Penaksir

Lihat di TensorFlow.org Jalankan di Google Colab Lihat sumber di GitHub Unduh buku catatan

Dokumen ini memperkenalkan tf.estimator TensorFlow tingkat tinggi. Penaksir merangkum tindakan berikut:

  • Pelatihan
  • Evaluasi
  • Ramalan
  • Ekspor untuk disajikan

TensorFlow mengimplementasikan beberapa Estimator yang sudah jadi. Penaksir kustom masih didukung, tetapi terutama sebagai ukuran kompatibilitas mundur. Penaksir khusus tidak boleh digunakan untuk kode baru . Semua Penaksir—yang dibuat sebelumnya atau yang dibuat khusus—adalah kelas yang didasarkan pada kelas tf.estimator.Estimator .

Untuk contoh cepat, coba tutorial Estimator . Untuk ikhtisar desain API, periksa buku putih .

Mempersiapkan

pip install -U tensorflow_datasets
import tempfile
import os

import tensorflow as tf
import tensorflow_datasets as tfds

Keuntungan

Mirip dengan tf.keras.Model , estimator adalah abstraksi tingkat model. tf.estimator menyediakan beberapa kemampuan yang saat ini masih dalam pengembangan untuk tf.keras . Ini adalah:

  • Pelatihan berbasis server parameter
  • Integrasi TFX penuh

Kemampuan Penaksir

Penaksir memberikan manfaat sebagai berikut:

  • Anda dapat menjalankan model berbasis Estimator pada host lokal atau pada lingkungan multi-server terdistribusi tanpa mengubah model Anda. Selanjutnya, Anda dapat menjalankan model berbasis Estimator pada CPU, GPU, atau TPU tanpa mengode ulang model Anda.
  • Penaksir menyediakan loop pelatihan terdistribusi yang aman yang mengontrol bagaimana dan kapan:
    • Muat data
    • Menangani pengecualian
    • Buat file pos pemeriksaan dan pulihkan dari kegagalan
    • Simpan ringkasan untuk TensorBoard

Saat menulis aplikasi dengan Estimator, Anda harus memisahkan jalur input data dari model. Pemisahan ini menyederhanakan eksperimen dengan kumpulan data yang berbeda.

Menggunakan Pengukur yang sudah jadi

Penaksir yang dibuat sebelumnya memungkinkan Anda bekerja pada tingkat konseptual yang jauh lebih tinggi daripada API TensorFlow dasar. Anda tidak perlu lagi khawatir membuat grafik atau sesi komputasi karena Estimator menangani semua "pipa ledeng" untuk Anda. Selanjutnya, Estimator yang dibuat sebelumnya memungkinkan Anda bereksperimen dengan arsitektur model yang berbeda dengan hanya membuat sedikit perubahan kode. tf.estimator.DNNClassifier , misalnya, adalah kelas Estimator yang dibuat sebelumnya yang melatih model klasifikasi berdasarkan jaringan saraf feed-forward yang padat.

Program TensorFlow yang mengandalkan Estimator yang dibuat sebelumnya biasanya terdiri dari empat langkah berikut:

1. Tulis fungsi input

Misalnya, Anda dapat membuat satu fungsi untuk mengimpor set pelatihan dan fungsi lain untuk mengimpor set pengujian. Penaksir mengharapkan input mereka diformat sebagai sepasang objek:

  • Kamus di mana kuncinya adalah nama fitur dan nilainya adalah Tensor (atau SparseTensors) yang berisi data fitur yang sesuai
  • Tensor yang berisi satu atau lebih label

input_fn harus mengembalikan tf.data.Dataset yang menghasilkan pasangan dalam format itu.

