Python tensorflow.InvalidArgumentError() Examples

The following are 11 code examples of tensorflow.InvalidArgumentError(). 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 .
Example #1
Source File: track_perplexity.py    From DOTA_models with Apache License 2.0 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
  """Evaluates the latest model checkpoint.

  Args:
    model: Instance of SkipThoughtsModel; the model to evaluate.
    losses: Tensor; the target cross entropy losses for the current batch.
    weights: A Tensor of weights corresponding to losses.
    saver: Instance of tf.train.Saver for restoring model Variables.
    summary_writer: Instance of FileWriter.
    summary_op: Op for generating model summaries.
  """
  model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
  if not model_path:
    tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                    FLAGS.checkpoint_dir)
    return

  with tf.Session() as sess:
    # Load model from checkpoint.
    tf.logging.info("Loading model from checkpoint: %s", model_path)
    saver.restore(sess, model_path)
    global_step = tf.train.global_step(sess, model.global_step.name)
    tf.logging.info("Successfully loaded %s at global step = %d.",
                    os.path.basename(model_path), global_step)
    if global_step < FLAGS.min_global_step:
      tf.logging.info("Skipping evaluation. Global step = %d < %d", global_step,
                      FLAGS.min_global_step)
      return

    # Start the queue runners.
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    num_eval_batches = int(
        math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

    # Run evaluation on the latest checkpoint.
    try:
      evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                     summary_writer, summary_op)
    except tf.InvalidArgumentError:
      tf.logging.error(
          "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
    finally:
      coord.request_stop()
      coord.join(threads, stop_grace_period_secs=10)
Example #2
Source File: track_perplexity.py    From yolo_v2 with Apache License 2.0 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
  """Evaluates the latest model checkpoint.

  Args:
    model: Instance of SkipThoughtsModel; the model to evaluate.
    losses: Tensor; the target cross entropy losses for the current batch.
    weights: A Tensor of weights corresponding to losses.
    saver: Instance of tf.train.Saver for restoring model Variables.
    summary_writer: Instance of FileWriter.
    summary_op: Op for generating model summaries.
  """
  model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
  if not model_path:
    tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                    FLAGS.checkpoint_dir)
    return

  with tf.Session() as sess:
    # Load model from checkpoint.
    tf.logging.info("Loading model from checkpoint: %s", model_path)
    saver.restore(sess, model_path)
    global_step = tf.train.global_step(sess, model.global_step.name)
    tf.logging.info("Successfully loaded %s at global step = %d.",
                    os.path.basename(model_path), global_step)
    if global_step < FLAGS.min_global_step:
      tf.logging.info("Skipping evaluation. Global step = %d < %d", global_step,
                      FLAGS.min_global_step)
      return

    # Start the queue runners.
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    num_eval_batches = int(
        math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

    # Run evaluation on the latest checkpoint.
    try:
      evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                     summary_writer, summary_op)
    except tf.InvalidArgumentError:
      tf.logging.error(
          "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
    finally:
      coord.request_stop()
      coord.join(threads, stop_grace_period_secs=10)
Example #3
Source File: track_perplexity.py    From parallax with Apache License 2.0 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
    """Evaluates the latest model checkpoint.

    Args:
      model: Instance of SkipThoughtsModel; the model to evaluate.
      losses: Tensor; the target cross entropy losses for the current batch.
      weights: A Tensor of weights corresponding to losses.
      saver: Instance of tf.train.Saver for restoring model Variables.
      summary_writer: Instance of FileWriter.
      summary_op: Op for generating model summaries.
    """
    model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
    if not model_path:
        tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                        FLAGS.checkpoint_dir)
        return

    with tf.Session() as sess:
        # Load model from checkpoint.
        tf.logging.info("Loading model from checkpoint: %s", model_path)
        saver.restore(sess, model_path)
        global_step = tf.train.global_step(sess, model.global_step.name)
        tf.logging.info("Successfully loaded %s at global step = %d.",
                        os.path.basename(model_path), global_step)
        if global_step < FLAGS.min_global_step:
            tf.logging.info("Skipping evaluation. Global step = %d < %d",
                            global_step,
                            FLAGS.min_global_step)
            return

        # Start the queue runners.
        coord = tf.train.Coordinator()
        threads = tf.train.start_queue_runners(coord=coord)

        num_eval_batches = int(
            math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

        # Run evaluation on the latest checkpoint.
        try:
            evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                           summary_writer, summary_op)
        except tf.InvalidArgumentError:
            tf.logging.error(
                "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
        finally:
            coord.request_stop()
            coord.join(threads, stop_grace_period_secs=10)
Example #4
Source File: track_perplexity.py    From Gun-Detector with Apache License 2.0 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
  """Evaluates the latest model checkpoint.

