2d. Distributed training and monitoring

In this notebook, we refactor to call train_and_evaluate instead of hand-coding our ML pipeline. This allows us to carry out evaluation as part of our training loop instead of as a separate step. It also adds in failure-handling that is necessary for distributed training capabilities.

We also use TensorBoard to monitor the training.



In [ ]:

    
import tensorflow as tf
import numpy as np
import shutil
print(tf.__version__)

Input

Read data created in Lab1a, but this time make it more general, so that we are reading in batches. Instead of using Pandas, we will use Datasets.



In [ ]:

    
CSV_COLUMNS = ['fare_amount', 'pickuplon','pickuplat','dropofflon','dropofflat','passengers', 'key']
LABEL_COLUMN = 'fare_amount'
DEFAULTS = [[0.0], [-74.0], [40.0], [-74.0], [40.7], [1.0], ['nokey']]

def read_dataset(filename, mode, batch_size = 512):
  def _input_fn():
    def decode_csv(value_column):
      columns = tf.decode_csv(value_column, record_defaults = DEFAULTS)
      features = dict(zip(CSV_COLUMNS, columns))
      label = features.pop(LABEL_COLUMN)
      return features, label
    
    # Create list of files that match pattern
    file_list = tf.gfile.Glob(filename)

    # Create dataset from file list
    dataset = tf.data.TextLineDataset(file_list).map(decode_csv)
    
    if mode == tf.estimator.ModeKeys.TRAIN:
        num_epochs = None # indefinitely
        dataset = dataset.shuffle(buffer_size = 10 * batch_size)
    else:
        num_epochs = 1 # end-of-input after this
 
    dataset = dataset.repeat(num_epochs).batch(batch_size)
    return dataset.make_one_shot_iterator().get_next()
  return _input_fn

Create features out of input data

For now, pass these through. (same as previous lab)



In [ ]:

    
INPUT_COLUMNS = [
    tf.feature_column.numeric_column('pickuplon'),
    tf.feature_column.numeric_column('pickuplat'),
    tf.feature_column.numeric_column('dropofflat'),
    tf.feature_column.numeric_column('dropofflon'),
    tf.feature_column.numeric_column('passengers'),
]

def add_more_features(feats):
  # Nothing to add (yet!)
  return feats

feature_cols = add_more_features(INPUT_COLUMNS)

train_and_evaluate



In [ ]:

    
def serving_input_fn():
  feature_placeholders = {
    'pickuplon' : tf.placeholder(tf.float32, [None]),
    'pickuplat' : tf.placeholder(tf.float32, [None]),
    'dropofflat' : tf.placeholder(tf.float32, [None]),
    'dropofflon' : tf.placeholder(tf.float32, [None]),
    'passengers' : tf.placeholder(tf.float32, [None]),
  }
  features = {
      key: tf.expand_dims(tensor, -1)
      for key, tensor in feature_placeholders.items()
  }
  return tf.estimator.export.ServingInputReceiver(features, feature_placeholders)



In [ ]:

    
def train_and_evaluate(output_dir, num_train_steps):
  estimator = tf.estimator.LinearRegressor(
                       model_dir = output_dir,
                       feature_columns = feature_cols)
  train_spec=tf.estimator.TrainSpec(
                       input_fn = read_dataset('./taxi-train.csv', mode = tf.estimator.ModeKeys.TRAIN),
                       max_steps = num_train_steps)
  exporter = tf.estimator.LatestExporter('exporter', serving_input_fn)
  eval_spec=tf.estimator.EvalSpec(
                       input_fn = read_dataset('./taxi-valid.csv', mode = tf.estimator.ModeKeys.EVAL),
                       steps = None,
                       start_delay_secs = 1, # start evaluating after N seconds
                       throttle_secs = 10,  # evaluate every N seconds
                       exporters = exporter)
  tf.estimator.train_and_evaluate(estimator, train_spec, eval_spec)



In [ ]:

    
# Run training    
OUTDIR = 'taxi_trained'
shutil.rmtree(OUTDIR, ignore_errors = True) # start fresh each time
train_and_evaluate(OUTDIR, num_train_steps = 5000)

Monitor training with TensorBoard

To activate TensorBoard within the JupyterLab UI navigate to "File" - "New Launcher". Then double-click the 'Tensorboard' icon on the bottom row.

TensorBoard 1 will appear in the new tab. Navigate through the three tabs to see the active TensorBoard. The 'Graphs' and 'Projector' tabs offer very interesting information including the ability to replay the tests.

You may close the TensorBoard tab when you are finished exploring.

Copyright 2017 Google Inc. 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 http://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