Misalnya, kode berikut membuat tf.data.Dataset dari file train.csv dataset Titanic:

def train_input_fn():
  titanic_file = tf.keras.utils.get_file("train.csv", "https://storage.googleapis.com/tf-datasets/titanic/train.csv")
  titanic = tf.data.experimental.make_csv_dataset(
      titanic_file, batch_size=32,
      label_name="survived")
  titanic_batches = (
      titanic.cache().repeat().shuffle(500)
      .prefetch(tf.data.AUTOTUNE))
  return titanic_batches

input_fn dieksekusi dalam tf.Graph dan juga dapat secara langsung mengembalikan pasangan (features_dics, labels) yang berisi tensor grafik, tetapi ini rawan kesalahan di luar kasus sederhana seperti mengembalikan konstanta.

2. Tentukan kolom fitur.

Setiap tf.feature_column mengidentifikasi nama fitur, jenisnya, dan pra-pemrosesan input apa pun.

Misalnya, cuplikan berikut membuat tiga kolom fitur.

  • Yang pertama menggunakan fitur age secara langsung sebagai input titik-mengambang.
  • Yang kedua menggunakan fitur class sebagai input kategoris.
  • Yang ketiga menggunakan embark_town sebagai input kategoris, tetapi menggunakan hashing trick untuk menghindari kebutuhan untuk menghitung opsi, dan untuk mengatur jumlah opsi.

Untuk informasi lebih lanjut, lihat tutorial kolom fitur .

age = tf.feature_column.numeric_column('age')
cls = tf.feature_column.categorical_column_with_vocabulary_list('class', ['First', 'Second', 'Third']) 
embark = tf.feature_column.categorical_column_with_hash_bucket('embark_town', 32)

3. Buat Instansiasi Penaksir pra-dibuat yang relevan.

Misalnya, berikut adalah contoh instantiasi dari Estimator yang dibuat sebelumnya bernama LinearClassifier :

model_dir = tempfile.mkdtemp()
model = tf.estimator.LinearClassifier(
    model_dir=model_dir,
    feature_columns=[embark, cls, age],
    n_classes=2
)
INFO:tensorflow:Using default config.
INFO:tensorflow:Using config: {'_model_dir': '/tmp/tmpl24pp3cp', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_checkpoint_save_graph_def': True, '_service': None, '_cluster_spec': ClusterSpec({}), '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}

Untuk informasi lebih lanjut, Anda dapat membuka tutorial pengklasifikasi linier .

4. Memanggil metode pelatihan, evaluasi, atau inferensi.

Semua Pengukur menyediakan metode train , evaluate , dan predict .

model = model.train(input_fn=train_input_fn, steps=100)
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.7/site-packages/tensorflow/python/training/training_util.py:236: Variable.initialized_value (from tensorflow.python.ops.variables) is deprecated and will be removed in a future version.
Instructions for updating:
Use Variable.read_value. Variables in 2.X are initialized automatically both in eager and graph (inside tf.defun) contexts.
INFO:tensorflow:Calling model_fn.
/tmpfs/src/tf_docs_env/lib/python3.7/site-packages/keras/engine/base_layer_v1.py:1684: UserWarning: `layer.add_variable` is deprecated and will be removed in a future version. Please use `layer.add_weight` method instead.
  warnings.warn('`layer.add_variable` is deprecated and '
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.7/site-packages/keras/optimizer_v2/ftrl.py:147: calling Constant.__init__ (from tensorflow.python.ops.init_ops) with dtype is deprecated and will be removed in a future version.
Instructions for updating:
Call initializer instance with the dtype argument instead of passing it to the constructor
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 0...
INFO:tensorflow:Saving checkpoints for 0 into /tmp/tmpl24pp3cp/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 0...
INFO:tensorflow:loss = 0.6931472, step = 0
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 100...
INFO:tensorflow:Saving checkpoints for 100 into /tmp/tmpl24pp3cp/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 100...
INFO:tensorflow:Loss for final step: 0.6319582.
2021-09-22 20:49:10.453286: W tensorflow/core/kernels/data/cache_dataset_ops.cc:768] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset  will be discarded. This can happen if you have an input pipeline similar to `dataset.cache().take(k).repeat()`. You should use `dataset.take(k).cache().repeat()` instead.
result = model.evaluate(train_input_fn, steps=10)

