Python model.NamignizerModel() Examples

The following are 30 code examples of model.NamignizerModel(). 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 model , or try the search function .
Example #1
Source File: names.py    From HumanRecognition with MIT License 5 votes vote down vote up 
def namignator(checkpoint_path, config):
    """Generates names randomly according to a given model

    Args:
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # mutate the config to become a name generator config
    config.num_steps = 1
    config.batch_size = 1

    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                  {m.input_data: np.zeros((1, 1)),
                                                   m.targets: np.zeros((1, 1)),
                                                   m.weights: np.ones(1)})

        # sample from our softmax activations
        next_letter = np.random.choice(27, p=activations[0])
        name = [next_letter]
        while next_letter != 0:
            activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                      {m.input_data: [[next_letter]],
                                                       m.targets: np.zeros((1, 1)),
                                                       m.initial_state: final_state,
                                                       m.weights: np.ones(1)})

            next_letter = np.random.choice(27, p=activations[0])
            name += [next_letter]

        print(map(lambda x: chr(x + 96), name))
Example #2
Source File: names.py    From DOTA_models with Apache License 2.0 5 votes vote down vote up 
def run_epoch(session, m, names, counts, epoch_size, eval_op, verbose=False):
    """Runs the model on the given data for one epoch

    Args:
        session: the tf session holding the model graph
        m: an instance of the NamignizerModel
        names: a set of lowercase names of 26 characters
        counts: a list of the frequency of the above names
        epoch_size: the number of batches to run
        eval_op: whether to change the params or not, and how to do it
    Kwargs:
        verbose: whether to print out state of training during the epoch
    Returns:
        cost: the average cost during the last stage of the epoch
    """
    start_time = time.time()
    costs = 0.0
    iters = 0
    for step, (x, y) in enumerate(data_utils.namignizer_iterator(names, counts,
                                                                 m.batch_size, m.num_steps, epoch_size)):

        cost, _ = session.run([m.cost, eval_op],
                              {m.input_data: x,
                               m.targets: y,
                               m.weights: np.ones(m.batch_size * m.num_steps)})
        costs += cost
        iters += m.num_steps

        if verbose and step % (epoch_size // 10) == 9:
            print("%.3f perplexity: %.3f speed: %.0f lps" %
                  (step * 1.0 / epoch_size, np.exp(costs / iters),
                   iters * m.batch_size / (time.time() - start_time)))

        if step >= epoch_size:
            break

    return np.exp(costs / iters)
Example #3
Source File: names.py    From object_detection_kitti with Apache License 2.0 5 votes vote down vote up 
def namignator(checkpoint_path, config):
    """Generates names randomly according to a given model

    Args:
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # mutate the config to become a name generator config
    config.num_steps = 1
    config.batch_size = 1

    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                  {m.input_data: np.zeros((1, 1)),
                                                   m.targets: np.zeros((1, 1)),
                                                   m.weights: np.ones(1)})

        # sample from our softmax activations
        next_letter = np.random.choice(27, p=activations[0])
        name = [next_letter]
        while next_letter != 0:
            activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                      {m.input_data: [[next_letter]],
                                                       m.targets: np.zeros((1, 1)),
                                                       m.initial_state: final_state,
                                                       m.weights: np.ones(1)})

            next_letter = np.random.choice(27, p=activations[0])
            name += [next_letter]

        print(map(lambda x: chr(x + 96), name))
Example #4
Source File: names.py    From object_detection_with_tensorflow with MIT License 5 votes vote down vote up 
def run_epoch(session, m, names, counts, epoch_size, eval_op, verbose=False):
    """Runs the model on the given data for one epoch

