In [ ]:
#@title Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
tf.Example|
|
Note: This is an archived TF1 notebook. These are configured to run in TF2's compatbility mode but will run in TF1 as well. To use TF1 in Colab, use the magic.
To read data efficiently it can be helpful to serialize your data and store it in a set of files (100-200MB each) that can each be read linearly. This is especially true if the data is being streamed over a network. This can also be useful for caching any data-preprocessing.
The TFRecord format is a simple format for storing a sequence of binary records.
Protocol buffers are a cross-platform, cross-language library for efficient serialization of structured data.
Protocol messages are defined by .proto files, these are often the easiest way to understand a message type.
The tf.Example message (or protobuf) is a flexible message type that represents a {"string": value} mapping. It is designed for use with TensorFlow and is used throughout the higher-level APIs such as TFX.
This notebook will demonstrate how to create, parse, and use the tf.Example message, and then serialize, write, and read tf.Example messages to and from .tfrecord files.
Note: While useful, these structures are optional. There is no need to convert existing code to use TFRecords, unless you are using tf.data and reading data is still the bottleneck to training. See Data Input Pipeline Performance for dataset performance tips.
In [ ]:
import tensorflow.compat.v1 as tf
import numpy as np
import IPython.display as display
Fundamentally a tf.Example is a {"string": tf.train.Feature} mapping.
The tf.train.Feature message type can accept one of the following three types (See the .proto file for reference). Most other generic types can be coerced into one of these.
tf.train.BytesList (the following types can be coerced)
stringbytetf.train.FloatList (the following types can be coerced)
float (float32)double (float64)tf.train.Int64List (the following types can be coerced)
boolenumint32uint32int64uint64In order to convert a standard TensorFlow type to a tf.Example-compatible tf.train.Feature, you can use the following shortcut functions:
Each function takes a scalar input value and returns a tf.train.Feature containing one of the three list types above.
In [ ]:
# The following functions can be used to convert a value to a type compatible
# with tf.Example.
def _bytes_feature(value):
"""Returns a bytes_list from a string / byte."""
return tf.train.Feature(bytes_list=tf.train.BytesList(value=[value]))
def _float_feature(value):
"""Returns a float_list from a float / double."""
return tf.train.Feature(float_list=tf.train.FloatList(value=[value]))
def _int64_feature(value):
"""Returns an int64_list from a bool / enum / int / uint."""
return tf.train.Feature(int64_list=tf.train.Int64List(value=[value]))
Note: To stay simple, this example only uses scalar inputs. The simplest way to handle non-scalar features is to use tf.serialize_tensor to convert tensors to binary-strings. Strings are scalars in tensorflow. Use tf.parse_tensor to convert the binary-string back to a tensor.
Below are some examples of how these functions work. Note the varying input types and the standardized output types. If the input type for a function does not match one of the coercible types stated above, the function will raise an exception (e.g. _int64_feature(1.0) will error out, since 1.0 is a float, so should be used with the _float_feature function instead).
In [ ]:
print(_bytes_feature(b'test_string'))
print(_bytes_feature(u'test_bytes'.encode('utf-8')))
print(_float_feature(np.exp(1)))
print(_int64_feature(True))
print(_int64_feature(1))
All proto messages can be serialized to a binary-string using the .SerializeToString method.
In [ ]:
feature = _float_feature(np.exp(1))
feature.SerializeToString()
Suppose you want to create a tf.Example message from existing data. In practice, the dataset may come from anywhere, but the procedure of creating the tf.Example message from a single observation will be the same.
Within each observation, each value needs to be converted to a tf.train.Feature containing one of the 3 compatible types, using one of the functions above.
We create a map (dictionary) from the feature name string to the encoded feature value produced in #1.
The map produced in #2 is converted to a Features message.
In this notebook, we will create a dataset using NumPy.
This dataset will have 4 features.
False or True with equal probability[0, 5)Consider a sample consisting of 10,000 independently and identically distributed observations from each of the above distributions.