  Args:
    model: Instance of SkipThoughtsModel; the model to evaluate.
    losses: Tensor; the target cross entropy losses for the current batch.
    weights: A Tensor of weights corresponding to losses.
    saver: Instance of tf.train.Saver for restoring model Variables.
    summary_writer: Instance of FileWriter.
    summary_op: Op for generating model summaries.
  """
  model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
  if not model_path:
    tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                    FLAGS.checkpoint_dir)
    return

  with tf.Session() as sess:
    # Load model from checkpoint.
    tf.logging.info("Loading model from checkpoint: %s", model_path)
    saver.restore(sess, model_path)
    global_step = tf.train.global_step(sess, model.global_step.name)
    tf.logging.info("Successfully loaded %s at global step = %d.",
                    os.path.basename(model_path), global_step)
    if global_step < FLAGS.min_global_step:
      tf.logging.info("Skipping evaluation. Global step = %d < %d", global_step,
                      FLAGS.min_global_step)
      return

    # Start the queue runners.
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    num_eval_batches = int(
        math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

    # Run evaluation on the latest checkpoint.
    try:
      evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                     summary_writer, summary_op)
    except tf.InvalidArgumentError:
      tf.logging.error(
          "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
    finally:
      coord.request_stop()
      coord.join(threads, stop_grace_period_secs=10)
Example #5
Source File: track_perplexity.py    From hands-detection with MIT License 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
  """Evaluates the latest model checkpoint.

  Args:
    model: Instance of SkipThoughtsModel; the model to evaluate.
    losses: Tensor; the target cross entropy losses for the current batch.
    weights: A Tensor of weights corresponding to losses.
    saver: Instance of tf.train.Saver for restoring model Variables.
    summary_writer: Instance of FileWriter.
    summary_op: Op for generating model summaries.
  """
  model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
  if not model_path:
    tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                    FLAGS.checkpoint_dir)
    return

  with tf.Session() as sess:
    # Load model from checkpoint.
    tf.logging.info("Loading model from checkpoint: %s", model_path)
    saver.restore(sess, model_path)
    global_step = tf.train.global_step(sess, model.global_step.name)
    tf.logging.info("Successfully loaded %s at global step = %d.",
                    os.path.basename(model_path), global_step)
    if global_step < FLAGS.min_global_step:
      tf.logging.info("Skipping evaluation. Global step = %d < %d", global_step,
                      FLAGS.min_global_step)
      return

    # Start the queue runners.
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    num_eval_batches = int(
        math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

    # Run evaluation on the latest checkpoint.
    try:
      evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                     summary_writer, summary_op)
    except tf.InvalidArgumentError:
      tf.logging.error(
          "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
    finally:
      coord.request_stop()
      coord.join(threads, stop_grace_period_secs=10)
Example #6
Source File: track_perplexity.py    From object_detection_kitti with Apache License 2.0 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
  """Evaluates the latest model checkpoint.

  Args:
    model: Instance of SkipThoughtsModel; the model to evaluate.
    losses: Tensor; the target cross entropy losses for the current batch.
    weights: A Tensor of weights corresponding to losses.
    saver: Instance of tf.train.Saver for restoring model Variables.
    summary_writer: Instance of FileWriter.
    summary_op: Op for generating model summaries.
  """
  model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
  if not model_path:
    tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                    FLAGS.checkpoint_dir)
    return

  with tf.Session() as sess:
    # Load model from checkpoint.
    tf.logging.info("Loading model from checkpoint: %s", model_path)
    saver.restore(sess, model_path)
    global_step = tf.train.global_step(sess, model.global_step.name)
    tf.logging.info("Successfully loaded %s at global step = %d.",
                    os.path.basename(model_path), global_step)
    if global_step < FLAGS.min_global_step:
      tf.logging.info("Skipping evaluation. Global step = %d < %d", global_step,
                      FLAGS.min_global_step)
      return

