Führen Sie TensorFlow Federated auf einer GPU mit Colab aus

Question

Gibt es eine Möglichkeit, die von Colab bereitgestellte GPU zu verwenden, um die Trainingseinheiten von TFF schneller durchzuführen? Das Training von Federated Models dauert mehr als 1 Stunde und es scheint, dass die Verwendung einer GPU-Laufzeit überhaupt keinen Nutzen bringt.

Die TFF-Seite von High-Performance Simulation ist noch leer und ich kann keine Anleitung zur Verwendung der GPU mit TFF finden.

Irgendein Vorschlag? Vielen Dank!

tf- und tff-versionen:

2.4.0-dev20200917 
0.16.1

Anzahl der Kunden in jeder Runde:

70

Eingabedatenelementspezifikation:

OrderedDict([('x',
          OrderedDict([('start_place',
                        TensorSpec(shape=(8, 8), dtype=tf.int32, name=None)),
                       ('start_hour_sin',
                        TensorSpec(shape=(8, 8), dtype=tf.float64, name=None)),
                       ('start_hour_cos',
                        TensorSpec(shape=(8, 8), dtype=tf.float64, name=None)),
                       ('week_day_sin',
                        TensorSpec(shape=(8, 8), dtype=tf.float64, name=None)),
                       ('week_day_cos',
                        TensorSpec(shape=(8, 8), dtype=tf.float64, name=None)),
                       ('weekend',
                        TensorSpec(shape=(8, 8), dtype=tf.int32, name=None)),
                       ('month',
                        TensorSpec(shape=(8, 8), dtype=tf.int32, name=None))])),
         ('y', TensorSpec(shape=(8, 8), dtype=tf.int32, name=None))])

Ähnlich wie im Tutorial zur Texterstellung, in dem ich mit einer Abfolge von Orten arbeite, ist das Modell ziemlich ähnlich:

    # Create a model
def create_keras_model(number_of_places, batch_size):
  
        # Shortcut to the layers package
  l = tf.keras.layers


  # Now we need to define an input dictionary.
    # Where the keys are the column names
    # This is a model with multiple inputs, so we need to declare and input layer for each feature
  feature_inputs = {
    'start_hour_sin': tf.keras.Input((N-1, ), batch_size=batch_size, name='start_hour_sin'),
    'start_hour_cos': tf.keras.Input((N-1, ), batch_size=batch_size, name='start_hour_cos'),
    'weekend': tf.keras.Input((N-1, ), batch_size=batch_size, name='weekend'),
    'week_day_sin': tf.keras.Input((N-1, ), batch_size=batch_size, name='week_day_sin'),
    'week_day_cos': tf.keras.Input((N-1, ), batch_size=batch_size, name='week_day_cos'),
  }

  
  # We cannot use anarray of features as always because we have sequences and we cannot match the shape otherwise
  # We have to do one by one
  start_hour_sin = feature_column.numeric_column("start_hour_sin", shape=(N-1))
  hour_sin_feature = l.DenseFeatures(start_hour_sin)(feature_inputs)

  start_hour_cos = feature_column.numeric_column("start_hour_cos", shape=(N-1))
  hour_cos_feature = l.DenseFeatures(start_hour_cos)(feature_inputs)

  weekend = feature_column.numeric_column("weekend", shape=(N-1))
  weekend_feature = l.DenseFeatures(weekend)(feature_inputs)
  
  week_day_sin = feature_column.numeric_column("week_day_sin", shape=(N-1))
  week_day_sin_feature = l.DenseFeatures(week_day_sin)(feature_inputs)

  week_day_cos = feature_column.numeric_column("week_day_cos", shape=(N-1))
  week_day_cos_feature = l.DenseFeatures(week_day_cos)(feature_inputs)

  
    # We have also to add a dimension to then concatenate
  hour_sin_feature = tf.expand_dims(hour_sin_feature, -1)
  hour_cos_feature = tf.expand_dims(hour_cos_feature, -1)
  weekend_feature = tf.expand_dims(weekend_feature, -1)
  week_day_sin_feature = tf.expand_dims(week_day_sin_feature, -1)
  week_day_cos_feature = tf.expand_dims(week_day_cos_feature, -1)