    Args:
        session: the tf session holding the model graph
        m: an instance of the NamignizerModel
        names: a set of lowercase names of 26 characters
        counts: a list of the frequency of the above names
        epoch_size: the number of batches to run
        eval_op: whether to change the params or not, and how to do it
    Kwargs:
        verbose: whether to print out state of training during the epoch
    Returns:
        cost: the average cost during the last stage of the epoch
    """
    start_time = time.time()
    costs = 0.0
    iters = 0
    for step, (x, y) in enumerate(data_utils.namignizer_iterator(names, counts,
                                                                 m.batch_size, m.num_steps, epoch_size)):

        cost, _ = session.run([m.cost, eval_op],
                              {m.input_data: x,
                               m.targets: y,
                               m.weights: np.ones(m.batch_size * m.num_steps)})
        costs += cost
        iters += m.num_steps

        if verbose and step % (epoch_size // 10) == 9:
            print("%.3f perplexity: %.3f speed: %.0f lps" %
                  (step * 1.0 / epoch_size, np.exp(costs / iters),
                   iters * m.batch_size / (time.time() - start_time)))

        if step >= epoch_size:
            break

    return np.exp(costs / iters)
Example #5
Source File: names.py    From object_detection_with_tensorflow with MIT License 5 votes vote down vote up 
def train(data_dir, checkpoint_path, config):
    """Trains the model with the given data

    Args:
        data_dir: path to the data for the model (see data_utils for data
            format)
        checkpoint_path: the path to save the trained model checkpoints
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # Prepare Name data.
    print("Reading Name data in %s" % data_dir)
    names, counts = data_utils.read_names(data_dir)

    with tf.Graph().as_default(), tf.Session() as session:
        initializer = tf.random_uniform_initializer(-config.init_scale,
                                                    config.init_scale)
        with tf.variable_scope("model", reuse=None, initializer=initializer):
            m = NamignizerModel(is_training=True, config=config)

        tf.global_variables_initializer().run()

        for i in range(config.max_max_epoch):
            lr_decay = config.lr_decay ** max(i - config.max_epoch, 0.0)
            m.assign_lr(session, config.learning_rate * lr_decay)

            print("Epoch: %d Learning rate: %.3f" % (i + 1, session.run(m.lr)))
            train_perplexity = run_epoch(session, m, names, counts, config.epoch_size, m.train_op,
                                         verbose=True)
            print("Epoch: %d Train Perplexity: %.3f" %
                  (i + 1, train_perplexity))

            m.saver.save(session, checkpoint_path, global_step=i)
Example #6
Source File: names.py    From object_detection_with_tensorflow with MIT License 5 votes vote down vote up 
def namignize(names, checkpoint_path, config):
    """Recognizes names and prints the Perplexity of the model for each names
    in the list

    Args:
        names: a list of names in the model format
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        for name in names:
            x, y = data_utils.name_to_batch(name, m.batch_size, m.num_steps)

            cost, loss, _ = session.run([m.cost, m.loss, tf.no_op()],
                                  {m.input_data: x,
                                   m.targets: y,
                                   m.weights: np.concatenate((
                                       np.ones(len(name)), np.zeros(m.batch_size * m.num_steps - len(name))))})

            print("Name {} gives us a perplexity of {}".format(
                name, np.exp(cost)))
Example #7
Source File: names.py    From object_detection_with_tensorflow with MIT License 5 votes vote down vote up 
def namignator(checkpoint_path, config):
    """Generates names randomly according to a given model

    Args:
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # mutate the config to become a name generator config
    config.num_steps = 1
    config.batch_size = 1

    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                  {m.input_data: np.zeros((1, 1)),
                                                   m.targets: np.zeros((1, 1)),
                                                   m.weights: np.ones(1)})

        # sample from our softmax activations
        next_letter = np.random.choice(27, p=activations[0])
        name = [next_letter]
        while next_letter != 0:
            activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                      {m.input_data: [[next_letter]],
                                                       m.targets: np.zeros((1, 1)),
                                                       m.initial_state: final_state,
                                                       m.weights: np.ones(1)})

            next_letter = np.random.choice(27, p=activations[0])
            name += [next_letter]

        print(map(lambda x: chr(x + 96), name))
Example #8
Source File: names.py    From AI_Reader with Apache License 2.0 5 votes vote down vote up 
def run_epoch(session, m, names, counts, epoch_size, eval_op, verbose=False):
    """Runs the model on the given data for one epoch