for key, value in result.items():
  print(key, ":", value)
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Starting evaluation at 2021-09-22T20:49:11
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /tmp/tmpl24pp3cp/model.ckpt-100
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Evaluation [1/10]
INFO:tensorflow:Evaluation [2/10]
INFO:tensorflow:Evaluation [3/10]
INFO:tensorflow:Evaluation [4/10]
INFO:tensorflow:Evaluation [5/10]
INFO:tensorflow:Evaluation [6/10]
INFO:tensorflow:Evaluation [7/10]
INFO:tensorflow:Evaluation [8/10]
INFO:tensorflow:Evaluation [9/10]
INFO:tensorflow:Evaluation [10/10]
INFO:tensorflow:Inference Time : 0.74609s
INFO:tensorflow:Finished evaluation at 2021-09-22-20:49:12
INFO:tensorflow:Saving dict for global step 100: accuracy = 0.734375, accuracy_baseline = 0.640625, auc = 0.7373913, auc_precision_recall = 0.64306235, average_loss = 0.563341, global_step = 100, label/mean = 0.359375, loss = 0.563341, precision = 0.734375, prediction/mean = 0.3463129, recall = 0.40869564
INFO:tensorflow:Saving 'checkpoint_path' summary for global step 100: /tmp/tmpl24pp3cp/model.ckpt-100
accuracy : 0.734375
accuracy_baseline : 0.640625
auc : 0.7373913
auc_precision_recall : 0.64306235
average_loss : 0.563341
label/mean : 0.359375
loss : 0.563341
precision : 0.734375
prediction/mean : 0.3463129
recall : 0.40869564
global_step : 100
2021-09-22 20:49:12.168629: W tensorflow/core/kernels/data/cache_dataset_ops.cc:768] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset  will be discarded. This can happen if you have an input pipeline similar to `dataset.cache().take(k).repeat()`. You should use `dataset.take(k).cache().repeat()` instead.
for pred in model.predict(train_input_fn):
  for key, value in pred.items():
    print(key, ":", value)
  break
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /tmp/tmpl24pp3cp/model.ckpt-100
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
logits : [-1.5173098]
logistic : [0.17985801]
probabilities : [0.820142   0.17985801]
class_ids : [0]
classes : [b'0']
all_class_ids : [0 1]
all_classes : [b'0' b'1']
2021-09-22 20:49:13.076528: W tensorflow/core/kernels/data/cache_dataset_ops.cc:768] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset  will be discarded. This can happen if you have an input pipeline similar to `dataset.cache().take(k).repeat()`. You should use `dataset.take(k).cache().repeat()` instead.

Manfaat Pengukur yang sudah jadi

Penaksir pra-dibuat menyandikan praktik terbaik, memberikan manfaat berikut:

  • Praktik terbaik untuk menentukan di mana bagian yang berbeda dari grafik komputasi harus dijalankan, menerapkan strategi pada satu mesin atau pada sebuah cluster.
  • Praktik terbaik untuk penulisan acara (ringkasan) dan ringkasan yang berguna secara universal.

Jika Anda tidak menggunakan Estimator yang sudah jadi, Anda harus mengimplementasikan sendiri fitur-fitur sebelumnya.

Penaksir Kustom

Inti dari setiap Estimator—baik yang dibuat sebelumnya maupun yang dibuat khusus—adalah fungsi modelnya , model_fn , yang merupakan metode yang membuat grafik untuk pelatihan, evaluasi, dan prediksi. Saat Anda menggunakan Estimator yang dibuat sebelumnya, orang lain telah mengimplementasikan fungsi model. Saat mengandalkan Penaksir khusus, Anda harus menulis sendiri fungsi model.

Buat Penaksir dari model Keras

Anda dapat mengonversi model Keras yang ada ke Penaksir dengan tf.keras.estimator.model_to_estimator . Ini berguna jika Anda ingin memodernisasi kode model Anda, tetapi alur pelatihan Anda masih memerlukan Estimator.