In [ ]:
# the number of observations in the dataset
n_observations = int(1e4)
# boolean feature, encoded as False or True
feature0 = np.random.choice([False, True], n_observations)
# integer feature, random from 0 .. 4
feature1 = np.random.randint(0, 5, n_observations)
# string feature
strings = np.array([b'cat', b'dog', b'chicken', b'horse', b'goat'])
feature2 = strings[feature1]
# float feature, from a standard normal distribution
feature3 = np.random.randn(n_observations)
Each of these features can be coerced into a tf.Example-compatible type using one of _bytes_feature, _float_feature, _int64_feature. We can then create a tf.Example message from these encoded features.
In [ ]:
def serialize_example(feature0, feature1, feature2, feature3):
"""
Creates a tf.Example message ready to be written to a file.
"""
# Create a dictionary mapping the feature name to the tf.Example-compatible
# data type.
feature = {
'feature0': _int64_feature(feature0),
'feature1': _int64_feature(feature1),
'feature2': _bytes_feature(feature2),
'feature3': _float_feature(feature3),
}
# Create a Features message using tf.train.Example.
example_proto = tf.train.Example(features=tf.train.Features(feature=feature))
return example_proto.SerializeToString()
For example, suppose we have a single observation from the dataset, [False, 4, bytes('goat'), 0.9876]. We can create and print the tf.Example message for this observation using create_message(). Each single observation will be written as a Features message as per the above. Note that the tf.Example message is just a wrapper around the Features message.
In [ ]:
# This is an example observation from the dataset.
example_observation = []
serialized_example = serialize_example(False, 4, b'goat', 0.9876)
serialized_example
To decode the message use the tf.train.Example.FromString method.
In [ ]:
example_proto = tf.train.Example.FromString(serialized_example)
example_proto
A TFRecord file contains a sequence of records. The file can only be read sequentially.
Each record contains a byte-string, for the data-payload, plus the data-length, and CRC32C (32-bit CRC using the Castagnoli polynomial) hashes for integrity checking.
Each record has the format
uint64 length
uint32 masked_crc32_of_length
byte data[length]
uint32 masked_crc32_of_data
The records are concatenated together to produce the file. CRCs are described here, and the mask of a CRC is
masked_crc = ((crc >> 15) | (crc << 17)) + 0xa282ead8ul
Note: There is no requirement to use tf.Example in TFRecord files. tf.Example is just a method of serializing dictionaries to byte-strings. Lines of text, encoded image data, or serialized tensors (using tf.io.serialize_tensor, and
tf.io.parse_tensor when loading). See the tf.io module for more options.
The tf.data module also provides tools for reading and writing data in tensorflow.
In [ ]:
tf.data.Dataset.from_tensor_slices(feature1)
Applies to a tuple of arrays, it returns a dataset of tuples:
In [ ]:
features_dataset = tf.data.Dataset.from_tensor_slices((feature0, feature1, feature2, feature3))
features_dataset
In [ ]:
# Use `take(1)` to only pull one example from the dataset.
for f0,f1,f2,f3 in features_dataset.take(1):
print(f0)
print(f1)
print(f2)
print(f3)
Use the tf.data.Dataset.map method to apply a function to each element of a Dataset.
The mapped function must operate in TensorFlow graph mode: It must operate on and return tf.Tensors. A non-tensor function, like serialize_example, can be wrapped with tf.py_function to make it compatible.
We define a similar function serialize_example_pyfunction below with a minor change - converting eagerTensor objects returned by the tf.py_function to numpy arrays as required by our _bytes_feature, _float_feature and _int64_feature functions.
Using tf.py_function requires that you specify the shape and type information that is otherwise unavailable:
In [ ]:
def serialize_example_pyfunction(feature0, feature1, feature2, feature3):
"""
Creates a tf.Example message ready to be written to a file.
"""
# Create a dictionary mapping the feature name to the tf.Example-compatible
# data type.
feature = {
'feature0': _int64_feature(feature0.numpy()),
'feature1': _int64_feature(feature1.numpy()),
'feature2': _bytes_feature(feature2.numpy()),
'feature3': _float_feature(feature3.numpy()),
}
# Create a Features message using tf.train.Example.
example_proto = tf.train.Example(features=tf.train.Features(feature=feature))
return example_proto.SerializeToString()
In [ ]:
def tf_serialize_example(f0,f1,f2,f3):
tf_string = tf.py_function(
serialize_example_pyfunction,
(f0,f1,f2,f3), # pass these args to the above function.
tf.string) # the return type is `tf.string`.
return tf.reshape(tf_string, ()) # The result is a scalar
Apply this function to each element in the dataset:
In [ ]:
serialized_features_dataset = features_dataset.map(tf_serialize_example)
serialized_features_dataset
And write them to a TFRecord file:
In [ ]:
filename = 'test.tfrecord'
writer = tf.data.experimental.TFRecordWriter(filename)
writer.write(serialized_features_dataset)
We can also read the TFRecord file using the tf.data.TFRecordDataset class.