    Args:
        session: the tf session holding the model graph
        m: an instance of the NamignizerModel
        names: a set of lowercase names of 26 characters
        counts: a list of the frequency of the above names
        epoch_size: the number of batches to run
        eval_op: whether to change the params or not, and how to do it
    Kwargs:
        verbose: whether to print out state of training during the epoch
    Returns:
        cost: the average cost during the last stage of the epoch
    """
    start_time = time.time()
    costs = 0.0
    iters = 0
    for step, (x, y) in enumerate(data_utils.namignizer_iterator(names, counts,
                                                                 m.batch_size, m.num_steps, epoch_size)):

        cost, _ = session.run([m.cost, eval_op],
                              {m.input_data: x,
                               m.targets: y,
                               m.initial_state: m.initial_state.eval(),
                               m.weights: np.ones(m.batch_size * m.num_steps)})
        costs += cost
        iters += m.num_steps

        if verbose and step % (epoch_size // 10) == 9:
            print("%.3f perplexity: %.3f speed: %.0f lps" %
                  (step * 1.0 / epoch_size, np.exp(costs / iters),
                   iters * m.batch_size / (time.time() - start_time)))

        if step >= epoch_size:
            break

    return np.exp(costs / iters)
Example #9
Source File: names.py    From AI_Reader with Apache License 2.0 5 votes vote down vote up 
def train(data_dir, checkpoint_path, config):
    """Trains the model with the given data

    Args:
        data_dir: path to the data for the model (see data_utils for data
            format)
        checkpoint_path: the path to save the trained model checkpoints
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # Prepare Name data.
    print("Reading Name data in %s" % data_dir)
    names, counts = data_utils.read_names(data_dir)

    with tf.Graph().as_default(), tf.Session() as session:
        initializer = tf.random_uniform_initializer(-config.init_scale,
                                                    config.init_scale)
        with tf.variable_scope("model", reuse=None, initializer=initializer):
            m = NamignizerModel(is_training=True, config=config)

        tf.initialize_all_variables().run()

        for i in range(config.max_max_epoch):
            lr_decay = config.lr_decay ** max(i - config.max_epoch, 0.0)
            m.assign_lr(session, config.learning_rate * lr_decay)

            print("Epoch: %d Learning rate: %.3f" % (i + 1, session.run(m.lr)))
            train_perplexity = run_epoch(session, m, names, counts, config.epoch_size, m.train_op,
                                         verbose=True)
            print("Epoch: %d Train Perplexity: %.3f" %
                  (i + 1, train_perplexity))

            m.saver.save(session, checkpoint_path, global_step=i)
Example #10
Source File: names.py    From AI_Reader with Apache License 2.0 5 votes vote down vote up 
def namignize(names, checkpoint_path, config):
    """Recognizes names and prints the Perplexity of the model for each names
    in the list

    Args:
        names: a list of names in the model format
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        for name in names:
            x, y = data_utils.name_to_batch(name, m.batch_size, m.num_steps)

            cost, loss, _ = session.run([m.cost, m.loss, tf.no_op()],
                                  {m.input_data: x,
                                   m.targets: y,
                                   m.initial_state: m.initial_state.eval(),
                                   m.weights: np.concatenate((
                                       np.ones(len(name)), np.zeros(m.batch_size * m.num_steps - len(name))))})

            print("Name {} gives us a perplexity of {}".format(
                name, np.exp(cost)))
Example #11
Source File: names.py    From HumanRecognition with MIT License 5 votes vote down vote up 
def run_epoch(session, m, names, counts, epoch_size, eval_op, verbose=False):
    """Runs the model on the given data for one epoch

    Args:
        session: the tf session holding the model graph
        m: an instance of the NamignizerModel
        names: a set of lowercase names of 26 characters
        counts: a list of the frequency of the above names
        epoch_size: the number of batches to run
        eval_op: whether to change the params or not, and how to do it
    Kwargs:
        verbose: whether to print out state of training during the epoch
    Returns:
        cost: the average cost during the last stage of the epoch
    """
    start_time = time.time()
    costs = 0.0
    iters = 0
    for step, (x, y) in enumerate(data_utils.namignizer_iterator(names, counts,
                                                                 m.batch_size, m.num_steps, epoch_size)):

        cost, _ = session.run([m.cost, eval_op],
                              {m.input_data: x,
                               m.targets: y,
                               m.weights: np.ones(m.batch_size * m.num_steps)})
        costs += cost
        iters += m.num_steps

        if verbose and step % (epoch_size // 10) == 9:
            print("%.3f perplexity: %.3f speed: %.0f lps" %
                  (step * 1.0 / epoch_size, np.exp(costs / iters),
                   iters * m.batch_size / (time.time() - start_time)))

        if step >= epoch_size:
            break

    return np.exp(costs / iters)
Example #12
Source File: names.py    From HumanRecognition with MIT License 5 votes vote down vote up 
def train(data_dir, checkpoint_path, config):
    """Trains the model with the given data

