Python tensorflow.no_regularizer() Examples
The following are 7
code examples of tensorflow.no_regularizer().
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
tensorflow
, or try the search function
.
![](https://www.programcreek.com/common/static/images/search.png)
Example #1
Source File: base_memory.py From iter-reason with MIT License | 6 votes |
def _mem_init(self, is_training, name): mem_initializer = tf.constant_initializer(0.0) # Kinda like bias if cfg.TRAIN.BIAS_DECAY: mem_regularizer = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY) else: mem_regularizer = tf.no_regularizer with tf.variable_scope('SMN'): with tf.variable_scope(name): mem_init = tf.get_variable('mem_init', [1, cfg.MEM.INIT_H, cfg.MEM.INIT_W, cfg.MEM.C], initializer=mem_initializer, trainable=is_training, regularizer=mem_regularizer) self._score_summaries[0].append(mem_init) # resize it to the image-specific size mem_init = tf.image.resize_bilinear(mem_init, self._memory_size, name="resize_init") return mem_init
Example #2
Source File: vgg.py From mtl-ssl with Apache License 2.0 | 6 votes |
def vgg_arg_scope(weight_decay=0.0005): """Defines the VGG arg scope. Args: weight_decay: The l2 regularization coefficient. Returns: An arg_scope. """ with slim.arg_scope([slim.conv2d, slim.fully_connected], activation_fn=tf.nn.relu, weights_regularizer=slim.l2_regularizer(weight_decay), biases_regularizer=tf.no_regularizer, biases_initializer=tf.zeros_initializer()): with slim.arg_scope([slim.conv2d], padding='SAME') as arg_sc: return arg_sc
Example #3
Source File: variable_scope_test.py From deep_image_model with Apache License 2.0 | 5 votes |
def testVarScopeRegularizer(self): with self.test_session() as sess: init = tf.constant_initializer(0.3) def regularizer1(v): return tf.reduce_mean(v) + 0.1 def regularizer2(v): return tf.reduce_mean(v) + 0.2 with tf.variable_scope("tower", regularizer=regularizer1) as tower: with tf.variable_scope("foo", initializer=init): v = tf.get_variable("v", []) sess.run(tf.initialize_variables([v])) losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES) self.assertEqual(1, len(losses)) self.assertAllClose(losses[0].eval(), 0.4) with tf.variable_scope(tower, initializer=init) as vs: u = tf.get_variable("u", []) vs.set_regularizer(regularizer2) w = tf.get_variable("w", []) # Next 3 variable not regularized to test disabling regularization. x = tf.get_variable("x", [], regularizer=tf.no_regularizer) with tf.variable_scope("baz", regularizer=tf.no_regularizer): y = tf.get_variable("y", []) vs.set_regularizer(tf.no_regularizer) z = tf.get_variable("z", []) # Check results. losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES) self.assertEqual(3, len(losses)) sess.run(tf.initialize_variables([u, w, x, y, z])) self.assertAllClose(losses[0].eval(), 0.4) self.assertAllClose(losses[1].eval(), 0.4) self.assertAllClose(losses[2].eval(), 0.5) with tf.variable_scope("foo", reuse=True): v = tf.get_variable("v", []) # "v" is alredy there, reused losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES) self.assertEqual(3, len(losses)) # No new loss added.
Example #4
Source File: network.py From MSDS-RCNN with MIT License | 4 votes |
def create_architecture_demo(self, mode, num_classes, tag=None, anchor_scales=(8, 16, 32), anchor_ratios=(0.5, 1, 2)): assert mode == 'TEST', 'only for demo' self._image = tf.placeholder(tf.float32, shape=[1, None, None, 3]) self._lwir = tf.placeholder(tf.float32, shape=[1, None, None, 3]) self._im_info = tf.placeholder(tf.float32, shape=[3]) self._tag = tag self._num_classes = num_classes self._mode = mode self._anchor_scales = anchor_scales self._num_scales = len(anchor_scales) self._anchor_ratios = anchor_ratios self._num_ratios = len(anchor_ratios) self._num_anchors = self._num_scales * self._num_ratios training = mode == 'TRAIN' testing = mode == 'TEST' assert tag != None # handle most of the regularizers here weights_regularizer = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY) if cfg.TRAIN.BIAS_DECAY: biases_regularizer = weights_regularizer else: biases_regularizer = tf.no_regularizer # select initializers if cfg.TRAIN.TRUNCATED: initializer = tf.truncated_normal_initializer(mean=0.0, stddev=0.01) initializer_bbox = tf.truncated_normal_initializer(mean=0.0, stddev=0.001) else: initializer = tf.random_normal_initializer(mean=0.0, stddev=0.01) initializer_bbox = tf.random_normal_initializer(mean=0.0, stddev=0.001) # list as many types of layers as possible, even if they are not used now with arg_scope([slim.conv2d, slim.conv2d_in_plane, slim.conv2d_transpose, slim.separable_conv2d, slim.fully_connected], weights_regularizer=weights_regularizer, biases_regularizer=biases_regularizer, biases_initializer=tf.constant_initializer(0.0)): self._build_network(training, initializer, initializer_bbox) layers_to_output = {} return layers_to_output
Example #5
Source File: network.py From iter-reason with MIT License | 4 votes |