    # Start the queue runners.
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    num_eval_batches = int(
        math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

    # Run evaluation on the latest checkpoint.
    try:
      evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                     summary_writer, summary_op)
    except tf.InvalidArgumentError:
      tf.logging.error(
          "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
    finally:
      coord.request_stop()
      coord.join(threads, stop_grace_period_secs=10)
Example #7
Source File: track_perplexity.py    From object_detection_with_tensorflow with MIT License 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
  """Evaluates the latest model checkpoint.

  Args:
    model: Instance of SkipThoughtsModel; the model to evaluate.
    losses: Tensor; the target cross entropy losses for the current batch.
    weights: A Tensor of weights corresponding to losses.
    saver: Instance of tf.train.Saver for restoring model Variables.
    summary_writer: Instance of FileWriter.
    summary_op: Op for generating model summaries.
  """
  model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
  if not model_path:
    tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                    FLAGS.checkpoint_dir)
    return

  with tf.Session() as sess:
    # Load model from checkpoint.
    tf.logging.info("Loading model from checkpoint: %s", model_path)
    saver.restore(sess, model_path)
    global_step = tf.train.global_step(sess, model.global_step.name)
    tf.logging.info("Successfully loaded %s at global step = %d.",
                    os.path.basename(model_path), global_step)
    if global_step < FLAGS.min_global_step:
      tf.logging.info("Skipping evaluation. Global step = %d < %d", global_step,
                      FLAGS.min_global_step)
      return

    # Start the queue runners.
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    num_eval_batches = int(
        math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

    # Run evaluation on the latest checkpoint.
    try:
      evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                     summary_writer, summary_op)
    except tf.InvalidArgumentError:
      tf.logging.error(
          "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
    finally:
      coord.request_stop()
      coord.join(threads, stop_grace_period_secs=10)
Example #8
Source File: track_perplexity.py    From HumanRecognition with MIT License 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
  """Evaluates the latest model checkpoint.

  Args:
    model: Instance of SkipThoughtsModel; the model to evaluate.
    losses: Tensor; the target cross entropy losses for the current batch.
    weights: A Tensor of weights corresponding to losses.
    saver: Instance of tf.train.Saver for restoring model Variables.
    summary_writer: Instance of FileWriter.
    summary_op: Op for generating model summaries.
  """
  model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
  if not model_path:
    tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                    FLAGS.checkpoint_dir)
    return

  with tf.Session() as sess:
    # Load model from checkpoint.
    tf.logging.info("Loading model from checkpoint: %s", model_path)
    saver.restore(sess, model_path)
    global_step = tf.train.global_step(sess, model.global_step.name)
    tf.logging.info("Successfully loaded %s at global step = %d.",
                    os.path.basename(model_path), global_step)
    if global_step < FLAGS.min_global_step:
      tf.logging.info("Skipping evaluation. Global step = %d < %d", global_step,
                      FLAGS.min_global_step)
      return

    # Start the queue runners.
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    num_eval_batches = int(
        math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

    # Run evaluation on the latest checkpoint.
    try:
      evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                     summary_writer, summary_op)
    except tf.InvalidArgumentError:
      tf.logging.error(
          "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
    finally:
      coord.request_stop()
      coord.join(threads, stop_grace_period_secs=10)
Example #9
Source File: track_perplexity.py    From g-tensorflow-models with Apache License 2.0 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
  """Evaluates the latest model checkpoint.

