A Component-by-Component Introduction to TensorFlow Extended (TFX)
View on TensorFlow.org
Run in Google Colab
View source on GitHubThis Colab-based tutorial will interactively walk through each built-in component of TensorFlow Extended (TFX).
It covers every step in an end-to-end machine learning pipeline, from data ingestion to pushing a model to serving.
When you're done, the contents of this notebook can be automatically exported as TFX pipeline source code, which you can orchestrate with Apache Airflow and Apache Beam.
Background
This notebook demonstrates how to use TFX in a Jupyter/Colab environment. Here, we walk through the Chicago Taxi example in an interactive notebook.
Working in an interactive notebook is a useful way to become familiar with the structure of a TFX pipeline. It's also useful when doing development of your own pipelines as a lightweight development environment, but you should be aware that there are differences in the way interactive notebooks are orchestrated, and how they access metadata artifacts.
Orchestration
In a production deployment of TFX, you will use an orchestrator such as Apache Airflow, Kubeflow Pipelines, or Apache Beam to orchestrate a pre-defined pipeline graph of TFX components. In an interactive notebook, the notebook itself is the orchestrator, running each TFX component as you execute the notebook cells.
Metadata
In a production deployment of TFX, you will access metadata through the ML Metadata (MLMD) API. MLMD stores metadata properties in a database such as MySQL or SQLite, and stores the metadata payloads in a persistent store such as on your filesystem. In an interactive notebook, both properties and payloads are stored in an ephemeral SQLite database in the /tmp directory on the Jupyter notebook or Colab server.
Setup
First, we install and import the necessary packages, set up paths, and download data.
Install TFX
Note: Because of package updates, the first time you run this cell you must then restart the runtime (Runtime > Restart runtime ...).
# TODO(ccy): remove "pyzmq==17.0.0" pin after bug in Colab is fixed.
!pip install -q "tfx>=0.21.1,<0.22" "tensorflow>=2.1,<2.2" "tensorboard>=2.1,<2.2" "pyzmq==17.0.0"
ERROR: qtconsole 4.7.2 has requirement pyzmq>=17.1, but you'll have pyzmq 17.0.0 which is incompatible.
Did you restart the runtime?
The first time that you run the cell above, you must restart the runtime (Runtime > Restart runtime ...). This is because of the way that Colab loads packages.
Import packages
We import necessary packages, including standard TFX component classes.
import os
import pprint
import tempfile
import urllib
import absl
import tensorflow as tf
import tensorflow_model_analysis as tfma
tf.get_logger().propagate = False
pp = pprint.PrettyPrinter()
import tfx
from tfx.components import CsvExampleGen
from tfx.components import Evaluator
from tfx.components import ExampleValidator
from tfx.components import Pusher
from tfx.components import ResolverNode
from tfx.components import SchemaGen
from tfx.components import StatisticsGen
from tfx.components import Trainer
from tfx.components import Transform
from tfx.components.base import executor_spec
from tfx.components.trainer.executor import GenericExecutor
from tfx.dsl.experimental import latest_blessed_model_resolver
from tfx.orchestration import metadata
from tfx.orchestration import pipeline
from tfx.orchestration.experimental.interactive.interactive_context import InteractiveContext
from tfx.proto import pusher_pb2
from tfx.proto import trainer_pb2
from tfx.types import Channel
from tfx.types.standard_artifacts import Model
from tfx.types.standard_artifacts import ModelBlessing
from tfx.utils.dsl_utils import external_input
%load_ext tfx.orchestration.experimental.interactive.notebook_extensions.skip
/tmpfs/src/tf_docs_env/lib/python3.6/site-packages/tfx/orchestration/data_types.py:191: UserWarning: RuntimeParameter is only supported on KubeflowDagRunner currently.
warnings.warn('RuntimeParameter is only supported on KubeflowDagRunner '
Let's check the library versions.
print('TensorFlow version: {}'.format(tf.__version__))
print('TFX version: {}'.format(tfx.__version__))
TensorFlow version: 2.1.0 TFX version: 0.21.2
Set up pipeline paths
# This is the root directory for your TFX pip package installation.
_tfx_root = tfx.__path__[0]
# This is the directory containing the TFX Chicago Taxi Pipeline example.
_taxi_root = os.path.join(_tfx_root, 'examples/chicago_taxi_pipeline')
# This is the path where your model will be pushed for serving.
_serving_model_dir = os.path.join(
tempfile.mkdtemp(), 'serving_model/taxi_simple')
# Set up logging.
absl.logging.set_verbosity(absl.logging.INFO)
Download example data
We download the example dataset for use in our TFX pipeline.
The dataset we're using is the Taxi Trips dataset released by the City of Chicago. The columns in this dataset are:
| pickup_community_area | fare | trip_start_month |
| trip_start_hour | trip_start_day | trip_start_timestamp |
| pickup_latitude | pickup_longitude | dropoff_latitude |
| dropoff_longitude | trip_miles | pickup_census_tract |
| dropoff_census_tract | payment_type | company |
| trip_seconds | dropoff_community_area | tips |
With this dataset, we will build a model that predicts the tips of a trip.
_data_root = tempfile.mkdtemp(prefix='tfx-data')
DATA_PATH = 'https://raw.githubusercontent.com/tensorflow/tfx/master/tfx/examples/chicago_taxi_pipeline/data/simple/data.csv'
_data_filepath = os.path.join(_data_root, "data.csv")
urllib.request.urlretrieve(DATA_PATH, _data_filepath)
('/tmp/tfx-databrtt19ey/data.csv', <http.client.HTTPMessage at 0x7f2640765d68>)
Take a quick look at the CSV file.
head {_data_filepath}
pickup_community_area,fare,trip_start_month,trip_start_hour,trip_start_day,trip_start_timestamp,pickup_latitude,pickup_longitude,dropoff_latitude,dropoff_longitude,trip_miles,pickup_census_tract,dropoff_census_tract,payment_type,company,trip_seconds,dropoff_community_area,tips 60,27.05,10,2,3,1380593700,41.836150155,-87.648787952,,,12.6,,,Cash,Taxi Affiliation Services,1380,,0.0 10,5.85,10,1,2,1382319000,41.985015101,-87.804532006,,,0.0,,,Cash,Taxi Affiliation Services,180,,0.0 14,16.65,5,7,5,1369897200,41.968069,-87.721559063,,,0.0,,,Cash,Dispatch Taxi Affiliation,1080,,0.0 13,16.45,11,12,3,1446554700,41.983636307,-87.723583185,,,6.9,,,Cash,,780,,0.0 16,32.05,12,1,1,1417916700,41.953582125,-87.72345239,,,15.4,,,Cash,,1200,,0.0 30,38.45,10,10,5,1444301100,41.839086906,-87.714003807,,,14.6,,,Cash,,2580,,0.0 11,14.65,1,1,3,1358213400,41.978829526,-87.771166703,,,5.81,,,Cash,,1080,,0.0 33,3.25,5,17,1,1368985500,41.849246754,-87.624135298,,,0.0,,,Cash,Taxi Affiliation Services,0,,0.0 19,47.65,6,15,4,1372258800,41.927260956,-87.765501609,,,0.0,,,Cash,Taxi Affiliation Services,3480,,0.0
Disclaimer: This site provides applications using data that has been modified for use from its original source, www.cityofchicago.org, the official website of the City of Chicago. The City of Chicago makes no claims as to the content, accuracy, timeliness, or completeness of any of the data provided at this site. The data provided at this site is subject to change at any time. It is understood that the data provided at this site is being used at one’s own risk.
Create the InteractiveContext
Last, we create an InteractiveContext, which will allow us to run TFX components interactively in this notebook.