    Args:
        data_dir: path to the data for the model (see data_utils for data
            format)
        checkpoint_path: the path to save the trained model checkpoints
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # Prepare Name data.
    print("Reading Name data in %s" % data_dir)
    names, counts = data_utils.read_names(data_dir)

    with tf.Graph().as_default(), tf.Session() as session:
        initializer = tf.random_uniform_initializer(-config.init_scale,
                                                    config.init_scale)
        with tf.variable_scope("model", reuse=None, initializer=initializer):
            m = NamignizerModel(is_training=True, config=config)

        tf.global_variables_initializer().run()

        for i in range(config.max_max_epoch):
            lr_decay = config.lr_decay ** max(i - config.max_epoch, 0.0)
            m.assign_lr(session, config.learning_rate * lr_decay)

            print("Epoch: %d Learning rate: %.3f" % (i + 1, session.run(m.lr)))
            train_perplexity = run_epoch(session, m, names, counts, config.epoch_size, m.train_op,
                                         verbose=True)
            print("Epoch: %d Train Perplexity: %.3f" %
                  (i + 1, train_perplexity))

            m.saver.save(session, checkpoint_path, global_step=i)
Example #13
Source File: names.py    From HumanRecognition with MIT License 5 votes vote down vote up 
def namignize(names, checkpoint_path, config):
    """Recognizes names and prints the Perplexity of the model for each names
    in the list

    Args:
        names: a list of names in the model format
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        for name in names:
            x, y = data_utils.name_to_batch(name, m.batch_size, m.num_steps)

            cost, loss, _ = session.run([m.cost, m.loss, tf.no_op()],
                                  {m.input_data: x,
                                   m.targets: y,
                                   m.weights: np.concatenate((
                                       np.ones(len(name)), np.zeros(m.batch_size * m.num_steps - len(name))))})

            print("Name {} gives us a perplexity of {}".format(
                name, np.exp(cost)))
Example #14
Source File: names.py    From object_detection_kitti with Apache License 2.0 5 votes vote down vote up 
def namignize(names, checkpoint_path, config):
    """Recognizes names and prints the Perplexity of the model for each names
    in the list

    Args:
        names: a list of names in the model format
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        for name in names:
            x, y = data_utils.name_to_batch(name, m.batch_size, m.num_steps)

            cost, loss, _ = session.run([m.cost, m.loss, tf.no_op()],
                                  {m.input_data: x,
                                   m.targets: y,
                                   m.weights: np.concatenate((
                                       np.ones(len(name)), np.zeros(m.batch_size * m.num_steps - len(name))))})

            print("Name {} gives us a perplexity of {}".format(
                name, np.exp(cost)))
Example #15
Source File: names.py    From g-tensorflow-models with Apache License 2.0 5 votes vote down vote up 
def run_epoch(session, m, names, counts, epoch_size, eval_op, verbose=False):
    """Runs the model on the given data for one epoch

    Args:
        session: the tf session holding the model graph
        m: an instance of the NamignizerModel
        names: a set of lowercase names of 26 characters
        counts: a list of the frequency of the above names
        epoch_size: the number of batches to run
        eval_op: whether to change the params or not, and how to do it
    Kwargs:
        verbose: whether to print out state of training during the epoch
    Returns:
        cost: the average cost during the last stage of the epoch
    """
    start_time = time.time()
    costs = 0.0
    iters = 0
    for step, (x, y) in enumerate(data_utils.namignizer_iterator(names, counts,
                                                                 m.batch_size, m.num_steps, epoch_size)):

        cost, _ = session.run([m.cost, eval_op],
                              {m.input_data: x,
                               m.targets: y,
                               m.weights: np.ones(m.batch_size * m.num_steps)})
        costs += cost
        iters += m.num_steps

        if verbose and step % (epoch_size // 10) == 9:
            print("%.3f perplexity: %.3f speed: %.0f lps" %
                  (step * 1.0 / epoch_size, np.exp(costs / iters),
                   iters * m.batch_size / (time.time() - start_time)))

        if step >= epoch_size:
            break

    return np.exp(costs / iters)
Example #16
Source File: names.py    From g-tensorflow-models with Apache License 2.0 5 votes vote down vote up 
def train(data_dir, checkpoint_path, config):
    """Trains the model with the given data