  Args:
    model: Instance of SkipThoughtsModel; the model to evaluate.
    losses: Tensor; the target cross entropy losses for the current batch.
    weights: A Tensor of weights corresponding to losses.
    saver: Instance of tf.train.Saver for restoring model Variables.
    summary_writer: Instance of FileWriter.
    summary_op: Op for generating model summaries.
  """
  model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
  if not model_path:
    tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                    FLAGS.checkpoint_dir)
    return

  with tf.Session() as sess:
    # Load model from checkpoint.
    tf.logging.info("Loading model from checkpoint: %s", model_path)
    saver.restore(sess, model_path)
    global_step = tf.train.global_step(sess, model.global_step.name)
    tf.logging.info("Successfully loaded %s at global step = %d.",
                    os.path.basename(model_path), global_step)
    if global_step < FLAGS.min_global_step:
      tf.logging.info("Skipping evaluation. Global step = %d < %d", global_step,
                      FLAGS.min_global_step)
      return

    # Start the queue runners.
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    num_eval_batches = int(
        math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

    # Run evaluation on the latest checkpoint.
    try:
      evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                     summary_writer, summary_op)
    except tf.InvalidArgumentError:
      tf.logging.error(
          "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
    finally:
      coord.request_stop()
      coord.join(threads, stop_grace_period_secs=10)
Example #10
Source File: track_perplexity.py    From models with Apache License 2.0 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
  """Evaluates the latest model checkpoint.

  Args:
    model: Instance of SkipThoughtsModel; the model to evaluate.
    losses: Tensor; the target cross entropy losses for the current batch.
    weights: A Tensor of weights corresponding to losses.
    saver: Instance of tf.train.Saver for restoring model Variables.
    summary_writer: Instance of FileWriter.
    summary_op: Op for generating model summaries.
  """
  model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
  if not model_path:
    tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                    FLAGS.checkpoint_dir)
    return

  with tf.Session() as sess:
    # Load model from checkpoint.
    tf.logging.info("Loading model from checkpoint: %s", model_path)
    saver.restore(sess, model_path)
    global_step = tf.train.global_step(sess, model.global_step.name)
    tf.logging.info("Successfully loaded %s at global step = %d.",
                    os.path.basename(model_path), global_step)
    if global_step < FLAGS.min_global_step:
      tf.logging.info("Skipping evaluation. Global step = %d < %d", global_step,
                      FLAGS.min_global_step)
      return

    # Start the queue runners.
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    num_eval_batches = int(
        math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

    # Run evaluation on the latest checkpoint.
    try:
      evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                     summary_writer, summary_op)
    except tf.InvalidArgumentError:
      tf.logging.error(
          "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
    finally:
      coord.request_stop()
      coord.join(threads, stop_grace_period_secs=10)
Example #11
Source File: track_perplexity.py    From multilabel-image-classification-tensorflow with MIT License 4 votes vote down vote up 
def run_once(model, losses, weights, saver, summary_writer, summary_op):
  """Evaluates the latest model checkpoint.

  Args:
    model: Instance of SkipThoughtsModel; the model to evaluate.
    losses: Tensor; the target cross entropy losses for the current batch.
    weights: A Tensor of weights corresponding to losses.
    saver: Instance of tf.train.Saver for restoring model Variables.
    summary_writer: Instance of FileWriter.
    summary_op: Op for generating model summaries.
  """
  model_path = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
  if not model_path:
    tf.logging.info("Skipping evaluation. No checkpoint found in: %s",
                    FLAGS.checkpoint_dir)
    return

  with tf.Session() as sess:
    # Load model from checkpoint.
    tf.logging.info("Loading model from checkpoint: %s", model_path)
    saver.restore(sess, model_path)
    global_step = tf.train.global_step(sess, model.global_step.name)
    tf.logging.info("Successfully loaded %s at global step = %d.",
                    os.path.basename(model_path), global_step)
    if global_step < FLAGS.min_global_step:
      tf.logging.info("Skipping evaluation. Global step = %d < %d", global_step,
                      FLAGS.min_global_step)
      return

    # Start the queue runners.
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    num_eval_batches = int(
        math.ceil(FLAGS.num_eval_examples / model.config.batch_size))

    # Run evaluation on the latest checkpoint.
    try:
      evaluate_model(sess, losses, weights, num_eval_batches, global_step,
                     summary_writer, summary_op)
    except tf.InvalidArgumentError:
      tf.logging.error(
          "Evaluation raised InvalidArgumentError (e.g. due to Nans).")
    finally:
      coord.request_stop()
      coord.join(threads, stop_grace_period_secs=10)