# Here, we create an InteractiveContext using default parameters. This will
# use a temporary directory with an ephemeral ML Metadata database instance.
# To use your own pipeline root or database, the optional properties
# `pipeline_root` and `metadata_connection_config` may be passed to
# InteractiveContext. Calls to InteractiveContext are no-ops outside of the
# notebook.
context = InteractiveContext()
WARNING:absl:InteractiveContext pipeline_root argument not provided: using temporary directory /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu as root for pipeline outputs. WARNING:absl:InteractiveContext metadata_connection_config not provided: using SQLite ML Metadata database at /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/metadata.sqlite.
Run TFX components interactively
In the cells that follow, we create TFX components one-by-one, run each of them, and visualize their output artifacts.
ExampleGen
The ExampleGen component is usually at the start of a TFX pipeline. It will:
- Split data into training and evaluation sets (by default, 2/3 training + 1/3 eval)
- Convert data into the
tf.Exampleformat - Copy data into the
_tfx_rootdirectory for other components to access
ExampleGen takes as input the path to your data source. In our case, this is the _data_root path that contains the downloaded CSV.
example_gen = CsvExampleGen(input=external_input(_data_root))
context.run(example_gen)
INFO:absl:Running driver for CsvExampleGen INFO:absl:MetadataStore with DB connection initialized INFO:absl:Running executor for CsvExampleGen INFO:absl:Generating examples. INFO:absl:Using 1 process(es) for Beam pipeline execution. INFO:absl:Processing input csv data /tmp/tfx-databrtt19ey/* to TFExample. WARNING:root:Couldn't find python-snappy so the implementation of _TFRecordUtil._masked_crc32c is not as fast as it could be. INFO:absl:Examples generated. INFO:absl:Running publisher for CsvExampleGen INFO:absl:MetadataStore with DB connection initialized
Let's examine the output artifacts of ExampleGen. This component produces two artifacts, training examples and evaluation examples:
%%skip_for_export
artifact = example_gen.outputs['examples'].get()[0]
print(artifact.split_names, artifact.uri)
["train", "eval"] /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/CsvExampleGen/examples/1 This cell will be skipped during export to pipeline.
We can also take a look at the first three training examples:
%%skip_for_export
# Get the URI of the output artifact representing the training examples, which is a directory
train_uri = os.path.join(example_gen.outputs['examples'].get()[0].uri, 'train')
# Get the list of files in this directory (all compressed TFRecord files)
tfrecord_filenames = [os.path.join(train_uri, name)
for name in os.listdir(train_uri)]
# Create a `TFRecordDataset` to read these files
dataset = tf.data.TFRecordDataset(tfrecord_filenames, compression_type="GZIP")
# Iterate over the first 3 records and decode them.
for tfrecord in dataset.take(3):
serialized_example = tfrecord.numpy()
example = tf.train.Example()
example.ParseFromString(serialized_example)
pp.pprint(example)
features {
feature {
key: "company"
value {
bytes_list {
value: "Taxi Affiliation Services"
}
}
}
feature {
key: "dropoff_census_tract"
value {
}
}
feature {
key: "dropoff_community_area"
value {
}
}
feature {
key: "dropoff_latitude"
value {
}
}
feature {
key: "dropoff_longitude"
value {
}
}
feature {
key: "fare"
value {
float_list {
value: 27.049999237060547
}
}
}
feature {
key: "payment_type"
value {
bytes_list {
value: "Cash"
}
}
}
feature {
key: "pickup_census_tract"
value {
}
}
feature {
key: "pickup_community_area"
value {
int64_list {
value: 60
}
}
}
feature {
key: "pickup_latitude"
value {
float_list {
value: 41.836151123046875
}
}
}
feature {
key: "pickup_longitude"
value {
float_list {
value: -87.64878845214844
}
}
}
feature {
key: "tips"
value {
float_list {
value: 0.0
}
}
}
feature {
key: "trip_miles"
value {
float_list {
value: 12.600000381469727
}
}
}
feature {
key: "trip_seconds"
value {
int64_list {
value: 1380
}
}
}
feature {
key: "trip_start_day"
value {
int64_list {
value: 3
}
}
}
feature {
key: "trip_start_hour"
value {
int64_list {
value: 2
}
}
}
feature {
key: "trip_start_month"
value {
int64_list {
value: 10
}
}
}
feature {
key: "trip_start_timestamp"
value {
int64_list {
value: 1380593700
}
}
}
}
features {
feature {
key: "company"
value {
bytes_list {
value: "Dispatch Taxi Affiliation"
}
}
}
feature {
key: "dropoff_census_tract"
value {
}
}
feature {
key: "dropoff_community_area"
value {
}
}
feature {
key: "dropoff_latitude"
value {
}
}
feature {
key: "dropoff_longitude"
value {
}
}
feature {
key: "fare"
value {
float_list {
value: 16.649999618530273
}
}
}
feature {
key: "payment_type"
value {
bytes_list {
value: "Cash"
}
}
}
feature {
key: "pickup_census_tract"
value {
}
}
feature {
key: "pickup_community_area"
value {
int64_list {
value: 14
}
}
}
feature {
key: "pickup_latitude"
value {
float_list {
value: 41.96806716918945
}
}
}
feature {
key: "pickup_longitude"
value {
float_list {
value: -87.7215576171875
}
}
}
feature {
key: "tips"
value {
float_list {
value: 0.0
}
}
}
feature {
key: "trip_miles"
value {
float_list {
value: 0.0
}
}
}
feature {
key: "trip_seconds"
value {
int64_list {
value: 1080
}
}
}
feature {
key: "trip_start_day"
value {
int64_list {
value: 5
}
}
}
feature {
key: "trip_start_hour"
value {
int64_list {
value: 7
}
}
}
feature {
key: "trip_start_month"
value {
int64_list {
value: 5
}
}
}
feature {
key: "trip_start_timestamp"
value {
int64_list {
value: 1369897200
}
}
}
}
features {
feature {
key: "company"
value {
}
}
feature {
key: "dropoff_census_tract"
value {
}
}
feature {
key: "dropoff_community_area"
value {
}
}
feature {
key: "dropoff_latitude"
value {
}
}
feature {
key: "dropoff_longitude"
value {
}
}
feature {
key: "fare"
value {
float_list {
value: 16.450000762939453
}
}
}
feature {
key: "payment_type"
value {
bytes_list {
value: "Cash"
}
}
}
feature {
key: "pickup_census_tract"
value {
}
}
feature {
key: "pickup_community_area"
value {
int64_list {
value: 13
}
}
}
feature {
key: "pickup_latitude"
value {
float_list {
value: 41.98363494873047
}
}
}
feature {
key: "pickup_longitude"
value {
float_list {
value: -87.72357940673828
}
}
}
feature {
key: "tips"
value {
float_list {
value: 0.0
}
}
}
feature {
key: "trip_miles"
value {
float_list {
value: 6.900000095367432
}
}
}
feature {
key: "trip_seconds"
value {
int64_list {
value: 780
}
}
}
feature {
key: "trip_start_day"
value {
int64_list {
value: 3
}
}
}
feature {
key: "trip_start_hour"
value {
int64_list {
value: 12
}
}
}
feature {
key: "trip_start_month"
value {
int64_list {
value: 11
}
}
}
feature {
key: "trip_start_timestamp"
value {
int64_list {
value: 1446554700
}
}
}
}
This cell will be skipped during export to pipeline.
Now that ExampleGen has finished ingesting the data, the next step is data analysis.
StatisticsGen
The StatisticsGen component computes statistics over your dataset for data analysis, as well as for use in downstream components. It uses the TensorFlow Data Validation library.