    Args:
        data_dir: path to the data for the model (see data_utils for data
            format)
        checkpoint_path: the path to save the trained model checkpoints
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # Prepare Name data.
    print("Reading Name data in %s" % data_dir)
    names, counts = data_utils.read_names(data_dir)

    with tf.Graph().as_default(), tf.Session() as session:
        initializer = tf.random_uniform_initializer(-config.init_scale,
                                                    config.init_scale)
        with tf.variable_scope("model", reuse=None, initializer=initializer):
            m = NamignizerModel(is_training=True, config=config)

        tf.global_variables_initializer().run()

        for i in range(config.max_max_epoch):
            lr_decay = config.lr_decay ** max(i - config.max_epoch, 0.0)
            m.assign_lr(session, config.learning_rate * lr_decay)

            print("Epoch: %d Learning rate: %.3f" % (i + 1, session.run(m.lr)))
            train_perplexity = run_epoch(session, m, names, counts, config.epoch_size, m.train_op,
                                         verbose=True)
            print("Epoch: %d Train Perplexity: %.3f" %
                  (i + 1, train_perplexity))

            m.saver.save(session, checkpoint_path, global_step=i)
Example #17
Source File: names.py    From g-tensorflow-models with Apache License 2.0 5 votes vote down vote up 
def namignize(names, checkpoint_path, config):
    """Recognizes names and prints the Perplexity of the model for each names
    in the list

    Args:
        names: a list of names in the model format
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        for name in names:
            x, y = data_utils.name_to_batch(name, m.batch_size, m.num_steps)

            cost, loss, _ = session.run([m.cost, m.loss, tf.no_op()],
                                  {m.input_data: x,
                                   m.targets: y,
                                   m.weights: np.concatenate((
                                       np.ones(len(name)), np.zeros(m.batch_size * m.num_steps - len(name))))})

            print("Name {} gives us a perplexity of {}".format(
                name, np.exp(cost)))
Example #18
Source File: names.py    From g-tensorflow-models with Apache License 2.0 5 votes vote down vote up 
def namignator(checkpoint_path, config):
    """Generates names randomly according to a given model

    Args:
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # mutate the config to become a name generator config
    config.num_steps = 1
    config.batch_size = 1

    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                  {m.input_data: np.zeros((1, 1)),
                                                   m.targets: np.zeros((1, 1)),
                                                   m.weights: np.ones(1)})

        # sample from our softmax activations
        next_letter = np.random.choice(27, p=activations[0])
        name = [next_letter]
        while next_letter != 0:
            activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                      {m.input_data: [[next_letter]],
                                                       m.targets: np.zeros((1, 1)),
                                                       m.initial_state: final_state,
                                                       m.weights: np.ones(1)})

            next_letter = np.random.choice(27, p=activations[0])
            name += [next_letter]

        print(map(lambda x: chr(x + 96), name))
Example #19
Source File: names.py    From models with Apache License 2.0 5 votes vote down vote up 
def run_epoch(session, m, names, counts, epoch_size, eval_op, verbose=False):
    """Runs the model on the given data for one epoch

    Args:
        session: the tf session holding the model graph
        m: an instance of the NamignizerModel
        names: a set of lowercase names of 26 characters
        counts: a list of the frequency of the above names
        epoch_size: the number of batches to run
        eval_op: whether to change the params or not, and how to do it
    Kwargs:
        verbose: whether to print out state of training during the epoch
    Returns:
        cost: the average cost during the last stage of the epoch
    """
    start_time = time.time()
    costs = 0.0
    iters = 0
    for step, (x, y) in enumerate(data_utils.namignizer_iterator(names, counts,
                                                                 m.batch_size, m.num_steps, epoch_size)):

        cost, _ = session.run([m.cost, eval_op],
                              {m.input_data: x,
                               m.targets: y,
                               m.weights: np.ones(m.batch_size * m.num_steps)})
        costs += cost
        iters += m.num_steps

        if verbose and step % (epoch_size // 10) == 9:
            print("%.3f perplexity: %.3f speed: %.0f lps" %
                  (step * 1.0 / epoch_size, np.exp(costs / iters),
                   iters * m.batch_size / (time.time() - start_time)))

        if step >= epoch_size:
            break

    return np.exp(costs / iters)
Example #20
Source File: names.py    From models with Apache License 2.0 5 votes vote down vote up 
def train(data_dir, checkpoint_path, config):
    """Trains the model with the given data

