How to use the babi.model.Tower function in babi

To help you get started, we’ve selected a few babi examples, based on popular ways it is used in public projects.

Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.

github uwnlp / qrn / babi / main.py View on Github external
def get_best_trial_idx_with_acc(_val_accs):
        return max(enumerate(_val_accs), key=lambda x: x[1])[0]



    val_losses = []
    val_accs = []
    test_accs = []
    for trial_idx in range(1, num_trials+1):
        if config.train:
            print("-" * 80)
            print("Task {} trial {}".format(config.task, trial_idx))
        mkdirs(config, trial_idx)
        graph = tf.Graph()
        # TODO : initialize BaseTower-subclassed objects
        towers = [Tower(config) for _ in range(config.num_devices)]
        sess = tf.Session(graph=graph, config=tf.ConfigProto(allow_soft_placement=True))
        # TODO : initialize BaseRunner-subclassed object
        runner = Runner(config, sess, towers)
        with graph.as_default(), tf.device("/cpu:0"):
            runner.initialize()
            if config.train:
                if config.load:
                    runner.load()
                val_loss, val_acc = runner.train(comb_train_ds, config.num_epochs, val_data_set=comb_dev_ds,
                                                 num_batches=config.train_num_batches,
                                                 val_num_batches=config.val_num_batches, eval_ph_names=eval_ph_names)
                val_losses.append(val_loss)
                val_accs.append(val_acc)
            else:
                runner.load()
            test_loss, test_acc = runner.eval(comb_test_ds, eval_tensor_names=eval_tensor_names,