| /external/tensorflow/tensorflow/contrib/losses/python/losses/ |
| loss_ops_test.py | 51 self._predictions, self._predictions, weights=None)
64 weights = 2.3
66 weights)
68 self.assertAlmostEqual(5.5 * weights, loss.eval(), 3)
71 weights = 2.3
73 constant_op.constant(weights))
75 self.assertAlmostEqual(5.5 * weights, loss.eval(), 3)
78 weights = constant_op.constant([1.2, 0.0], shape=[2,])
80 weights)
85 weights = constant_op.constant([1.2, 0.0], shape=[2, 1]
[all...] |
| /external/tensorflow/tensorflow/python/ops/ |
| metrics_impl.py | 53 def _remove_squeezable_dimensions(predictions, labels, weights):
58 Squeezes or expands last dim of `weights` if its rank differs by 1 from the
61 If `weights` is scalar, it is kept scalar.
69 weights: Optional weight scalar or `Tensor` whose dimensions match
73 Tuple of `predictions`, `labels` and `weights`. Each of them possibly has
74 the last dimension squeezed, `weights` could be extended by one dimension.
82 if weights is None:
85 weights = ops.convert_to_tensor(weights)
86 weights_shape = weights.get_shape(
[all...] |
| /external/tensorflow/tensorflow/contrib/metrics/python/ops/ |
| metric_ops.py | 68 weights=None,
72 """Sum the weights of true_positives.
74 If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
81 weights: Optional `Tensor` whose rank is either 0, or the same rank as
97 `weights` is not `None` and its shape doesn't match `predictions`, or if
104 weights=weights,
112 weights=None
[all...] |
| /external/apache-commons-math/src/main/java/org/apache/commons/math/stat/descriptive/summary/ |
| Sum.java | 138 * <li>the weights array is null</li>
139 * <li>the weights array does not have the same length as the values array</li>
140 * <li>the weights array contains one or more infinite values</li>
141 * <li>the weights array contains one or more NaN values</li>
142 * <li>the weights array contains negative values</li>
147 * weighted sum = Σ(values[i] * weights[i])
151 * @param weights the weights array
158 public double evaluate(final double[] values, final double[] weights,
161 if (test(values, weights, begin, length))
[all...] |
| /external/tensorflow/tensorflow/core/lib/random/ |
| weighted_picker_test.cc | 34 static void TestPickAt(int items, const int32* weights);
64 VLOG(0) << "======= Grown picker with zero weights";
73 VLOG(0) << "======= Shrink picker and check weights";
95 VLOG(0) << "======= Check uniform with big weights";
103 static const int32 weights[] = {1, 0, 200, 5, 42}; local
104 TestPickAt(TF_ARRAYSIZE(weights), weights); local
131 // Create zero weights array
132 std::vector<int32> weights(size);
134 weights[elem] = 0
[all...] |
| /external/tensorflow/tensorflow/contrib/boosted_trees/lib/utils/ |
| dropout_utils.h | 33 // indices and the weights they had when this method ran.
39 // weights: weights of those trees
41 // weights.
45 const std::vector<float>& weights, std::vector<int32>* dropped_trees,
48 // Recalculates the weights of the trees when the new trees are added to
56 // ensemble. Returns current_weights: updated vector of the tree weights.