StatisticsGen takes as input the dataset we just ingested using ExampleGen.
statistics_gen = StatisticsGen(
examples=example_gen.outputs['examples'])
context.run(statistics_gen)
INFO:absl:Running driver for StatisticsGen INFO:absl:MetadataStore with DB connection initialized INFO:absl:Running executor for StatisticsGen INFO:absl:Using 1 process(es) for Beam pipeline execution. INFO:absl:Generating statistics for split train INFO:absl:Statistics for split train written to /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/StatisticsGen/statistics/2/train. INFO:absl:Generating statistics for split eval INFO:absl:Statistics for split eval written to /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/StatisticsGen/statistics/2/eval. /tmpfs/src/tf_docs_env/lib/python3.6/site-packages/tensorflow_data_validation/arrow/arrow_util.py:239: FutureWarning: Calling .data on ChunkedArray is provided for compatibility after Column was removed, simply drop this attribute types.FeaturePath([column_name]), column.data.chunk(0), weights): INFO:absl:Running publisher for StatisticsGen INFO:absl:MetadataStore with DB connection initialized
After StatisticsGen finishes running, we can visualize the outputted statistics. Try playing with the different plots!
%%skip_for_export
context.show(statistics_gen.outputs['statistics'])
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.6/site-packages/tensorflow_data_validation/utils/stats_gen_lib.py:366: tf_record_iterator (from tensorflow.python.lib.io.tf_record) is deprecated and will be removed in a future version. Instructions for updating: Use eager execution and: `tf.data.TFRecordDataset(path)`
This cell will be skipped during export to pipeline.
SchemaGen
The SchemaGen component generates a schema based on your data statistics. (A schema defines the expected bounds, types, and properties of the features in your dataset.) It also uses the TensorFlow Data Validation library.
SchemaGen will take as input the statistics that we generated with StatisticsGen, looking at the training split by default.
schema_gen = SchemaGen(
statistics=statistics_gen.outputs['statistics'],
infer_feature_shape=False)
context.run(schema_gen)
INFO:absl:Running driver for SchemaGen INFO:absl:MetadataStore with DB connection initialized INFO:absl:Running executor for SchemaGen INFO:absl:Infering schema from statistics. INFO:absl:Schema written to /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/SchemaGen/schema/3/schema.pbtxt. INFO:absl:Running publisher for SchemaGen INFO:absl:MetadataStore with DB connection initialized
After SchemaGen finishes running, we can visualize the generated schema as a table.
%%skip_for_export
context.show(schema_gen.outputs['schema'])
This cell will be skipped during export to pipeline.
Each feature in your dataset shows up as a row in the schema table, alongside its properties. The schema also captures all the values that a categorical feature takes on, denoted as its domain.
To learn more about schemas, see the SchemaGen documentation.
ExampleValidator
The ExampleValidator component detects anomalies in your data, based on the expectations defined by the schema. It also uses the TensorFlow Data Validation library.
ExampleValidator will take as input the statistics from StatisticsGen, and the schema from SchemaGen.
By default, it compares the statistics from the evaluation split to the schema from the training split.
example_validator = ExampleValidator(
statistics=statistics_gen.outputs['statistics'],
schema=schema_gen.outputs['schema'])
context.run(example_validator)
INFO:absl:Running driver for ExampleValidator INFO:absl:MetadataStore with DB connection initialized INFO:absl:Running executor for ExampleValidator INFO:absl:Validating schema against the computed statistics. INFO:absl:Validation complete. Anomalies written to /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/ExampleValidator/anomalies/4. INFO:absl:Running publisher for ExampleValidator INFO:absl:MetadataStore with DB connection initialized
After ExampleValidator finishes running, we can visualize the anomalies as a table.
%%skip_for_export
context.show(example_validator.outputs['anomalies'])
This cell will be skipped during export to pipeline.
In the anomalies table, we can see that the company feature takes on new values that were not in the training split. This information can be used to debug model performance, understand how your data evolves over time, and identify data errors.
In our case, this company anomaly is innocuous, but the payment_type could be fixed. For now we move on to the next step of transforming the data.
Transform
The Transform component performs feature engineering for both training and serving. It uses the TensorFlow Transform library.
Transform will take as input the data from ExampleGen, the schema from SchemaGen, as well as a module that contains user-defined Transform code.
Let's see an example of user-defined Transform code below (for an introduction to the TensorFlow Transform APIs, see the tutorial). First, we define a few constants for feature engineering:
_taxi_constants_module_file = 'taxi_constants.py'
%%skip_for_export
%%writefile {_taxi_constants_module_file}
# Categorical features are assumed to each have a maximum value in the dataset.
MAX_CATEGORICAL_FEATURE_VALUES = [24, 31, 12]
CATEGORICAL_FEATURE_KEYS = [
'trip_start_hour', 'trip_start_day', 'trip_start_month',
'pickup_census_tract', 'dropoff_census_tract', 'pickup_community_area',
'dropoff_community_area'
]
DENSE_FLOAT_FEATURE_KEYS = ['trip_miles', 'fare', 'trip_seconds']
# Number of buckets used by tf.transform for encoding each feature.
FEATURE_BUCKET_COUNT = 10
BUCKET_FEATURE_KEYS = [
'pickup_latitude', 'pickup_longitude', 'dropoff_latitude',
'dropoff_longitude'
]
# Number of vocabulary terms used for encoding VOCAB_FEATURES by tf.transform
VOCAB_SIZE = 1000
# Count of out-of-vocab buckets in which unrecognized VOCAB_FEATURES are hashed.
OOV_SIZE = 10
VOCAB_FEATURE_KEYS = [
'payment_type',
'company',
]
# Keys
LABEL_KEY = 'tips'
FARE_KEY = 'fare'
def transformed_name(key):
return key + '_xf'
Writing taxi_constants.py This cell will be skipped during export to pipeline.
Next, we write a preprocessing_fn that takes in raw data as input, and returns transformed features that our model can train on:
_taxi_transform_module_file = 'taxi_transform.py'
%%skip_for_export
%%writefile {_taxi_transform_module_file}
import tensorflow as tf
import tensorflow_transform as tft
import taxi_constants
_DENSE_FLOAT_FEATURE_KEYS = taxi_constants.DENSE_FLOAT_FEATURE_KEYS
_VOCAB_FEATURE_KEYS = taxi_constants.VOCAB_FEATURE_KEYS
_VOCAB_SIZE = taxi_constants.VOCAB_SIZE
_OOV_SIZE = taxi_constants.OOV_SIZE
_FEATURE_BUCKET_COUNT = taxi_constants.FEATURE_BUCKET_COUNT
_BUCKET_FEATURE_KEYS = taxi_constants.BUCKET_FEATURE_KEYS
_CATEGORICAL_FEATURE_KEYS = taxi_constants.CATEGORICAL_FEATURE_KEYS
_FARE_KEY = taxi_constants.FARE_KEY
_LABEL_KEY = taxi_constants.LABEL_KEY
_transformed_name = taxi_constants.transformed_name
def preprocessing_fn(inputs):
"""tf.transform's callback function for preprocessing inputs.
Args:
inputs: map from feature keys to raw not-yet-transformed features.
Returns:
Map from string feature key to transformed feature operations.
"""
outputs = {}
for key in _DENSE_FLOAT_FEATURE_KEYS:
# Preserve this feature as a dense float, setting nan's to the mean.
outputs[_transformed_name(key)] = tft.scale_to_z_score(
_fill_in_missing(inputs[key]))
for key in _VOCAB_FEATURE_KEYS:
# Build a vocabulary for this feature.
outputs[_transformed_name(key)] = tft.compute_and_apply_vocabulary(
_fill_in_missing(inputs[key]),
top_k=_VOCAB_SIZE,
num_oov_buckets=_OOV_SIZE)
for key in _BUCKET_FEATURE_KEYS:
outputs[_transformed_name(key)] = tft.bucketize(
_fill_in_missing(inputs[key]), _FEATURE_BUCKET_COUNT,
always_return_num_quantiles=False)
for key in _CATEGORICAL_FEATURE_KEYS:
outputs[_transformed_name(key)] = _fill_in_missing(inputs[key])
# Was this passenger a big tipper?
taxi_fare = _fill_in_missing(inputs[_FARE_KEY])
tips = _fill_in_missing(inputs[_LABEL_KEY])
outputs[_transformed_name(_LABEL_KEY)] = tf.where(
tf.math.is_nan(taxi_fare),
tf.cast(tf.zeros_like(taxi_fare), tf.int64),
# Test if the tip was > 20% of the fare.
tf.cast(
tf.greater(tips, tf.multiply(taxi_fare, tf.constant(0.2))), tf.int64))
return outputs
def _fill_in_missing(x):
"""Replace missing values in a SparseTensor.