    Args:
        data_dir: path to the data for the model (see data_utils for data
            format)
        checkpoint_path: the path to save the trained model checkpoints
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # Prepare Name data.
    print("Reading Name data in %s" % data_dir)
    names, counts = data_utils.read_names(data_dir)

    with tf.Graph().as_default(), tf.Session() as session:
        initializer = tf.random_uniform_initializer(-config.init_scale,
                                                    config.init_scale)
        with tf.variable_scope("model", reuse=None, initializer=initializer):
            m = NamignizerModel(is_training=True, config=config)

        tf.global_variables_initializer().run()

        for i in range(config.max_max_epoch):
            lr_decay = config.lr_decay ** max(i - config.max_epoch, 0.0)
            m.assign_lr(session, config.learning_rate * lr_decay)

            print("Epoch: %d Learning rate: %.3f" % (i + 1, session.run(m.lr)))
            train_perplexity = run_epoch(session, m, names, counts, config.epoch_size, m.train_op,
                                         verbose=True)
            print("Epoch: %d Train Perplexity: %.3f" %
                  (i + 1, train_perplexity))

            m.saver.save(session, checkpoint_path, global_step=i)
Example #21
Source File: names.py    From models with Apache License 2.0 5 votes vote down vote up 
def namignize(names, checkpoint_path, config):
    """Recognizes names and prints the Perplexity of the model for each names
    in the list

    Args:
        names: a list of names in the model format
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        for name in names:
            x, y = data_utils.name_to_batch(name, m.batch_size, m.num_steps)

            cost, loss, _ = session.run([m.cost, m.loss, tf.no_op()],
                                  {m.input_data: x,
                                   m.targets: y,
                                   m.weights: np.concatenate((
                                       np.ones(len(name)), np.zeros(m.batch_size * m.num_steps - len(name))))})

            print("Name {} gives us a perplexity of {}".format(
                name, np.exp(cost)))
Example #22
Source File: names.py    From models with Apache License 2.0 5 votes vote down vote up 
def namignator(checkpoint_path, config):
    """Generates names randomly according to a given model

    Args:
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # mutate the config to become a name generator config
    config.num_steps = 1
    config.batch_size = 1

    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                  {m.input_data: np.zeros((1, 1)),
                                                   m.targets: np.zeros((1, 1)),
                                                   m.weights: np.ones(1)})

        # sample from our softmax activations
        next_letter = np.random.choice(27, p=activations[0])
        name = [next_letter]
        while next_letter != 0:
            activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                      {m.input_data: [[next_letter]],
                                                       m.targets: np.zeros((1, 1)),
                                                       m.initial_state: final_state,
                                                       m.weights: np.ones(1)})

            next_letter = np.random.choice(27, p=activations[0])
            name += [next_letter]

        print(map(lambda x: chr(x + 96), name))
Example #23
Source File: names.py    From multilabel-image-classification-tensorflow with MIT License 5 votes vote down vote up 
def run_epoch(session, m, names, counts, epoch_size, eval_op, verbose=False):
    """Runs the model on the given data for one epoch