57 // Weights of dropped trees are updated. Note that the size of returned vector
59 // weights of the new trees to be added) if new_trees_first_index
66 // Current weights and num_updates will be updated as a result of thi
[all...] |
| /external/tensorflow/tensorflow/python/keras/_impl/keras/applications/ |
| nasnet_test.py | 28 model = keras.applications.NASNetMobile(weights=None)
32 model = keras.applications.NASNetMobile(weights=None, include_top=False)
36 model = keras.applications.NASNetMobile(weights=None,
43 keras.applications.NASNetMobile(weights='unknown',
46 keras.applications.NASNetMobile(weights='imagenet',
53 model = keras.applications.NASNetLarge(weights=None)
57 model = keras.applications.NASNetLarge(weights=None, include_top=False)
61 model = keras.applications.NASNetLarge(weights=None,
68 keras.applications.NASNetLarge(weights='unknown',
71 keras.applications.NASNetLarge(weights='imagenet'
[all...] |
| /external/tensorflow/tensorflow/python/keras/_impl/keras/engine/ |
| topology.py | 77 trainable: Boolean, whether the layer weights
99 weights: The concatenation of the lists trainable_weights and
112 set_weights(weights)
145 'weights',
187 if 'weights' in kwargs:
188 self._initial_weights = kwargs['weights']
413 def set_weights(self, weights):
414 """Sets the weights of the layer, from Numpy arrays.
417 weights: a list of Numpy arrays. The number
419 number of the dimensions of the weights
[all...] |
| /external/apache-commons-math/src/main/java/org/apache/commons/math/optimization/ |
| LeastSquaresConverter.java | 49 * This class support combination of residuals with or without weights and correlations.
66 /** Optional weights for the residuals. */
67 private final double[] weights; field in class:LeastSquaresConverter
80 this.weights = null;
84 /** Build a simple converter for uncorrelated residuals with the specific weights.
92 * Weights can be used for example to combine residuals with different standard
96 * In this case, the weights array should be initialized with value
102 * weights array must have consistent sizes or a {@link FunctionEvaluationException} will be
107 * @param weights weights to apply to the residual
[all...] |
| /external/tensorflow/tensorflow/contrib/training/python/training/ |
| resample_test.py | 54 """Tests `resample(x, weights)` and resample(resample(x, rate), 1/rate)`."""
58 weights = self.get_weights(count)
61 [foo, bar], constant_op.constant(weights), rate, seed=123)
86 self.assert_expected(weights, rate, counts_resampled, n)
90 [1.0 for _ in weights],
104 weights = self.get_weights(count)
107 constant_op.constant(weights),
131 # sum(vals) * n ~= weighted_sum(resampled, 1.0/weights)
159 """Test that we can define the ops with float64 weights."""
162 weights = math_ops.cast(self.get_weights(count), dtypes.float64
[all...] |
| /external/apache-commons-math/src/main/java/org/apache/commons/math/ode/nonstiff/ |
| DormandPrince853StepInterpolator.java | 44 /** Propagation weights, element 1. */
49 /** Propagation weights, element 6. */
52 /** Propagation weights, element 7. */
55 /** Propagation weights, element 8. */
58 /** Propagation weights, element 9. */
61 /** Propagation weights, element 10. */
64 /** Propagation weights, element 11. */
67 /** Propagation weights, element 12. */
73 /** Internal weights for stage 14, element 1. */
78 /** Internal weights for stage 14, element 6. *
[all...] |
| /external/tensorflow/tensorflow/contrib/kernel_methods/python/ |
| losses_test.py | 50 """An error is raised when weights have invalid shape."""
54 weights = constant_op.constant([1.5, 0.2], shape=(2, 1, 1))
56 _ = losses.sparse_multiclass_hinge_loss(labels, logits, weights)
67 """An error is raised when weights are None."""
72 _ = losses.sparse_multiclass_hinge_loss(labels, logits, weights=None)
75 """Error raised when weights and labels have same ranks, different sizes."""
79 weights = constant_op.constant([1.1, 2.0], shape=(2, 1))
81 _ = losses.sparse_multiclass_hinge_loss(labels, logits, weights)
84 """Error raised when weights and labels have different ranks and sizes."""
88 weights = constant_op.constant([1.1, 2.0, 2.8], shape=(3,)
[all...] |
| losses.py | 33 weights=1.0,
56 weights: Optional (python) scalar or `Tensor`. If a non-scalar `Tensor`, its
67 ValueError: If `logits`, `labels` or `weights` have invalid or inconsistent
91 # Check labels and weights have valid ranks and are consistent.