Fills in missing values of `x` with '' or 0, and converts to a dense tensor.
Args:
x: A `SparseTensor` of rank 2. Its dense shape should have size at most 1
in the second dimension.
Returns:
A rank 1 tensor where missing values of `x` have been filled in.
"""
default_value = '' if x.dtype == tf.string else 0
return tf.squeeze(
tf.sparse.to_dense(
tf.SparseTensor(x.indices, x.values, [x.dense_shape[0], 1]),
default_value),
axis=1)
Writing taxi_transform.py This cell will be skipped during export to pipeline.
Now, we pass in this feature engineering code to the Transform component and run it to transform your data.
transform = Transform(
examples=example_gen.outputs['examples'],
schema=schema_gen.outputs['schema'],
module_file=os.path.abspath(_taxi_transform_module_file))
context.run(transform)
INFO:absl:Running driver for Transform INFO:absl:MetadataStore with DB connection initialized INFO:absl:Running executor for Transform WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.6/site-packages/tfx/components/transform/executor.py:501: Schema (from tensorflow_transform.tf_metadata.dataset_schema) is deprecated and will be removed in a future version. Instructions for updating: Schema is a deprecated, use schema_utils.schema_from_feature_spec to create a `Schema` INFO:absl:Using 1 process(es) for Beam pipeline execution. WARNING:tensorflow:Tensorflow version (2.1.0) found. Note that Tensorflow Transform support for TF 2.0 is currently in beta, and features such as tf.function may not work as intended. WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.6/site-packages/tensorflow_core/python/saved_model/signature_def_utils_impl.py:201: build_tensor_info (from tensorflow.python.saved_model.utils_impl) is deprecated and will be removed in a future version. Instructions for updating: This function will only be available through the v1 compatibility library as tf.compat.v1.saved_model.utils.build_tensor_info or tf.compat.v1.saved_model.build_tensor_info. INFO:tensorflow:Assets added to graph. INFO:tensorflow:No assets to write. WARNING:tensorflow:Issue encountered when serializing tft_mapper_use. Type is unsupported, or the types of the items don't match field type in CollectionDef. Note this is a warning and probably safe to ignore. 'Counter' object has no attribute 'name' INFO:tensorflow:SavedModel written to: /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Transform/transform_graph/5/.temp_path/tftransform_tmp/202a58dcd4314f28a055632ed2c2a236/saved_model.pb INFO:tensorflow:Assets added to graph. INFO:tensorflow:No assets to write. WARNING:tensorflow:Issue encountered when serializing tft_mapper_use. Type is unsupported, or the types of the items don't match field type in CollectionDef. Note this is a warning and probably safe to ignore. 'Counter' object has no attribute 'name' INFO:tensorflow:SavedModel written to: /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Transform/transform_graph/5/.temp_path/tftransform_tmp/1afb6213616c400a9dd3460f7eedb97a/saved_model.pb WARNING:tensorflow:Tensorflow version (2.1.0) found. Note that Tensorflow Transform support for TF 2.0 is currently in beta, and features such as tf.function may not work as intended. WARNING:tensorflow:Tensorflow version (2.1.0) found. Note that Tensorflow Transform support for TF 2.0 is currently in beta, and features such as tf.function may not work as intended. INFO:tensorflow:Saver not created because there are no variables in the graph to restore INFO:tensorflow:Saver not created because there are no variables in the graph to restore INFO:tensorflow:Assets added to graph. INFO:tensorflow:Assets written to: /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Transform/transform_graph/5/.temp_path/tftransform_tmp/8315801b407142dfb7794de9305f0be8/assets INFO:tensorflow:SavedModel written to: /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Transform/transform_graph/5/.temp_path/tftransform_tmp/8315801b407142dfb7794de9305f0be8/saved_model.pb WARNING:tensorflow:Expected binary or unicode string, got type_url: "type.googleapis.com/tensorflow.AssetFileDef" value: "\n\013\n\tConst_7:0\022-vocab_compute_and_apply_vocabulary_vocabulary" WARNING:tensorflow:Expected binary or unicode string, got type_url: "type.googleapis.com/tensorflow.AssetFileDef" value: "\n\013\n\tConst_9:0\022/vocab_compute_and_apply_vocabulary_1_vocabulary" INFO:tensorflow:Saver not created because there are no variables in the graph to restore WARNING:tensorflow:Expected binary or unicode string, got type_url: "type.googleapis.com/tensorflow.AssetFileDef" value: "\n\013\n\tConst_7:0\022-vocab_compute_and_apply_vocabulary_vocabulary" WARNING:tensorflow:Expected binary or unicode string, got type_url: "type.googleapis.com/tensorflow.AssetFileDef" value: "\n\013\n\tConst_9:0\022/vocab_compute_and_apply_vocabulary_1_vocabulary" INFO:tensorflow:Saver not created because there are no variables in the graph to restore WARNING:tensorflow:Expected binary or unicode string, got type_url: "type.googleapis.com/tensorflow.AssetFileDef" value: "\n\013\n\tConst_7:0\022-vocab_compute_and_apply_vocabulary_vocabulary" WARNING:tensorflow:Expected binary or unicode string, got type_url: "type.googleapis.com/tensorflow.AssetFileDef" value: "\n\013\n\tConst_9:0\022/vocab_compute_and_apply_vocabulary_1_vocabulary" INFO:tensorflow:Saver not created because there are no variables in the graph to restore INFO:absl:Running publisher for Transform INFO:absl:MetadataStore with DB connection initialized
Let's examine the output artifacts of Transform. This component produces two types of outputs:
transform_graphis the graph that can perform the preprocessing operations (this graph will be included in the serving and evaluation models).transformed_examplesrepresents the preprocessed training and evaluation data.
transform.outputs
{'transform_graph': Channel(
type_name: TransformGraph
artifacts: [Artifact(type_name: TransformGraph, uri: /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Transform/transform_graph/5, id: 6)]
), 'transformed_examples': Channel(
type_name: Examples
artifacts: [Artifact(type_name: Examples, uri: /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Transform/transformed_examples/5, id: 7)]
)}
Take a peek at the transform_graph artifact. It points to a directory containing three subdirectories.
train_uri = transform.outputs['transform_graph'].get()[0].uri
os.listdir(train_uri)
['transform_fn', 'transformed_metadata', 'metadata']
The transformed_metadata subdirectory contains the schema of the preprocessed data. The transform_fn subdirectory contains the actual preprocessing graph. The metadata subdirectory contains the schema of the original data.
