Python tensorflow.Records方法代码示例

# 需要导入模块: import tensorflow [as 别名]
# 或者: from tensorflow import Records [as 别名]
def get_input_tensors(batch_size, tf_records, num_repeats=None,
                      shuffle_records=True, shuffle_examples=True,
                      shuffle_buffer_size=None,
                      filter_amount=0.05):
    '''Read tf.Records and prepare them for ingestion by dual_net.  See
    `read_tf_records` for parameter documentation.

    Returns a dict of tensors (see return value of batch_parse_tf_example)
    '''
    if shuffle_buffer_size is None:
        shuffle_buffer_size = SHUFFLE_BUFFER_SIZE
    dataset = read_tf_records(batch_size, tf_records, num_repeats=num_repeats,
                              shuffle_records=shuffle_records,
                              shuffle_examples=shuffle_examples,
                              shuffle_buffer_size=shuffle_buffer_size,
                              filter_amount=filter_amount)
    dataset = dataset.filter(lambda t: tf.equal(tf.shape(t)[0], batch_size))
    dataset = dataset.map(functools.partial(
        batch_parse_tf_example, batch_size))
    return dataset.make_one_shot_iterator().get_next()

# End-to-end utility functions 

# 需要导入模块: import tensorflow [as 别名]
# 或者: from tensorflow import Records [as 别名]
def check_data(self, tfrecords_filename):
        """Checks a specified tf.Records file for coreect dataformat.
        Check if the data format in the example files is correct. Prints the shape of the data
        stored in a tf.Records file.

        Args
          tfrecords_filename: `str`, the path to the `tf.records` file to check.
        """
        record_iterator = tf.python_io.tf_record_iterator(path=tfrecords_filename)

        for string_record in record_iterator:
            # Parse the next example
            example = tf.train.Example()
            example.ParseFromString(string_record)

            # Get the features you stored (change to match your tfrecord writing code)
            seq = (example.features.feature['seq_raw']
                   .bytes_list
                   .value[0])

            label = (example.features.feature['label_raw']
                     .bytes_list
                     .value[0])

            # Convert to a numpy array (change dtype to the datatype you stored)
            seq_array = np.fromstring(seq, dtype=np.float64)
            label_array = np.fromstring(label, dtype=np.float64)

            # Print the image shape; does it match your expectations?
            print(seq_array.shape)
            print(label_array.shape) 

# 需要导入模块: import tensorflow [as 别名]
# 或者: from tensorflow import Records [as 别名]
def get_input_tensors(batch_size, feature_layout, tf_records, num_repeats=1,
                      shuffle_records=True, shuffle_examples=True,
                      shuffle_buffer_size=None,
                      filter_amount=0.05, random_rotation=True):
    """Read tf.Records and prepare them for ingestion by dual_net.

    See `read_tf_records` for parameter documentation.

    Returns a dict of tensors (see return value of batch_parse_tf_example)
    """
    print("Reading tf_records from {} inputs".format(len(tf_records)))
    dataset = read_tf_records(
        batch_size,
        tf_records,
        num_repeats=num_repeats,
        shuffle_records=shuffle_records,
        shuffle_examples=shuffle_examples,
        shuffle_buffer_size=shuffle_buffer_size,
        filter_amount=filter_amount,
        interleave=False)
    dataset = dataset.filter(lambda t: tf.equal(tf.shape(t)[0], batch_size))
    dataset = dataset.map(
        functools.partial(batch_parse_tf_example, batch_size, feature_layout))
    if random_rotation:
        # Unbatch the dataset so we can rotate it
        dataset = dataset.apply(tf.data.experimental.unbatch())
        dataset = dataset.apply(tf.data.experimental.map_and_batch(
            functools.partial(_random_rotation, feature_layout),
            batch_size))

    return dataset.make_one_shot_iterator().get_next()