CODEWITHSUNDEEP
pythonstringtensorflowmachine-learningdeep-learning
Peter Lawrence
Peter Lawrence

Reputation: 749

Tensorflow: cannot extract filename from tfrecord

I have written an image, label and filename to a tfrecords file. When I try to decode the file, I cannot convert the filename to a string from tf.string.

The code I wrote to convert it to a tfrecords file:

num_batches = 6
batch_size = math.ceil(X_training.shape[0] / num_batches)

for i in range(num_batches):
    train_path = os.path.join("data","batch_" + str(i) + '.tfrecords')
    writer = tf.python_io.TFRecordWriter(train_path)
    start_row = i * batch_size
    end_row = start_row + batch_size - 1

    for idx in range(start_row, end_row):
        try:
            label = y_tr[idx]
            filename = train_filenames[idx].tostring()
            image = X_tr[idx]
            image_raw = image.tostring()
        except:
            continue

        example = tf.train.Example(
            features=tf.train.Features(
              feature={
                'label': _int64_feature(label),
                'filename': _bytes_feature(filename),
                'image': _bytes_feature(image_raw),
              }))

        serialized = example.SerializeToString()
        writer.write(serialized)

To read and decode a tfrecords file I have the function:

def read_and_decode_single_example(filenames):
    filename_queue = tf.train.string_input_producer(filenames)

    reader = tf.TFRecordReader()

    _, serialized_example = reader.read(filename_queue)
    features = tf.parse_single_example(
        serialized_example,
        features={
            'label': tf.FixedLenFeature([], tf.int64),
            'filename': tf.FixedLenFeature([], tf.string),
            'image': tf.FixedLenFeature([], tf.string)
        })


    label = features['label']
    image = tf.decode_raw(features['image'], tf.uint8)
    image = tf.reshape(image, [499, 499, 1])
    filename = features['filename']

    return label, image, filename

When I decode the different batches, the filename that gets returned looks like:

b'P\x00\x00\x00_\x00\x00\x000\x00\x00\x000\x00\x00\x001\x00\x00\x004\x00\x00\x008\x00\x00\x00_\x00\x00\x00R\x00\x00\x00I\x00\x00\x00G\x00\x00\x00H\x00\x00\x00T\x00\x00\x00_\x00\x00\x00M\x00\x00\x00L\x00\x00\x00O\x00\x00\x00.\x00\x00\x00j\x00\x00\x00p\x00\x00\x00g\x00\x00\x00'

What am I doing wrong in decoding from a tf.string?

Upvotes: 2

Views: 888

Answers (1)

twolffpiggott
twolffpiggott

Reputation: 1103

Calling .decode().replace('\x00', '') on your bytestring produces 'P_00148_RIGHT_MLO.jpg'.

Adding the decode and replace in the function return should solve your problem.

Upvotes: 1

Related Questions