We can also take a look at the first three transformed examples:
# Get the URI of the output artifact representing the transformed examples, which is a directory
train_uri = os.path.join(transform.outputs['transformed_examples'].get()[0].uri, 'train')
# Get the list of files in this directory (all compressed TFRecord files)
tfrecord_filenames = [os.path.join(train_uri, name)
for name in os.listdir(train_uri)]
# Create a `TFRecordDataset` to read these files
dataset = tf.data.TFRecordDataset(tfrecord_filenames, compression_type="GZIP")
# Iterate over the first 3 records and decode them.
for tfrecord in dataset.take(3):
serialized_example = tfrecord.numpy()
example = tf.train.Example()
example.ParseFromString(serialized_example)
pp.pprint(example)
features {
feature {
key: "company_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "dropoff_census_tract_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "dropoff_community_area_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "dropoff_latitude_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "dropoff_longitude_xf"
value {
int64_list {
value: 12
}
}
}
feature {
key: "fare_xf"
value {
float_list {
value: 1.5174685716629028
}
}
}
feature {
key: "payment_type_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "pickup_census_tract_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "pickup_community_area_xf"
value {
int64_list {
value: 60
}
}
}
feature {
key: "pickup_latitude_xf"
value {
int64_list {
value: 1
}
}
}
feature {
key: "pickup_longitude_xf"
value {
int64_list {
value: 6
}
}
}
feature {
key: "tips_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "trip_miles_xf"
value {
float_list {
value: 1.6439942121505737
}
}
}
feature {
key: "trip_seconds_xf"
value {
float_list {
value: 0.7374526262283325
}
}
}
feature {
key: "trip_start_day_xf"
value {
int64_list {
value: 3
}
}
}
feature {
key: "trip_start_hour_xf"
value {
int64_list {
value: 2
}
}
}
feature {
key: "trip_start_month_xf"
value {
int64_list {
value: 10
}
}
}
}
features {
feature {
key: "company_xf"
value {
int64_list {
value: 1
}
}
}
feature {
key: "dropoff_census_tract_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "dropoff_community_area_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "dropoff_latitude_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "dropoff_longitude_xf"
value {
int64_list {
value: 12
}
}
}
feature {
key: "fare_xf"
value {
float_list {
value: 0.4919655919075012
}
}
}
feature {
key: "payment_type_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "pickup_census_tract_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "pickup_community_area_xf"
value {
int64_list {
value: 14
}
}
}
feature {
key: "pickup_latitude_xf"
value {
int64_list {
value: 8
}
}
}
feature {
key: "pickup_longitude_xf"
value {
int64_list {
value: 1
}
}
}
feature {
key: "tips_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "trip_miles_xf"
value {
float_list {
value: -0.44967928528785706
}
}
}
feature {
key: "trip_seconds_xf"
value {
float_list {
value: 0.37378188967704773
}
}
}
feature {
key: "trip_start_day_xf"
value {
int64_list {
value: 5
}
}
}
feature {
key: "trip_start_hour_xf"
value {
int64_list {
value: 7
}
}
}
feature {
key: "trip_start_month_xf"
value {
int64_list {
value: 5
}
}
}
}
features {
feature {
key: "company_xf"
value {
int64_list {
value: 56
}
}
}
feature {
key: "dropoff_census_tract_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "dropoff_community_area_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "dropoff_latitude_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "dropoff_longitude_xf"
value {
int64_list {
value: 12
}
}
}
feature {
key: "fare_xf"
value {
float_list {
value: 0.4722445011138916
}
}
}
feature {
key: "payment_type_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "pickup_census_tract_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "pickup_community_area_xf"
value {
int64_list {
value: 13
}
}
}
feature {
key: "pickup_latitude_xf"
value {
int64_list {
value: 9
}
}
}
feature {
key: "pickup_longitude_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "tips_xf"
value {
int64_list {
value: 0
}
}
}
feature {
key: "trip_miles_xf"
value {
float_list {
value: 0.6968561410903931
}
}
}
feature {
key: "trip_seconds_xf"
value {
float_list {
value: 0.010111115872859955
}
}
}
feature {
key: "trip_start_day_xf"
value {
int64_list {
value: 3
}
}
}
feature {
key: "trip_start_hour_xf"
value {
int64_list {
value: 12
}
}
}
feature {
key: "trip_start_month_xf"
value {
int64_list {
value: 11
}
}
}
}
After the Transform component has transformed your data into features, and the next step is to train a model.
Trainer
The Trainer component will train a model that you define in TensorFlow. Default Trainer support Estimator API, to use Keras API, you need to specify Generic Trainer by setup custom_executor_spec=executor_spec.ExecutorClassSpec(GenericExecutor) in Trainer's contructor.
Trainer takes as input the schema from SchemaGen, the transformed data and graph from Transform, training parameters, as well as a module that contains user-defined model code.
Let's see an example of user-defined model code below (for an introduction to the TensorFlow Keras APIs, see the tutorial):
_taxi_trainer_module_file = 'taxi_trainer.py'
%%skip_for_export
%%writefile {_taxi_trainer_module_file}
from typing import List, Text
import os
import absl
import datetime
import tensorflow as tf
import tensorflow_transform as tft
from tfx.components.trainer.executor import TrainerFnArgs
import taxi_constants
_DENSE_FLOAT_FEATURE_KEYS = taxi_constants.DENSE_FLOAT_FEATURE_KEYS
_VOCAB_FEATURE_KEYS = taxi_constants.VOCAB_FEATURE_KEYS
_VOCAB_SIZE = taxi_constants.VOCAB_SIZE
_OOV_SIZE = taxi_constants.OOV_SIZE
_FEATURE_BUCKET_COUNT = taxi_constants.FEATURE_BUCKET_COUNT
_BUCKET_FEATURE_KEYS = taxi_constants.BUCKET_FEATURE_KEYS
_CATEGORICAL_FEATURE_KEYS = taxi_constants.CATEGORICAL_FEATURE_KEYS
_MAX_CATEGORICAL_FEATURE_VALUES = taxi_constants.MAX_CATEGORICAL_FEATURE_VALUES
_LABEL_KEY = taxi_constants.LABEL_KEY
_transformed_name = taxi_constants.transformed_name
def _transformed_names(keys):
return [_transformed_name(key) for key in keys]
def _gzip_reader_fn(filenames):
"""Small utility returning a record reader that can read gzip'ed files."""
return tf.data.TFRecordDataset(
filenames,
compression_type='GZIP')
def _get_serve_tf_examples_fn(model, tf_transform_output):
"""Returns a function that parses a serialized tf.Example and applies TFT."""
model.tft_layer = tf_transform_output.transform_features_layer()
@tf.function
def serve_tf_examples_fn(serialized_tf_examples):
"""Returns the output to be used in the serving signature."""
feature_spec = tf_transform_output.raw_feature_spec()
feature_spec.pop(_LABEL_KEY)
parsed_features = tf.io.parse_example(serialized_tf_examples, feature_spec)
transformed_features = model.tft_layer(parsed_features)
transformed_features.pop(_transformed_name(_LABEL_KEY))
return model(transformed_features)
return serve_tf_examples_fn
def _input_fn(file_pattern: Text,
tf_transform_output: tft.TFTransformOutput,
batch_size: int = 200) -> tf.data.Dataset:
"""Generates features and label for tuning/training.
Args:
file_pattern: input tfrecord file pattern.
tf_transform_output: A TFTransformOutput.
batch_size: representing the number of consecutive elements of returned
dataset to combine in a single batch
Returns:
A dataset that contains (features, indices) tuple where features is a
dictionary of Tensors, and indices is a single Tensor of label indices.
"""
transformed_feature_spec = (
tf_transform_output.transformed_feature_spec().copy())
dataset = tf.data.experimental.make_batched_features_dataset(
file_pattern=file_pattern,
batch_size=batch_size,
features=transformed_feature_spec,
reader=_gzip_reader_fn,
label_key=_transformed_name(_LABEL_KEY))
return dataset
def _build_keras_model(hidden_units: List[int] = None) -> tf.keras.Model:
"""Creates a DNN Keras model for classifying taxi data.
Args:
hidden_units: [int], the layer sizes of the DNN (input layer first).
Returns:
A keras Model.
"""
real_valued_columns = [
tf.feature_column.numeric_column(key, shape=())
for key in _transformed_names(_DENSE_FLOAT_FEATURE_KEYS)
]
categorical_columns = [
tf.feature_column.categorical_column_with_identity(
key, num_buckets=_VOCAB_SIZE + _OOV_SIZE, default_value=0)
for key in _transformed_names(_VOCAB_FEATURE_KEYS)
]
categorical_columns += [
tf.feature_column.categorical_column_with_identity(
key, num_buckets=_FEATURE_BUCKET_COUNT, default_value=0)
for key in _transformed_names(_BUCKET_FEATURE_KEYS)
]
categorical_columns += [
tf.feature_column.categorical_column_with_identity( # pylint: disable=g-complex-comprehension
key,
num_buckets=num_buckets,
default_value=0) for key, num_buckets in zip(
_transformed_names(_CATEGORICAL_FEATURE_KEYS),
_MAX_CATEGORICAL_FEATURE_VALUES)
]
indicator_column = [
tf.feature_column.indicator_column(categorical_column)
for categorical_column in categorical_columns
]
model = _wide_and_deep_classifier(
# TODO(b/139668410) replace with premade wide_and_deep keras model
wide_columns=indicator_column,
deep_columns=real_valued_columns,
dnn_hidden_units=hidden_units or [100, 70, 50, 25])
return model
def _wide_and_deep_classifier(wide_columns, deep_columns, dnn_hidden_units):
"""Build a simple keras wide and deep model.