    Args:
        session: the tf session holding the model graph
        m: an instance of the NamignizerModel
        names: a set of lowercase names of 26 characters
        counts: a list of the frequency of the above names
        epoch_size: the number of batches to run
        eval_op: whether to change the params or not, and how to do it
    Kwargs:
        verbose: whether to print out state of training during the epoch
    Returns:
        cost: the average cost during the last stage of the epoch
    """
    start_time = time.time()
    costs = 0.0
    iters = 0
    for step, (x, y) in enumerate(data_utils.namignizer_iterator(names, counts,
                                                                 m.batch_size, m.num_steps, epoch_size)):

        cost, _ = session.run([m.cost, eval_op],
                              {m.input_data: x,
                               m.targets: y,
                               m.weights: np.ones(m.batch_size * m.num_steps)})
        costs += cost
        iters += m.num_steps

        if verbose and step % (epoch_size // 10) == 9:
            print("%.3f perplexity: %.3f speed: %.0f lps" %
                  (step * 1.0 / epoch_size, np.exp(costs / iters),
                   iters * m.batch_size / (time.time() - start_time)))

        if step >= epoch_size:
            break

    return np.exp(costs / iters)
Example #24
Source File: names.py    From multilabel-image-classification-tensorflow with MIT License 5 votes vote down vote up 
def train(data_dir, checkpoint_path, config):
    """Trains the model with the given data

    Args:
        data_dir: path to the data for the model (see data_utils for data
            format)
        checkpoint_path: the path to save the trained model checkpoints
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # Prepare Name data.
    print("Reading Name data in %s" % data_dir)
    names, counts = data_utils.read_names(data_dir)

    with tf.Graph().as_default(), tf.Session() as session:
        initializer = tf.random_uniform_initializer(-config.init_scale,
                                                    config.init_scale)
        with tf.variable_scope("model", reuse=None, initializer=initializer):
            m = NamignizerModel(is_training=True, config=config)

        tf.global_variables_initializer().run()

        for i in range(config.max_max_epoch):
            lr_decay = config.lr_decay ** max(i - config.max_epoch, 0.0)
            m.assign_lr(session, config.learning_rate * lr_decay)

            print("Epoch: %d Learning rate: %.3f" % (i + 1, session.run(m.lr)))
            train_perplexity = run_epoch(session, m, names, counts, config.epoch_size, m.train_op,
                                         verbose=True)
            print("Epoch: %d Train Perplexity: %.3f" %
                  (i + 1, train_perplexity))

            m.saver.save(session, checkpoint_path, global_step=i)
Example #25
Source File: names.py    From multilabel-image-classification-tensorflow with MIT License 5 votes vote down vote up 
def namignize(names, checkpoint_path, config):
    """Recognizes names and prints the Perplexity of the model for each names
    in the list

    Args:
        names: a list of names in the model format
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        for name in names:
            x, y = data_utils.name_to_batch(name, m.batch_size, m.num_steps)

            cost, loss, _ = session.run([m.cost, m.loss, tf.no_op()],
                                  {m.input_data: x,
                                   m.targets: y,
                                   m.weights: np.concatenate((
                                       np.ones(len(name)), np.zeros(m.batch_size * m.num_steps - len(name))))})

            print("Name {} gives us a perplexity of {}".format(
                name, np.exp(cost)))
Example #26
Source File: names.py    From multilabel-image-classification-tensorflow with MIT License 5 votes vote down vote up 
def namignator(checkpoint_path, config):
    """Generates names randomly according to a given model

    Args:
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # mutate the config to become a name generator config
    config.num_steps = 1
    config.batch_size = 1

    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                  {m.input_data: np.zeros((1, 1)),
                                                   m.targets: np.zeros((1, 1)),
                                                   m.weights: np.ones(1)})

        # sample from our softmax activations
        next_letter = np.random.choice(27, p=activations[0])
        name = [next_letter]
        while next_letter != 0:
            activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                      {m.input_data: [[next_letter]],
                                                       m.targets: np.zeros((1, 1)),
                                                       m.initial_state: final_state,
                                                       m.weights: np.ones(1)})

            next_letter = np.random.choice(27, p=activations[0])
            name += [next_letter]

        print(map(lambda x: chr(x + 96), name))
Example #27
Source File: names.py    From Action_Recognition_Zoo with MIT License 5 votes vote down vote up 
def train(data_dir, checkpoint_path, config):
    """Trains the model with the given data

