我们从Python开源项目中，提取了以下2个代码示例，用于说明如何使用tensorflow.contrib.slim.create_global_step()。
def __setup_training(self,images, labels): tf.logging.set_verbosity(tf.logging.INFO) logits, end_points = self.network_fn(images) ############################# # Specify the loss function # ############################# loss_1 = None if 'AuxLogits' in end_points: loss_1 = tf.losses.softmax_cross_entropy( logits=end_points['AuxLogits'], onehot_labels=labels, label_smoothing=self.label_smoothing, weights=0.4, scope='aux_loss') total_loss = tf.losses.softmax_cross_entropy( logits=logits, onehot_labels=labels, label_smoothing=self.label_smoothing, weights=1.0) if loss_1 is not None: total_loss = total_loss + loss_1 global_step = slim.create_global_step() # Variables to train. variables_to_train = self.__get_variables_to_train() learning_rate = self.__configure_learning_rate(self.dataset.num_samples, global_step) optimizer = self.__configure_optimizer(learning_rate) train_op = slim.learning.create_train_op(total_loss, optimizer, variables_to_train=variables_to_train) self.__add_summaries(end_points, learning_rate, total_loss) ########################### # Kicks off the training. # ########################### slim.learning.train( train_op, logdir=self.train_dir, init_fn=self.__get_init_fn(), number_of_steps=self.max_number_of_steps, log_every_n_steps=self.log_every_n_steps, save_summaries_secs=self.save_summaries_secs, save_interval_secs=self.save_interval_secs) return
def __start_training(self): tf.logging.set_verbosity(tf.logging.INFO) #get batched training training data image, filename,glabels,gbboxes,gdifficults,gclasses, localizations, gscores = self.get_voc_2007_2012_train_data() #get model outputs predictions, localisations, logits, end_points = g_ssd_model.get_model(image, weight_decay=self.weight_decay, is_training=True) #get model training losss total_loss = g_ssd_model.get_losses(logits, localisations, gclasses, localizations, gscores) global_step = slim.create_global_step() # Variables to train. variables_to_train = self.__get_variables_to_train() learning_rate = self.__configure_learning_rate(self.dataset.num_samples, global_step) optimizer = self.__configure_optimizer(learning_rate) train_op = slim.learning.create_train_op(total_loss, optimizer, variables_to_train=variables_to_train) self.__add_summaries(end_points, learning_rate, total_loss) self.setup_debugging(predictions, localizations, glabels, gbboxes, gdifficults) gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0.9) config = tf.ConfigProto(log_device_placement=False, gpu_options=gpu_options) ########################### # Kicks off the training. # ########################### slim.learning.train( train_op, self.train_dir, train_step_fn=self.train_step, saver=tf_saver.Saver(max_to_keep=500), init_fn=self.__get_init_fn(), number_of_steps=self.max_number_of_steps, log_every_n_steps=self.log_every_n_steps, save_summaries_secs=self.save_summaries_secs, # session_config=config, save_interval_secs=self.save_interval_secs) return