Args:
wide_columns: Feature columns wrapped in indicator_column for wide (linear)
part of the model.
deep_columns: Feature columns for deep part of the model.
dnn_hidden_units: [int], the layer sizes of the hidden DNN.
Returns:
A Wide and Deep Keras model
"""
# Following values are hard coded for simplicity in this example,
# However prefarably they should be passsed in as hparams.
# Keras needs the feature definitions at compile time.
# TODO(b/139081439): Automate generation of input layers from FeatureColumn.
input_layers = {
colname: tf.keras.layers.Input(name=colname, shape=(), dtype=tf.float32)
for colname in _transformed_names(_DENSE_FLOAT_FEATURE_KEYS)
}
input_layers.update({
colname: tf.keras.layers.Input(name=colname, shape=(), dtype='int32')
for colname in _transformed_names(_VOCAB_FEATURE_KEYS)
})
input_layers.update({
colname: tf.keras.layers.Input(name=colname, shape=(), dtype='int32')
for colname in _transformed_names(_BUCKET_FEATURE_KEYS)
})
input_layers.update({
colname: tf.keras.layers.Input(name=colname, shape=(), dtype='int32')
for colname in _transformed_names(_CATEGORICAL_FEATURE_KEYS)
})
# TODO(b/144500510): SparseFeatures for feature columns + Keras.
deep = tf.keras.layers.DenseFeatures(deep_columns)(input_layers)
for numnodes in dnn_hidden_units:
deep = tf.keras.layers.Dense(numnodes)(deep)
wide = tf.keras.layers.DenseFeatures(wide_columns)(input_layers)
output = tf.keras.layers.Dense(
1, activation='sigmoid')(
tf.keras.layers.concatenate([deep, wide]))
model = tf.keras.Model(input_layers, output)
model.compile(
loss='binary_crossentropy',
optimizer=tf.keras.optimizers.Adam(lr=0.001),
metrics=[tf.keras.metrics.BinaryAccuracy()])
model.summary(print_fn=absl.logging.info)
return model
# TFX Trainer will call this function.
def run_fn(fn_args: TrainerFnArgs):
"""Train the model based on given args.
Args:
fn_args: Holds args used to train the model as name/value pairs.
"""
# Number of nodes in the first layer of the DNN
first_dnn_layer_size = 100
num_dnn_layers = 4
dnn_decay_factor = 0.7
tf_transform_output = tft.TFTransformOutput(fn_args.transform_output)
train_dataset = _input_fn(fn_args.train_files, tf_transform_output, 40)
eval_dataset = _input_fn(fn_args.eval_files, tf_transform_output, 40)
model = _build_keras_model(
# Construct layers sizes with exponetial decay
hidden_units=[
max(2, int(first_dnn_layer_size * dnn_decay_factor**i))
for i in range(num_dnn_layers)
])
# This log path might change in the future.
log_dir = os.path.join(os.path.dirname(fn_args.serving_model_dir), 'logs')
tensorboard_callback = tf.keras.callbacks.TensorBoard(
log_dir=log_dir, update_freq='batch')
model.fit(
train_dataset,
steps_per_epoch=fn_args.train_steps,
validation_data=eval_dataset,
validation_steps=fn_args.eval_steps,
callbacks=[tensorboard_callback])
signatures = {
'serving_default':
_get_serve_tf_examples_fn(model,
tf_transform_output).get_concrete_function(
tf.TensorSpec(
shape=[None],
dtype=tf.string,
name='examples')),
}
model.save(fn_args.serving_model_dir, save_format='tf', signatures=signatures)
Writing taxi_trainer.py This cell will be skipped during export to pipeline.
Now, we pass in this model code to the Trainer component and run it to train the model.
trainer = Trainer(
module_file=os.path.abspath(_taxi_trainer_module_file),
custom_executor_spec=executor_spec.ExecutorClassSpec(GenericExecutor),
examples=transform.outputs['transformed_examples'],
transform_graph=transform.outputs['transform_graph'],
schema=schema_gen.outputs['schema'],
train_args=trainer_pb2.TrainArgs(num_steps=10000),
eval_args=trainer_pb2.EvalArgs(num_steps=5000))
context.run(trainer)
INFO:absl:Running driver for Trainer INFO:absl:MetadataStore with DB connection initialized INFO:absl:Running executor for Trainer INFO:absl:Training model. WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.6/site-packages/tensorflow_core/python/feature_column/feature_column_v2.py:4267: IndicatorColumn._variable_shape (from tensorflow.python.feature_column.feature_column_v2) is deprecated and will be removed in a future version. Instructions for updating: The old _FeatureColumn APIs are being deprecated. Please use the new FeatureColumn APIs instead. WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.6/site-packages/tensorflow_core/python/feature_column/feature_column_v2.py:4322: IdentityCategoricalColumn._num_buckets (from tensorflow.python.feature_column.feature_column_v2) is deprecated and will be removed in a future version. Instructions for updating: The old _FeatureColumn APIs are being deprecated. Please use the new FeatureColumn APIs instead. INFO:absl:Model: "model" INFO:absl:__________________________________________________________________________________________________ INFO:absl:Layer (type) Output Shape Param # Connected to INFO:absl:================================================================================================== INFO:absl:company_xf (InputLayer) [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:dropoff_census_tract_xf (InputL [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:dropoff_community_area_xf (Inpu [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:dropoff_latitude_xf (InputLayer [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:dropoff_longitude_xf (InputLaye [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:fare_xf (InputLayer) [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:payment_type_xf (InputLayer) [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:pickup_census_tract_xf (InputLa [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:pickup_community_area_xf (Input [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:pickup_latitude_xf (InputLayer) [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:pickup_longitude_xf (InputLayer [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:trip_miles_xf (InputLayer) [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:trip_seconds_xf (InputLayer) [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:trip_start_day_xf (InputLayer) [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:trip_start_hour_xf (InputLayer) [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:trip_start_month_xf (InputLayer [(None,)] 0 INFO:absl:__________________________________________________________________________________________________ INFO:absl:dense_features (DenseFeatures) (None, 3) 0 company_xf[0][0] INFO:absl: dropoff_census_tract_xf[0][0] INFO:absl: dropoff_community_area_xf[0][0] INFO:absl: dropoff_latitude_xf[0][0] INFO:absl: dropoff_longitude_xf[0][0] INFO:absl: fare_xf[0][0] INFO:absl: payment_type_xf[0][0] INFO:absl: pickup_census_tract_xf[0][0] INFO:absl: pickup_community_area_xf[0][0] INFO:absl: pickup_latitude_xf[0][0] INFO:absl: pickup_longitude_xf[0][0] INFO:absl: trip_miles_xf[0][0] INFO:absl: trip_seconds_xf[0][0] INFO:absl: trip_start_day_xf[0][0] INFO:absl: trip_start_hour_xf[0][0] INFO:absl: trip_start_month_xf[0][0] INFO:absl:__________________________________________________________________________________________________ INFO:absl:dense (Dense) (None, 100) 400 dense_features[0][0] INFO:absl:__________________________________________________________________________________________________ INFO:absl:dense_1 (Dense) (None, 70) 7070 dense[0][0] INFO:absl:__________________________________________________________________________________________________ INFO:absl:dense_2 (Dense) (None, 48) 3408 dense_1[0][0] INFO:absl:__________________________________________________________________________________________________ INFO:absl:dense_3 (Dense) (None, 34) 1666 dense_2[0][0] INFO:absl:__________________________________________________________________________________________________ INFO:absl:dense_features_1 (DenseFeatures (None, 2127) 0 company_xf[0][0] INFO:absl: dropoff_census_tract_xf[0][0] INFO:absl: dropoff_community_area_xf[0][0] INFO:absl: dropoff_latitude_xf[0][0] INFO:absl: dropoff_longitude_xf[0][0] INFO:absl: fare_xf[0][0] INFO:absl: payment_type_xf[0][0] INFO:absl: pickup_census_tract_xf[0][0] INFO:absl: pickup_community_area_xf[0][0] INFO:absl: pickup_latitude_xf[0][0] INFO:absl: pickup_longitude_xf[0][0] INFO:absl: trip_miles_xf[0][0] INFO:absl: trip_seconds_xf[0][0] INFO:absl: trip_start_day_xf[0][0] INFO:absl: trip_start_hour_xf[0][0] INFO:absl: trip_start_month_xf[0][0] INFO:absl:__________________________________________________________________________________________________ INFO:absl:concatenate (Concatenate) (None, 2161) 0 dense_3[0][0] INFO:absl: dense_features_1[0][0] INFO:absl:__________________________________________________________________________________________________ INFO:absl:dense_4 (Dense) (None, 1) 2162 concatenate[0][0] INFO:absl:================================================================================================== INFO:absl:Total params: 14,706 INFO:absl:Trainable params: 14,706 INFO:absl:Non-trainable params: 0 INFO:absl:__________________________________________________________________________________________________ Train for 10000 steps, validate for 5000 steps 10000/10000 [==============================] - 101s 10ms/step - loss: 0.2449 - binary_accuracy: 0.8770 - val_loss: 0.2138 - val_binary_accuracy: 0.8961 WARNING:tensorflow:AutoGraph could not transform <function _load_transform_saved_model at 0x7f2641711d90> and will run it as-is. Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: LIVE_VARS_IN WARNING: AutoGraph could not transform <function _load_transform_saved_model at 0x7f2641711d90> and will run it as-is. Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: LIVE_VARS_IN WARNING:tensorflow:Expected binary or unicode string, got type_url: "type.googleapis.com/tensorflow.AssetFileDef" value: "\n\013\n\tConst_7:0\022-vocab_compute_and_apply_vocabulary_vocabulary" WARNING:tensorflow:Expected binary or unicode string, got type_url: "type.googleapis.com/tensorflow.AssetFileDef" value: "\n\013\n\tConst_9:0\022/vocab_compute_and_apply_vocabulary_1_vocabulary" INFO:tensorflow:Saver not created because there are no variables in the graph to restore WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.6/site-packages/tensorflow_core/python/ops/resource_variable_ops.py:1786: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version. Instructions for updating: If using Keras pass *_constraint arguments to layers. INFO:tensorflow:Assets written to: /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Trainer/model/6/serving_model_dir/assets INFO:absl:Training complete. Model written to /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Trainer/model/6/serving_model_dir INFO:absl:Running publisher for Trainer INFO:absl:MetadataStore with DB connection initialized
Analyze Training with TensorBoard
Take a peek at the trainer artifact. It points to a directory containing the model subdirectories.
model_artifact_dir = trainer.outputs['model'].get()[0].uri
pp.pprint(os.listdir(model_artifact_dir))
model_dir = os.path.join(model_artifact_dir, 'serving_model_dir')
pp.pprint(os.listdir(model_dir))
['logs', 'serving_model_dir'] ['variables', 'assets', 'saved_model.pb']
Optionally, we can connect TensorBoard to the Trainer to analyze our model's training curves.
%%skip_for_export
log_dir = os.path.join(model_artifact_dir, 'logs')
%load_ext tensorboard
%tensorboard --logdir {log_dir}
This cell will be skipped during export to pipeline.
Evaluator
The Evaluator component computes model performance metrics over the evaluation set. It uses the TensorFlow Model Analysis library. The Evaluator can also optionally validate that a newly trained model is better than the previous model. This is useful in a production pipeline setting where you may automatically train and validate a model every day. In this notebook, we only train one model, so the Evaluator automatically will label the model as "good".
Evaluator will take as input the data from ExampleGen, the trained model from Trainer, and slicing configuration. The slicing configuration allows you to slice your metrics on feature values (e.g. how does your model perform on taxi trips that start at 8am versus 8pm?). See an example of this configuration below:
eval_config = tfma.EvalConfig(
model_specs=[
# This assumes a serving model with signature 'serving_default'. If
# using estimator based EvalSavedModel, add signature_name: 'eval' and
# remove the label_key.
tfma.ModelSpec(label_key='tips')
],
metrics_specs=[
tfma.MetricsSpec(
# The metrics added here are in addition to those saved with the
# model (assuming either a keras model or EvalSavedModel is used).
# Any metrics added into the saved model (for example using
# model.compile(..., metrics=[...]), etc) will be computed
# automatically.
metrics=[
tfma.MetricConfig(class_name='ExampleCount')
],
# To add validation thresholds for metrics saved with the model,
# add them keyed by metric name to the thresholds map.
thresholds = {
'binary_accuracy': tfma.MetricThreshold(
value_threshold=tfma.GenericValueThreshold(
lower_bound={'value': 0.5}),
change_threshold=tfma.GenericChangeThreshold(
direction=tfma.MetricDirection.HIGHER_IS_BETTER,
absolute={'value': -1e-10}))
}
)
],
slicing_specs=[
# An empty slice spec means the overall slice, i.e. the whole dataset.
tfma.SlicingSpec(),
# Data can be sliced along a feature column. In this case, data is
# sliced along feature column trip_start_hour.
tfma.SlicingSpec(feature_keys=['trip_start_hour'])
])
Next, we give this configuration to Evaluator and run it.
# Use TFMA to compute a evaluation statistics over features of a model and
# validate them against a baseline.
# The model resolver is only required if performing model validation in addition
# to evaluation. In this case we validate against the latest blessed model. If
# no model has been blessed before (as in this case) the evaluator will make our
# candidate the first blessed model.
model_resolver = ResolverNode(
instance_name='latest_blessed_model_resolver',
resolver_class=latest_blessed_model_resolver.LatestBlessedModelResolver,
model=Channel(type=Model),
model_blessing=Channel(type=ModelBlessing))
context.run(model_resolver)
evaluator = Evaluator(
examples=example_gen.outputs['examples'],
model=trainer.outputs['model'],
baseline_model=model_resolver.outputs['model'],
# Change threshold will be ignored if there is no baseline (first run).
eval_config=eval_config)
context.run(evaluator)
INFO:absl:Running driver for ResolverNode.latest_blessed_model_resolver
INFO:absl:MetadataStore with DB connection initialized
INFO:absl:Running publisher for ResolverNode.latest_blessed_model_resolver
INFO:absl:MetadataStore with DB connection initialized
INFO:absl:Running driver for Evaluator
INFO:absl:MetadataStore with DB connection initialized
INFO:absl:Running executor for Evaluator
INFO:absl:Request was made to ignore the baseline ModelSpec and any change thresholds. This is likely because a baseline model was not provided: updated_config=
model_specs {
label_key: "tips"
}
slicing_specs {
}
slicing_specs {
feature_keys: "trip_start_hour"
}
metrics_specs {
metrics {
class_name: "ExampleCount"
threshold {
}
}
thresholds {
key: "binary_accuracy"
value {
value_threshold {
lower_bound {
value: 0.5
}
}
}
}
}
INFO:absl:Using /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Trainer/model/6/serving_model_dir as model.
INFO:absl:Evaluating model.
INFO:absl:Using 1 process(es) for Beam pipeline execution.