  # Declare the dictionary for the places sequence as before
  sequence_input = {
      'start_place': tf.keras.Input((N-1,), batch_size=batch_size, dtype=tf.dtypes.int32, name='start_place') # add batch_size=batch_size in case of stateful GRU
  }


  # Handling the categorical feature sequence using one-hot
  places_one_hot = feature_column.sequence_categorical_column_with_vocabulary_list(
      'start_place', [i for i in range(number_of_places)])
  
  # Embed the one-hot encoding
  places_embed = feature_column.embedding_column(places_one_hot, embedding_dim)


  # With an input sequence we can't use the DenseFeature layer, we need to use the SequenceFeatures
  sequence_features, sequence_length = tf.keras.experimental.SequenceFeatures(places_embed)(sequence_input)

  input_sequence = l.Concatenate(axis=2)([ sequence_features, hour_sin_feature, hour_cos_feature, weekend_feature, week_day_sin_feature, week_day_cos_feature])

  # Rnn
  recurrent = l.GRU(rnn_units,
                        batch_size=batch_size, #in case of stateful
                        return_sequences=True,
                        dropout=0.5,
                        stateful=True,
                        recurrent_initializer='glorot_uniform')(input_sequence)


    # Last layer with an output for each places
  dense_1 = layers.Dense(number_of_places)(recurrent)

    # Softmax output layer
  output = l.Softmax()(dense_1)
    
    # To return the Model, we need to define it's inputs and outputs
    # In out case, we need to list all the input layers we have defined 
  inputs = list(feature_inputs.values()) + list(sequence_input.values())

    # Return the Model
  return tf.keras.Model(inputs=inputs, outputs=output)

Funktion zum Erstellen des Modells:

def create_tff_model():
  # TFF uses an `input_spec` so it knows the types and shapes
  # that your model expects.
  input_spec = preprocessed_example_dataset.element_spec
  keras_model_clone = create_keras_model(number_of_places, batch_size=BATCH_SIZE)
  return tff.learning.from_keras_model(
      keras_model_clone,
      input_spec=input_spec,
      loss=tf.keras.losses.SparseCategoricalCrossentropy(),

Federated Avg

# This command builds all the TensorFlow graphs and serializes them: 
fed_avg = tff.learning.build_federated_averaging_process(
    model_fn=create_tff_model,
    client_optimizer_fn=lambda: tf.keras.optimizers.Adam(learning_rate=0.001),
    server_optimizer_fn=lambda: tf.keras.optimizers.Adam(learning_rate=0.06))
          metrics=[tf.keras.metrics.SparseCategoricalAccuracy()])

State init:

state = fed_avg.initialize()

Trainingsschleife:

NUM_ROUNDS = 10

  for round_num in range(1, NUM_ROUNDS + 1):
    print('Round {r}'.format(r=round_num))
    state, metrics = fed_avg.next(state, train_data)
    train_metrics = metrics['train']
    print('\tTrain: loss={l:.3f}, accuracy={a:.3f}'.format(l=train_metrics['loss'], a=train_metrics['sparse_categorical_accuracy']))

Answer 1

Zu beachten ist, dass dieses Modell 0 Clients * 13 SGD-Schritte pro Runde (fast 1.000) ausführt, obwohl eine Stunde immer noch sehr lang zu sein scheint. 70 Clients auf einem einzelnen Computer stoßen an die Grenzen der Simulation. Wenn die Anzahl viel höher wird, beginnen wir mit dem Remote Executor, die Einrichtung mehrerer Computer zu untersuchen.

Einige Dinge zu untersuchen:

• Ist die Simulations- E / A gebunden? Wie schnell kann die Python-Umgebung über ein einzelnes Client-Dataset iterieren? In TF for batch in dataset:und Zeit kann es hier nützlich sein, wie lange es dauert.
• Ist die Simulation berechnet gebunden? Beobachten Sie möglicherweise die CPU- und GPU-Auslastung. Wie lange dauert die Ausführung keras_model.fit()auf einem einzelnen Client-Dataset? Die TFF-Simulation macht ungefähr 70x pro Runde (einmal für jeden Kunden).