102 weights = ops.convert_to_tensor(weights)
103 weights_rank = weights.get_shape().ndims
106 'non-scalar weights should have rank 1 ([batch_size]) or 2 '
110 weights = array_ops.reshape(weights, shape=[-1]
[all...] |
| /external/tensorflow/tensorflow/contrib/rnn/python/tools/ |
| checkpoint_convert.py | 62 ('basic_rnn_cell/weights', 'basic_rnn_cell/kernel'),
65 ('gru_cell/weights', 'gru_cell/kernel'),
67 ('gru_cell/gates/weights', 'gru_cell/gates/kernel'),
69 ('gru_cell/candidate/weights', 'gru_cell/candidate/kernel'),
72 ('basic_lstm_cell/weights', 'basic_lstm_cell/kernel'),
75 ('lstm_cell/weights', 'lstm_cell/kernel'),
77 ('lstm_cell/projection/weights', 'lstm_cell/projection/kernel'),
80 ('output_projection_wrapper/weights', 'output_projection_wrapper/kernel'),
83 ('input_projection_wrapper/weights', 'input_projection_wrapper/kernel'),
88 ('lstm_block_wrapper/weights', 'lstm_block_wrapper/kernel')
[all...] |
| /external/tensorflow/tensorflow/contrib/seq2seq/python/ops/ |
| loss.py | 32 weights,
53 weights: A Tensor of shape `[batch_size, sequence_length]` and dtype
54 float. `weights` constitutes the weighting of each prediction in the
55 sequence. When using `weights` as masking, set all valid timesteps to 1
75 dimensions or weights does not have 2 dimensions.
83 if len(weights.get_shape()) != 2:
84 raise ValueError("Weights must be a [batch_size x sequence_length] "
86 with ops.name_scope(name, "sequence_loss", [logits, targets, weights]):
95 crossent *= array_ops.reshape(weights, [-1])
98 total_size = math_ops.reduce_sum(weights)
[all...] |
| /external/apache-commons-math/src/main/java/org/apache/commons/math/stat/descriptive/ |
| AbstractUnivariateStatistic.java | 165 * and the weights are all non-negative, non-NaN, finite, and not all zero.
169 * positive length and the weights array contains legitimate values.</li>
172 * <li>the weights array is null</li>
173 * <li>the weights array does not have the same length as the values array</li>
174 * <li>the weights array contains one or more infinite values</li>
175 * <li>the weights array contains one or more NaN values</li>
176 * <li>the weights array contains negative values</li>
184 * @param weights the weights array
193 final double[] weights,
[all...] |
| /external/tensorflow/tensorflow/contrib/model_pruning/python/ |
| pruning_test.py | 90 weights = variables.Variable(
91 random_ops.random_normal([width, height], stddev=1), name="weights")
92 masked_weights = pruning.apply_mask(weights,
95 weights_val = weights.eval()
101 weights = variables.Variable(
102 math_ops.linspace(1.0, 100.0, 100), name="weights")
103 masked_weights = pruning.apply_mask(weights)
115 def _blockMasking(self, hparams, weights, expected_mask):
126 _, new_mask = p._maybe_update_block_mask(weights, threshold)
127 # Check if the mask is the same size as the weights
[all...] |
| /external/tensorflow/tensorflow/contrib/learn/python/learn/estimators/ |
| nonlinear_test.py | 49 classifier.get_variable_value("dnn/hiddenlayer_0/weights").shape,
52 classifier.get_variable_value("dnn/hiddenlayer_1/weights").shape,
55 classifier.get_variable_value("dnn/hiddenlayer_2/weights").shape,
58 classifier.get_variable_value("dnn/logits/weights").shape, (10, 3))
75 weights = ([regressor.get_variable_value("dnn/hiddenlayer_0/weights")] +
76 [regressor.get_variable_value("dnn/hiddenlayer_1/weights")] +
77 [regressor.get_variable_value("dnn/hiddenlayer_2/weights")] +
78 [regressor.get_variable_value("dnn/logits/weights")])
79 self.assertEqual(weights[0].shape, (13, 10)
[all...] |
| /external/apache-commons-math/src/main/java/org/apache/commons/math/stat/descriptive/moment/ |
| Mean.java | 180 * described above is used here, with weights applied in computing both the original
185 * <li>the weights array is null</li>
186 * <li>the weights array does not have the same length as the values array</li>
187 * <li>the weights array contains one or more infinite values</li>
188 * <li>the weights array contains one or more NaN values</li>
189 * <li>the weights array contains negative values</li>
194 * @param weights the weights array
201 public double evaluate(final double[] values, final double[] weights,
203 if (test(values, weights, begin, length))
[all...] |
| /external/tensorflow/tensorflow/python/estimator/canned/ |
| head.py | 60 # * a scalar `Tensor` representing the example weights
63 'LossSpec', ['training_loss', 'unreduced_loss', 'weights',
167 * the scalar `Tensor` representing the example weights
276 """Fetches weights from features and checks that the shape matches logits.
