflow/tensorflow/python/ops/ |
| D | nn_impl.py | 317 def relu_layer(x, weights, biases, name=None): argument
331 with ops.name_scope(name, "relu_layer", [x, weights, biases]) as name:
334 biases = ops.convert_to_tensor(biases, name="biases")
335 xw_plus_b = nn_ops.bias_add(math_ops.matmul(x, weights), biases)
1369 biases, argument
1434 weights + [biases, inputs, labels]):
1485 biases, all_ids, partition_strategy=partition_strategy)
1555 biases, argument
1645 biases,
1659 biases, argument
[all …]
|
| D | nn_test.py | 538 biases = np.random.randn(num_classes).astype(np.float32)
545 sampled_w, sampled_b = weights[sampled], biases[sampled]
546 true_w, true_b = weights[labels], biases[labels]
564 return weights, biases, hidden_acts, sampled_vals, exp_logits, exp_labels
566 def _ShardTestEmbeddings(self, weights, biases, num_shards): argument
586 initializer=constant_op.constant(biases))
599 (weights, biases, hidden_acts, sampled_vals, exp_logits,
610 biases=constant_op.constant(biases),
635 (weights, biases, hidden_acts, sampled_vals, exp_logits,
646 biases=constant_op.constant(biases),
[all …]
|
| /external/tensorflow/tensorflow/contrib/factorization/examples/ |
| D | mnist.py | 159 biases = tf.Variable(tf.zeros([hidden1_units]),
161 hidden1 = tf.nn.relu(tf.matmul(all_scores, weights) + biases)
168 biases = tf.Variable(tf.zeros([hidden2_units]),
170 hidden2 = tf.nn.relu(tf.matmul(hidden1, weights) + biases)
177 biases = tf.Variable(tf.zeros([NUM_CLASSES]),
179 logits = tf.matmul(hidden2, weights) + biases
|
| /external/tensorflow/tensorflow/contrib/distribute/python/ |
| D | step_fn_test.py | 57 weights, biases = [], []
61 biases.append(self.evaluate(layer.bias))
63 error = abs(numpy.add(numpy.squeeze(weights), numpy.squeeze(biases)) - 1)
|
| D | optimizer_v2_test.py | 58 weights, biases = [], []
63 biases.append(self.evaluate(layer.bias))
65 error = abs(numpy.add(numpy.squeeze(weights), numpy.squeeze(biases)) - 1)
|
| D | minimize_loss_test.py | 81 weights, biases = [], []
85 biases.append(self.evaluate(layer.bias))
87 error = abs(numpy.add(numpy.squeeze(weights), numpy.squeeze(biases)) - 1)
114 weights, biases = [], []
119 biases.append(self.evaluate(layer.bias))
121 error = abs(numpy.add(numpy.squeeze(weights), numpy.squeeze(biases)) - 1)
455 weights, biases, losses = [], [], []
460 biases.append(self.evaluate(layer.bias))
466 numpy.add(numpy.squeeze(weights), numpy.squeeze(biases)) - 1)
|
| /external/tensorflow/tensorflow/core/api_def/base_api/ |
| D | api_def_CudnnRNNCanonicalToParams.pbtxt | 10 biases.
18 biases: the canonical form of biases that can be used for saving
|
| D | api_def_CudnnRNNParamsToCanonical.pbtxt | 10 biases.
21 biases: the canonical form of biases that can be used for saving
|
| /external/tensorflow/tensorflow/examples/udacity/ |
| D | 2_fullyconnected.ipynb | 262 " # normal distribution. The biases get initialized to zero.\n",
265 " biases = tf.Variable(tf.zeros([num_labels]))\n",
268 " # We multiply the inputs with the weight matrix, and add biases. We compute\n",
272 " logits = tf.matmul(tf_train_dataset, weights) + biases\n",
285 " tf.matmul(tf_valid_dataset, weights) + biases)\n",
286 " test_prediction = tf.nn.softmax(tf.matmul(tf_test_dataset, weights) + biases)"
346 " # biases. \n",
446 " biases = tf.Variable(tf.zeros([num_labels]))\n",
449 " logits = tf.matmul(tf_train_dataset, weights) + biases\n",
459 " tf.matmul(tf_valid_dataset, weights) + biases)\n",
[all …]
|
| /external/tensorflow/tensorflow/contrib/training/python/training/ |
| D | training_test.py | 486 biases = variables_lib.get_variables_by_name('biases')
489 total_loss, optimizer, variables_to_train=biases)
530 weights, biases = variables_lib.get_variables()
536 total_loss, optimizer, variables_to_train=[biases])
543 weights_values, biases_values = session.run([weights, biases])
550 new_weights, new_biases = session.run([weights, biases])
561 new_weights, new_biases = session.run([weights, biases])
571 new_weights, new_biases = session.run([weights, biases])
|
| /external/tensorflow/tensorflow/core/profiler/g3doc/ |
| D | command_line.md | 229 pool_logit/biases (10, 10/10 params)
250 pool_logit/biases (10, 10/20 params)
251 pool_logit/biases/Momentum (10, 10/10 params)
289 entry.name = 'pool_logit/biases'
311 pool_logit/biases (10, 10/10 params)
335 pool_logit/biases (10, 10/20 params)
|