def create_architecture(self, mode, num_classes, tag=None): self._image = tf.placeholder(tf.float32, shape=[1, None, None, 3]) self._im_info = tf.placeholder(tf.float32, shape=[3]) self._memory_size = tf.placeholder(tf.int32, shape=[2]) self._gt_boxes = tf.placeholder(tf.float32, shape=[None, 5]) self._num_gt = tf.placeholder(tf.int32, shape=[]) self._tag = tag self._num_classes = num_classes self._mode = mode training = mode == 'TRAIN' testing = mode == 'TEST' assert tag is not None # handle most of the regularizers here weights_regularizer = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY) if cfg.TRAIN.BIAS_DECAY: biases_regularizer = weights_regularizer else: biases_regularizer = tf.no_regularizer # list as many types of layers as possible, even if they are not used now with slim.arg_scope([slim.conv2d, slim.conv2d_in_plane, \ slim.conv2d_transpose, slim.separable_conv2d, slim.fully_connected], weights_regularizer=weights_regularizer, biases_regularizer=biases_regularizer, biases_initializer=tf.constant_initializer(0.0)): rois, cls_prob = self._build_network(training) layers_to_output = {'rois': rois} if not testing: self._add_losses() layers_to_output.update(self._losses) val_summaries = [] with tf.device("/cpu:0"): val_summaries.append(self._add_gt_image_summary()) val_summaries.append(self._add_pred_summary()) for key, var in self._event_summaries.items(): val_summaries.append(tf.summary.scalar(key, var)) for key, var in self._score_summaries.items(): self._add_score_summary(key, var) for var in self._act_summaries: self._add_act_summary(var) self._summary_op = tf.summary.merge_all() self._summary_op_val = tf.summary.merge(val_summaries) layers_to_output.update(self._predictions) return layers_to_output
Example #6
Source File: base_memory.py From iter-reason with MIT License | 4 votes |
def create_architecture(self, mode, num_classes, tag=None): self._image = tf.placeholder(tf.float32, shape=[1, None, None, 3]) self._im_info = tf.placeholder(tf.float32, shape=[3]) self._memory_size = tf.placeholder(tf.int32, shape=[2]) self._gt_boxes = tf.placeholder(tf.float32, shape=[None, 5]) self._count_base = tf.ones([1, cfg.MEM.CROP_SIZE, cfg.MEM.CROP_SIZE, 1]) self._num_gt = tf.placeholder(tf.int32, shape=[]) self._tag = tag self._num_classes = num_classes self._mode = mode training = mode == 'TRAIN' testing = mode == 'TEST' assert tag is not None # handle most of the regularizers here weights_regularizer = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY) if cfg.TRAIN.BIAS_DECAY: biases_regularizer = weights_regularizer else: biases_regularizer = tf.no_regularizer # list as many types of layers as possible, even if they are not used now with slim.arg_scope([slim.conv2d, slim.conv2d_in_plane, \ slim.conv2d_transpose, slim.separable_conv2d, slim.fully_connected], weights_regularizer=weights_regularizer, biases_regularizer=biases_regularizer, biases_initializer=tf.constant_initializer(0.0)): rois = self._build_memory(training, testing) layers_to_output = {'rois': rois} if not testing: self._add_memory_losses("loss") layers_to_output.update(self._losses) self._create_summary() layers_to_output.update(self._predictions) return layers_to_output # take the last predicted output
Example #7
Source File: base.py From PoseFix_RELEASE with MIT License | 4 votes |
def _make_graph(self): self.logger.info("Generating training graph on {} GPUs ...".format(self.cfg.num_gpus)) weights_initializer = slim.xavier_initializer() biases_initializer = tf.constant_initializer(0.) biases_regularizer = tf.no_regularizer weights_regularizer = tf.contrib.layers.l2_regularizer(self.cfg.weight_decay) tower_grads = [] with tf.variable_scope(tf.get_variable_scope()): for i in range(self.cfg.num_gpus): with tf.device('/gpu:%d' % i): with tf.name_scope('tower_%d' % i) as name_scope: # Force all Variables to reside on the CPU. with slim.arg_scope([slim.model_variable, slim.variable], device='/device:CPU:0'): with slim.arg_scope([slim.conv2d, slim.conv2d_in_plane, \ slim.conv2d_transpose, slim.separable_conv2d, slim.fully_connected], weights_regularizer=weights_regularizer, biases_regularizer=biases_regularizer, weights_initializer=weights_initializer, biases_initializer=biases_initializer): # loss over single GPU self.net.make_network(is_train=True) if i == self.cfg.num_gpus - 1: loss = self.net.get_loss(include_wd=True) else: loss = self.net.get_loss() self._input_list.append( self.net.get_inputs() ) tf.get_variable_scope().reuse_variables() if i == 0: if self.cfg.num_gpus > 1 and self.cfg.bn_train is True: self.logger.warning("BN is calculated only on single GPU.") extra_update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS, name_scope) with tf.control_dependencies(extra_update_ops): grads = self._optimizer.compute_gradients(loss) else: grads = self._optimizer.compute_gradients(loss) final_grads = [] with tf.variable_scope('Gradient_Mult') as scope: for grad, var in grads: final_grads.append((grad, var)) tower_grads.append(final_grads) if len(tower_grads) > 1: grads = average_gradients(tower_grads) else: grads = tower_grads[0] apply_gradient_op = self._optimizer.apply_gradients(grads) train_op = tf.group(apply_gradient_op, *extra_update_ops) return train_op