278 Consider logits of shape [D0, D1, ... DN, logits_dimension]. Weights shape
282 * [D0, D1, ... DN]: In this case, weights is reshaped into
286 features: The features dict that contains weights.
289 allow_per_logit_weights: Boolean. Whether we allow weights along the logits
292 Validated and reshaped weights Tensor.
294 ValueError: If the weights `Tensor` cannot be cast into float
[all...] |
| /external/opencv/ml/src/ |
| mlcnn.cpp | 297 // 3) Update weights by the gradient descent
635 CvMat* connect_mask, CvMat* weights )
656 CV_CALL(layer->weights = cvCreateMat( n_output_planes, K*K+1, CV_32FC1 ));
659 if( weights )
661 if( !ICV_IS_MAT_OF_TYPE( weights, CV_32FC1 ) )
662 CV_ERROR( CV_StsBadSize, "Type of initial weights matrix must be CV_32FC1" );
663 if( !CV_ARE_SIZES_EQ( weights, layer->weights ) )
664 CV_ERROR( CV_StsBadSize, "Invalid size of initial weights matrix" );
665 CV_CALL(cvCopy( weights, layer->weights ))
[all...] |
| /external/tensorflow/tensorflow/tools/api/golden/ |
| tensorflow.keras.applications.inception_resnet_v2.pbtxt | 5 argspec: "args=[\'include_top\', \'weights\', \'input_tensor\', \'input_shape\', \'pooling\', \'classes\'], varargs=None, keywords=None, defaults=[\'True\', \'imagenet\', \'None\', \'None\', \'None\', \'1000\'], "
|
| tensorflow.keras.applications.inception_v3.pbtxt | 5 argspec: "args=[\'include_top\', \'weights\', \'input_tensor\', \'input_shape\', \'pooling\', \'classes\'], varargs=None, keywords=None, defaults=[\'True\', \'imagenet\', \'None\', \'None\', \'None\', \'1000\'], "
|
| tensorflow.keras.applications.mobilenet.pbtxt | 5 argspec: "args=[\'input_shape\', \'alpha\', \'depth_multiplier\', \'dropout\', \'include_top\', \'weights\', \'input_tensor\', \'pooling\', \'classes\'], varargs=None, keywords=None, defaults=[\'None\', \'1.0\', \'1\', \'0.001\', \'True\', \'imagenet\', \'None\', \'None\', \'1000\'], "
|
| tensorflow.keras.applications.resnet50.pbtxt | 5 argspec: "args=[\'include_top\', \'weights\', \'input_tensor\', \'input_shape\', \'pooling\', \'classes\'], varargs=None, keywords=None, defaults=[\'True\', \'imagenet\', \'None\', \'None\', \'None\', \'1000\'], "
|