More information on consuming TFRecord files using tf.data can be found here.
Using TFRecordDatasets can be useful for standardizing input data and optimizing performance.
In [ ]:
filenames = [filename]
raw_dataset = tf.data.TFRecordDataset(filenames)
raw_dataset
At this point the dataset contains serialized tf.train.Example messages. When iterated over it returns these as scalar string tensors.
Use the .take method to only show the first 10 records.
Note: iterating over a tf.data.Dataset only works with eager execution enabled.
In [ ]:
for raw_record in raw_dataset.take(10):
print(repr(raw_record))
These tensors can be parsed using the function below.
Note: The feature_description is necessary here because datasets use graph-execution, and need this description to build their shape and type signature.
In [ ]:
# Create a description of the features.
feature_description = {
'feature0': tf.FixedLenFeature([], tf.int64, default_value=0),
'feature1': tf.FixedLenFeature([], tf.int64, default_value=0),
'feature2': tf.FixedLenFeature([], tf.string, default_value=''),
'feature3': tf.FixedLenFeature([], tf.float32, default_value=0.0),
}
def _parse_function(example_proto):
# Parse the input tf.Example proto using the dictionary above.
return tf.parse_single_example(example_proto, feature_description)
Or use tf.parse example to parse a whole batch at once.
Apply this finction to each item in the dataset using the tf.data.Dataset.map method:
In [ ]:
parsed_dataset = raw_dataset.map(_parse_function)
parsed_dataset
Use eager execution to display the observations in the dataset. There are 10,000 observations in this dataset, but we only display the first 10. The data is displayed as a dictionary of features. Each item is a tf.Tensor, and the numpy element of this tensor displays the value of the feature.
In [ ]:
for parsed_record in parsed_dataset.take(10):
print(repr(parsed_record))
Here, the tf.parse_example function unpacks the tf.Example fields into standard tensors.
The tf.python_io module also contains pure-Python functions for reading and writing TFRecord files.
Now write the 10,000 observations to the file test.tfrecords. Each observation is converted to a tf.Example message, then written to file. We can then verify that the file test.tfrecords has been created.
In [ ]:
# Write the `tf.Example` observations to the file.
with tf.python_io.TFRecordWriter(filename) as writer:
for i in range(n_observations):
example = serialize_example(feature0[i], feature1[i], feature2[i], feature3[i])
writer.write(example)
In [ ]:
!ls
Suppose we now want to read this data back, to be input as data into a model.
The following example imports the data as is, as a tf.Example message. This can be useful to verify that a file contains the data that we expect. This can also be useful if the input data is stored as TFRecords but you would prefer to input NumPy data (or some other input data type), for example here, since this example allows us to read the values themselves.
We iterate through the TFRecords in the infile, extract the tf.Example message, and can read/store the values within.
In [ ]:
record_iterator = tf.python_io.tf_record_iterator(path=filename)
for string_record in record_iterator:
example = tf.train.Example()
example.ParseFromString(string_record)
print(example)
# Exit after 1 iteration as this is purely demonstrative.
break
The features of the example object (created above of type tf.Example) can be accessed using its getters (similarly to any protocol buffer message). example.features returns a repeated feature message, then getting the feature message returns a map of feature name to feature value (stored in Python as a dictionary).
In [ ]:
print(dict(example.features.feature))
From this dictionary, you can get any given value as with a dictionary.
In [ ]:
print(example.features.feature['feature3'])
Now, we can access the value using the getters again.
In [ ]:
print(example.features.feature['feature3'].float_list.value)
This is an example of how to read and write image data using TFRecords. The purpose of this is to show how, end to end, input data (in this case an image) and write the data as a TFRecord file, then read the file back and display the image.
