问题

My question is about how to get batch inputs from multiple (or sharded) tfrecords. I've read the example https://github.com/tensorflow/models/blob/master/inception/inception/image_processing.py#L410. The basic pipeline is, take the training set as as example, (1) first generate a series of tfrecords (e.g., train-000-of-005, train-001-of-005, ...), (2) from these filenames, generate a list and fed them into the tf.train.string_input_producer to get a queue, (3) simultaneously generate a tf.RandomShuffleQueue to do other stuff, (4) using tf.train.batch_join to generate batch inputs.

I think this is complex, and I'm not sure the logic of this procedure. In my case, I have a list of .npy files, and I want to generate sharded tfrecords(multiple seperated tfrecords, not just one single large file). Each of these .npy files contains different number of positive and negative samples (2 classes). A basic method is to generate one single large tfrecord file. But the file is too large (~20Gb). So I resort to sharded tfrecords. Are there any simpler way to do this? Thanks.

回答1:

The whole process is simplied using the Dataset API. Here are both the parts: (1): Convert numpy array to tfrecords and (2,3,4): read the tfrecords to generate batches.

1. Creation of tfrecords from a numpy array:

    def npy_to_tfrecords(...):
       # write records to a tfrecords file
       writer = tf.python_io.TFRecordWriter(output_file)

       # Loop through all the features you want to write
       for ... :
          let say X is of np.array([[...][...]])
          let say y is of np.array[[0/1]]

         # Feature contains a map of string to feature proto objects
         feature = {}
         feature['X'] = tf.train.Feature(float_list=tf.train.FloatList(value=X.flatten()))
         feature['y'] = tf.train.Feature(int64_list=tf.train.Int64List(value=y))

         # Construct the Example proto object
         example = tf.train.Example(features=tf.train.Features(feature=feature))

         # Serialize the example to a string
         serialized = example.SerializeToString()

         # write the serialized objec to the disk
         writer.write(serialized)
      writer.close()

2. Read the tfrecords using the Dataset API (tensorflow >=1.2):

    # Creates a dataset that reads all of the examples from filenames.
    filenames = ["file1.tfrecord", "file2.tfrecord", ..."fileN.tfrecord"]
    dataset = tf.contrib.data.TFRecordDataset(filenames)
    # for version 1.5 and above use tf.data.TFRecordDataset

    # example proto decode
    def _parse_function(example_proto):
      keys_to_features = {'X':tf.FixedLenFeature((shape_of_npy_array), tf.float32),
                          'y': tf.FixedLenFeature((), tf.int64, default_value=0)}
      parsed_features = tf.parse_single_example(example_proto, keys_to_features)
     return parsed_features['X'], parsed_features['y']

    # Parse the record into tensors.
    dataset = dataset.map(_parse_function)  

    # Shuffle the dataset
    dataset = dataset.shuffle(buffer_size=10000)

    # Repeat the input indefinitly
    dataset = dataset.repeat()  

    # Generate batches
    dataset = dataset.batch(batch_size)

    # Create a one-shot iterator
    iterator = dataset.make_one_shot_iterator()

    # Get batch X and y
    X, y = iterator.get_next()

来源：https://stackoverflow.com/questions/45427637/numpy-to-tfrecords-is-there-a-more-simple-way-to-handle-batch-inputs-from-tfrec

标签

python