WARNING:absl:inputs do not match those expected by the model: input_names=['examples'], found in extracts={}
INFO:absl:Evaluation complete. Results written to /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Evaluator/evaluation/8.
INFO:absl:Checking validation results.
INFO:absl:Blessing result True written to /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Evaluator/blessing/8.
INFO:absl:Running publisher for Evaluator
INFO:absl:MetadataStore with DB connection initialized
Now let's examine the output artifacts of Evaluator.
%%skip_for_export
evaluator.outputs
{'evaluation': Channel(
type_name: ModelEvaluation
artifacts: [Artifact(type_name: ModelEvaluation, uri: /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Evaluator/evaluation/8, id: 9)]
), 'blessing': Channel(
type_name: ModelBlessing
artifacts: [Artifact(type_name: ModelBlessing, uri: /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Evaluator/blessing/8, id: 10)]
)}
This cell will be skipped during export to pipeline.
Using the evaluation output we can show the default visualization of global metrics on the entire evaluation set.
%%skip_for_export
context.show(evaluator.outputs['evaluation'])
This cell will be skipped during export to pipeline.
To see the visualization for sliced evaluation metrics, we can directly call the TensorFlow Model Analysis library.
%%skip_for_export
import tensorflow_model_analysis as tfma
# Get the TFMA output result path and load the result.
PATH_TO_RESULT = evaluator.outputs['evaluation'].get()[0].uri
tfma_result = tfma.load_eval_result(PATH_TO_RESULT)
# Show data sliced along feature column trip_start_hour.
tfma.view.render_slicing_metrics(
tfma_result, slicing_column='trip_start_hour')
SlicingMetricsViewer(config={'weightedExamplesColumn': 'example_count'}, data=[{'slice': 'trip_start_hour:1', …
This cell will be skipped during export to pipeline.
This visualization shows the same metrics, but computed at every feature value of trip_start_hour instead of on the entire evaluation set.
TensorFlow Model Analysis supports many other visualizations, such as Fairness Indicators and plotting a time series of model performance. To learn more, see the tutorial.
Since we added thresholds to our config, validation output is also available. The precence of a blessing artifact indicates that our model passed validation. Since this is the first validation being performed the candidate is automatically blessed.
%%skip_for_export
blessing_uri = evaluator.outputs.blessing.get()[0].uri
!ls -l {blessing_uri}
total 0 -rw-rw-r-- 1 kbuilder kbuilder 0 Apr 8 09:15 BLESSED This cell will be skipped during export to pipeline.
Now can also verify the success by loading the validation result record:
%%skip_for_export
PATH_TO_RESULT = evaluator.outputs['evaluation'].get()[0].uri
print(tfma.load_validation_result(PATH_TO_RESULT))
validation_ok: true This cell will be skipped during export to pipeline.
Pusher
The Pusher component is usually at the end of a TFX pipeline. It checks whether a model has passed validation, and if so, exports the model to _serving_model_dir.
pusher = Pusher(
model=trainer.outputs['model'],
model_blessing=evaluator.outputs['blessing'],
push_destination=pusher_pb2.PushDestination(
filesystem=pusher_pb2.PushDestination.Filesystem(
base_directory=_serving_model_dir)))
context.run(pusher)
INFO:absl:Running driver for Pusher INFO:absl:MetadataStore with DB connection initialized INFO:absl:Running executor for Pusher INFO:absl:Model pushing. INFO:absl:Model version is 1586337359 INFO:absl:Model written to /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Pusher/pushed_model/9. INFO:absl:Model written to serving path /tmp/tmpd_6vlr21/serving_model/taxi_simple/1586337359. INFO:absl:Model pushed to /tmp/tmpd_6vlr21/serving_model/taxi_simple/1586337359. INFO:absl:Running publisher for Pusher INFO:absl:MetadataStore with DB connection initialized
Let's examine the output artifacts of Pusher.
%%skip_for_export
pusher.outputs
{'pushed_model': Channel(
type_name: PushedModel
artifacts: [Artifact(type_name: PushedModel, uri: /tmp/tfx-interactive-2020-04-08T09_13_36.436626-3xb2v2nu/Pusher/pushed_model/9, id: 11)]
)}
This cell will be skipped during export to pipeline.
In particular, the Pusher will export your model in the SavedModel format, which looks like this:
%%skip_for_export
push_uri = pusher.outputs.model_push.get()[0].uri
latest_version = max(os.listdir(push_uri))
latest_version_path = os.path.join(push_uri, latest_version)
model = tf.saved_model.load(latest_version_path)
for item in model.signatures.items():
pp.pprint(item)
('serving_default',
<tensorflow.python.saved_model.load._WrapperFunction object at 0x7f263831ba90>)
This cell will be skipped during export to pipeline.
We're finished our tour of built-in TFX components!
After you're happy with experimenting with TFX components and code in this notebook, you may want to export it as a pipeline to be orchestrated with Apache Airflow or Apache Beam. See the final section.
Export to pipeline
To export the contents of this notebook as a pipeline to be orchestrated with Airflow or Beam, follow the instructions below.
If you're using Colab, make sure to save this notebook to Google Drive (File → Save a Copy in Drive) before exporting.
1. Mount Google Drive (Colab-only)
If you're using Colab, this notebook needs to mount your Google Drive to be able to access its own .ipynb file.
%%skip_for_export
#@markdown Run this cell and enter the authorization code to mount Google Drive.
import sys
if 'google.colab' in sys.modules:
# Colab.
from google.colab import drive
drive.mount('/content/drive')
2. Select an orchestrator
_runner_type = 'beam' #@param ["beam", "airflow"]
_pipeline_name = 'chicago_taxi_%s' % _runner_type
3. Set up paths for the pipeline
# For Colab notebooks only.
# TODO(USER): Fill out the path to this notebook.
_notebook_filepath = (
'/content/drive/My Drive/Colab Notebooks/taxi_pipeline_interactive.ipynb')
# For Jupyter notebooks only.
# _notebook_filepath = os.path.join(os.getcwd(),
# 'taxi_pipeline_interactive.ipynb')
# TODO(USER): Fill out the paths for the exported pipeline.
_tfx_root = os.path.join(os.environ['HOME'], 'tfx')
_taxi_root = os.path.join(os.environ['HOME'], 'taxi')
_serving_model_dir = os.path.join(_taxi_root, 'serving_model')
_data_root = os.path.join(_taxi_root, 'data', 'simple')
_pipeline_root = os.path.join(_tfx_root, 'pipelines', _pipeline_name)
_metadata_path = os.path.join(_tfx_root, 'metadata', _pipeline_name,
'metadata.db')
4. Choose components to include in the pipeline
# TODO(USER): Specify components to be included in the exported pipeline.
components = [
example_gen, statistics_gen, schema_gen, example_validator, transform,
trainer, evaluator, pusher
]
5. Generate pipeline files
%%skip_for_export
#@markdown Run this cell to generate the pipeline files.
if get_ipython().magics_manager.auto_magic:
print('Warning: %automagic is ON. Line magics specified without the % prefix '
'will not be scrubbed during export to pipeline.')
_pipeline_export_filepath = 'export_%s.py' % _pipeline_name
context.export_to_pipeline(notebook_filepath=_notebook_filepath,
export_filepath=_pipeline_export_filepath,
runner_type=_runner_type)
6. Download pipeline files
%%skip_for_export
#@markdown Run this cell to download the pipeline files as a `.zip`.
if 'google.colab' in sys.modules:
from google.colab import files
import zipfile
zip_export_path = os.path.join(
tempfile.mkdtemp(), 'export.zip')
with zipfile.ZipFile(zip_export_path, mode='w') as export_zip:
export_zip.write(_pipeline_export_filepath)
export_zip.write(_taxi_constants_module_file)
export_zip.write(_taxi_transform_module_file)
export_zip.write(_taxi_trainer_module_file)
files.download(zip_export_path)
To learn how to run the orchestrated pipeline with Apache Airflow, please refer to the TFX Orchestration Tutorial.