This can be useful if, for example, you want to use several models on the same input dataset. Instead of storing the image data raw, it can be preprocessed into the TFRecords format, and that can be used in all further processing and modelling.
First, let's download this image of a cat in the snow and this photo of the Williamsburg Bridge, NYC under construction.
In [ ]:
cat_in_snow = tf.keras.utils.get_file('320px-Felis_catus-cat_on_snow.jpg', 'https://storage.googleapis.com/download.tensorflow.org/example_images/320px-Felis_catus-cat_on_snow.jpg')
williamsburg_bridge = tf.keras.utils.get_file('194px-New_East_River_Bridge_from_Brooklyn_det.4a09796u.jpg','https://storage.googleapis.com/download.tensorflow.org/example_images/194px-New_East_River_Bridge_from_Brooklyn_det.4a09796u.jpg')
In [ ]:
display.display(display.Image(filename=cat_in_snow))
display.display(display.HTML('Image cc-by: <a "href=https://commons.wikimedia.org/wiki/File:Felis_catus-cat_on_snow.jpg">Von.grzanka</a>'))
In [ ]:
display.display(display.Image(filename=williamsburg_bridge))
display.display(display.HTML('<a "href=https://commons.wikimedia.org/wiki/File:New_East_River_Bridge_from_Brooklyn_det.4a09796u.jpg">source</a>'))
As we did earlier, we can now encode the features as types compatible with tf.Example. In this case, we will not only store the raw image string as a feature, but we will store the height, width, depth, and an arbitrary label feature, which is used when we write the file to distinguish between the cat image and the bridge image. We will use 0 for the cat image, and 1 for the bridge image.
In [ ]:
image_labels = {
cat_in_snow : 0,
williamsburg_bridge : 1,
}
In [ ]:
# This is an example, just using the cat image.
image_string = open(cat_in_snow, 'rb').read()
label = image_labels[cat_in_snow]
# Create a dictionary with features that may be relevant.
def image_example(image_string, label):
image_shape = tf.image.decode_jpeg(image_string).shape
feature = {
'height': _int64_feature(image_shape[0]),
'width': _int64_feature(image_shape[1]),
'depth': _int64_feature(image_shape[2]),
'label': _int64_feature(label),
'image_raw': _bytes_feature(image_string),
}
return tf.train.Example(features=tf.train.Features(feature=feature))
for line in str(image_example(image_string, label)).split('\n')[:15]:
print(line)
print('...')
We see that all of the features are now stores in the tf.Example message. Now, we functionalize the code above and write the example messages to a file, images.tfrecords.
In [ ]:
# Write the raw image files to images.tfrecords.
# First, process the two images into tf.Example messages.
# Then, write to a .tfrecords file.
with tf.python_io.TFRecordWriter('images.tfrecords') as writer:
for filename, label in image_labels.items():
image_string = open(filename, 'rb').read()
tf_example = image_example(image_string, label)
writer.write(tf_example.SerializeToString())
In [ ]:
!ls
We now have the file images.tfrecords. We can now iterate over the records in the file to read back what we wrote. Since, for our use case we will just reproduce the image, the only feature we need is the raw image string. We can extract that using the getters described above, namely example.features.feature['image_raw'].bytes_list.value[0]. We also use the labels to determine which record is the cat as opposed to the bridge.
In [ ]:
raw_image_dataset = tf.data.TFRecordDataset('images.tfrecords')
# Create a dictionary describing the features.
image_feature_description = {
'height': tf.FixedLenFeature([], tf.int64),
'width': tf.FixedLenFeature([], tf.int64),
'depth': tf.FixedLenFeature([], tf.int64),
'label': tf.FixedLenFeature([], tf.int64),
'image_raw': tf.FixedLenFeature([], tf.string),
}
def _parse_image_function(example_proto):
# Parse the input tf.Example proto using the dictionary above.
return tf.parse_single_example(example_proto, image_feature_description)
parsed_image_dataset = raw_image_dataset.map(_parse_image_function)
parsed_image_dataset
Recover the images from the TFRecord file:
In [ ]:
for image_features in parsed_image_dataset:
image_raw = image_features['image_raw'].numpy()
display.display(display.Image(data=image_raw))