    Args:
        data_dir: path to the data for the model (see data_utils for data
            format)
        checkpoint_path: the path to save the trained model checkpoints
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # Prepare Name data.
    print("Reading Name data in %s" % data_dir)
    names, counts = data_utils.read_names(data_dir)

    with tf.Graph().as_default(), tf.Session() as session:
        initializer = tf.random_uniform_initializer(-config.init_scale,
                                                    config.init_scale)
        with tf.variable_scope("model", reuse=None, initializer=initializer):
            m = NamignizerModel(is_training=True, config=config)

        tf.initialize_all_variables().run()

        for i in range(config.max_max_epoch):
            lr_decay = config.lr_decay ** max(i - config.max_epoch, 0.0)
            m.assign_lr(session, config.learning_rate * lr_decay)

            print("Epoch: %d Learning rate: %.3f" % (i + 1, session.run(m.lr)))
            train_perplexity = run_epoch(session, m, names, counts, config.epoch_size, m.train_op,
                                         verbose=True)
            print("Epoch: %d Train Perplexity: %.3f" %
                  (i + 1, train_perplexity))

            m.saver.save(session, checkpoint_path, global_step=i)
Example #28
Source File: names.py    From DOTA_models with Apache License 2.0 5 votes vote down vote up 
def train(data_dir, checkpoint_path, config):
    """Trains the model with the given data

    Args:
        data_dir: path to the data for the model (see data_utils for data
            format)
        checkpoint_path: the path to save the trained model checkpoints
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # Prepare Name data.
    print("Reading Name data in %s" % data_dir)
    names, counts = data_utils.read_names(data_dir)

    with tf.Graph().as_default(), tf.Session() as session:
        initializer = tf.random_uniform_initializer(-config.init_scale,
                                                    config.init_scale)
        with tf.variable_scope("model", reuse=None, initializer=initializer):
            m = NamignizerModel(is_training=True, config=config)

        tf.global_variables_initializer().run()

        for i in range(config.max_max_epoch):
            lr_decay = config.lr_decay ** max(i - config.max_epoch, 0.0)
            m.assign_lr(session, config.learning_rate * lr_decay)

            print("Epoch: %d Learning rate: %.3f" % (i + 1, session.run(m.lr)))
            train_perplexity = run_epoch(session, m, names, counts, config.epoch_size, m.train_op,
                                         verbose=True)
            print("Epoch: %d Train Perplexity: %.3f" %
                  (i + 1, train_perplexity))

            m.saver.save(session, checkpoint_path, global_step=i)
Example #29
Source File: names.py    From DOTA_models with Apache License 2.0 5 votes vote down vote up 
def namignize(names, checkpoint_path, config):
    """Recognizes names and prints the Perplexity of the model for each names
    in the list

    Args:
        names: a list of names in the model format
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        for name in names:
            x, y = data_utils.name_to_batch(name, m.batch_size, m.num_steps)

            cost, loss, _ = session.run([m.cost, m.loss, tf.no_op()],
                                  {m.input_data: x,
                                   m.targets: y,
                                   m.weights: np.concatenate((
                                       np.ones(len(name)), np.zeros(m.batch_size * m.num_steps - len(name))))})

            print("Name {} gives us a perplexity of {}".format(
                name, np.exp(cost)))
Example #30
Source File: names.py    From DOTA_models with Apache License 2.0 5 votes vote down vote up 
def namignator(checkpoint_path, config):
    """Generates names randomly according to a given model

    Args:
        checkpoint_path: the path to restore the trained model from, should not
            include the model name, just the path to
        config: one of the above configs that specify the model and how it
            should be run and trained
    Returns:
        None
    """
    # mutate the config to become a name generator config
    config.num_steps = 1
    config.batch_size = 1

    with tf.Graph().as_default(), tf.Session() as session:

        with tf.variable_scope("model"):
            m = NamignizerModel(is_training=False, config=config)

        m.saver.restore(session, checkpoint_path)

        activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                  {m.input_data: np.zeros((1, 1)),
                                                   m.targets: np.zeros((1, 1)),
                                                   m.weights: np.ones(1)})

        # sample from our softmax activations
        next_letter = np.random.choice(27, p=activations[0])
        name = [next_letter]
        while next_letter != 0:
            activations, final_state, _ = session.run([m.activations, m.final_state, tf.no_op()],
                                                      {m.input_data: [[next_letter]],
                                                       m.targets: np.zeros((1, 1)),
                                                       m.initial_state: final_state,
                                                       m.weights: np.ones(1)})

            next_letter = np.random.choice(27, p=activations[0])
            name += [next_letter]

        print(map(lambda x: chr(x + 96), name))