Buat instance model Keras MobileNet V2 dan kompilasi model dengan pengoptimal, kehilangan, dan metrik untuk dilatih dengan:

keras_mobilenet_v2 = tf.keras.applications.MobileNetV2(
    input_shape=(160, 160, 3), include_top=False)
keras_mobilenet_v2.trainable = False

estimator_model = tf.keras.Sequential([
    keras_mobilenet_v2,
    tf.keras.layers.GlobalAveragePooling2D(),
    tf.keras.layers.Dense(1)
])

# Compile the model
estimator_model.compile(
    optimizer='adam',
    loss=tf.keras.losses.BinaryCrossentropy(from_logits=True),
    metrics=['accuracy'])
Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/mobilenet_v2/mobilenet_v2_weights_tf_dim_ordering_tf_kernels_1.0_160_no_top.h5
9412608/9406464 [==============================] - 0s 0us/step
9420800/9406464 [==============================] - 0s 0us/step

Buat Estimator dari model Keras yang dikompilasi. Status model awal model Keras dipertahankan dalam Estimator yang dibuat :

est_mobilenet_v2 = tf.keras.estimator.model_to_estimator(keras_model=estimator_model)
INFO:tensorflow:Using default config.
WARNING:tensorflow:Using temporary folder as model directory: /tmp/tmpmosnmied
INFO:tensorflow:Using the Keras model provided.
/tmpfs/src/tf_docs_env/lib/python3.7/site-packages/keras/backend.py:401: UserWarning: `tf.keras.backend.set_learning_phase` is deprecated and will be removed after 2020-10-11. To update it, simply pass a True/False value to the `training` argument of the `__call__` method of your layer or model.
  warnings.warn('`tf.keras.backend.set_learning_phase` is deprecated and '
/tmpfs/src/tf_docs_env/lib/python3.7/site-packages/keras/utils/generic_utils.py:497: CustomMaskWarning: Custom mask layers require a config and must override get_config. When loading, the custom mask layer must be passed to the custom_objects argument.
  category=CustomMaskWarning)
INFO:tensorflow:Using config: {'_model_dir': '/tmp/tmpmosnmied', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_checkpoint_save_graph_def': True, '_service': None, '_cluster_spec': ClusterSpec({}), '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}

Perlakukan Estimator turunan seperti yang Anda lakukan dengan Estimator lainnya .

IMG_SIZE = 160  # All images will be resized to 160x160

def preprocess(image, label):
  image = tf.cast(image, tf.float32)
  image = (image/127.5) - 1
  image = tf.image.resize(image, (IMG_SIZE, IMG_SIZE))
  return image, label
def train_input_fn(batch_size):
  data = tfds.load('cats_vs_dogs', as_supervised=True)
  train_data = data['train']
  train_data = train_data.map(preprocess).shuffle(500).batch(batch_size)
  return train_data

Untuk melatih, panggil fungsi kereta Estimator:

est_mobilenet_v2.train(input_fn=lambda: train_input_fn(32), steps=50)
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Warm-starting with WarmStartSettings: WarmStartSettings(ckpt_to_initialize_from='/tmp/tmpmosnmied/keras/keras_model.ckpt', vars_to_warm_start='.*', var_name_to_vocab_info={}, var_name_to_prev_var_name={})
INFO:tensorflow:Warm-starting with WarmStartSettings: WarmStartSettings(ckpt_to_initialize_from='/tmp/tmpmosnmied/keras/keras_model.ckpt', vars_to_warm_start='.*', var_name_to_vocab_info={}, var_name_to_prev_var_name={})
INFO:tensorflow:Warm-starting from: /tmp/tmpmosnmied/keras/keras_model.ckpt
INFO:tensorflow:Warm-starting from: /tmp/tmpmosnmied/keras/keras_model.ckpt
INFO:tensorflow:Warm-starting variables only in TRAINABLE_VARIABLES.
INFO:tensorflow:Warm-starting variables only in TRAINABLE_VARIABLES.
INFO:tensorflow:Warm-started 158 variables.
INFO:tensorflow:Warm-started 158 variables.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 0...
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 0...
INFO:tensorflow:Saving checkpoints for 0 into /tmp/tmpmosnmied/model.ckpt.
INFO:tensorflow:Saving checkpoints for 0 into /tmp/tmpmosnmied/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 0...
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 0...
INFO:tensorflow:loss = 0.6994096, step = 0
INFO:tensorflow:loss = 0.6994096, step = 0
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 50...
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 50...
INFO:tensorflow:Saving checkpoints for 50 into /tmp/tmpmosnmied/model.ckpt.
INFO:tensorflow:Saving checkpoints for 50 into /tmp/tmpmosnmied/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 50...
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 50...
INFO:tensorflow:Loss for final step: 0.68789804.
INFO:tensorflow:Loss for final step: 0.68789804.
<tensorflow_estimator.python.estimator.estimator.EstimatorV2 at 0x7f4b1c1e9890>

Demikian pula, untuk mengevaluasi, panggil fungsi evaluasi Estimator:

est_mobilenet_v2.evaluate(input_fn=lambda: train_input_fn(32), steps=10)
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Calling model_fn.
/tmpfs/src/tf_docs_env/lib/python3.7/site-packages/keras/engine/training.py:2470: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.
  warnings.warn('`Model.state_updates` will be removed in a future version. '
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Starting evaluation at 2021-09-22T20:49:36
INFO:tensorflow:Starting evaluation at 2021-09-22T20:49:36
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /tmp/tmpmosnmied/model.ckpt-50
INFO:tensorflow:Restoring parameters from /tmp/tmpmosnmied/model.ckpt-50
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Evaluation [1/10]
INFO:tensorflow:Evaluation [1/10]
INFO:tensorflow:Evaluation [2/10]
INFO:tensorflow:Evaluation [2/10]
INFO:tensorflow:Evaluation [3/10]
INFO:tensorflow:Evaluation [3/10]
INFO:tensorflow:Evaluation [4/10]
INFO:tensorflow:Evaluation [4/10]
INFO:tensorflow:Evaluation [5/10]
INFO:tensorflow:Evaluation [5/10]
INFO:tensorflow:Evaluation [6/10]
INFO:tensorflow:Evaluation [6/10]
INFO:tensorflow:Evaluation [7/10]
INFO:tensorflow:Evaluation [7/10]
INFO:tensorflow:Evaluation [8/10]
INFO:tensorflow:Evaluation [8/10]
INFO:tensorflow:Evaluation [9/10]
INFO:tensorflow:Evaluation [9/10]
INFO:tensorflow:Evaluation [10/10]
INFO:tensorflow:Evaluation [10/10]
INFO:tensorflow:Inference Time : 3.89658s
INFO:tensorflow:Inference Time : 3.89658s
INFO:tensorflow:Finished evaluation at 2021-09-22-20:49:39
INFO:tensorflow:Finished evaluation at 2021-09-22-20:49:39
INFO:tensorflow:Saving dict for global step 50: accuracy = 0.525, global_step = 50, loss = 0.6723582
INFO:tensorflow:Saving dict for global step 50: accuracy = 0.525, global_step = 50, loss = 0.6723582
INFO:tensorflow:Saving 'checkpoint_path' summary for global step 50: /tmp/tmpmosnmied/model.ckpt-50
INFO:tensorflow:Saving 'checkpoint_path' summary for global step 50: /tmp/tmpmosnmied/model.ckpt-50
{'accuracy': 0.525, 'loss': 0.6723582, 'global_step': 50}

Untuk lebih jelasnya, silakan lihat dokumentasi untuk tf.keras.estimator.model_to_estimator .

Menyimpan pos pemeriksaan berbasis objek dengan Estimator

Penaksir secara default menyimpan pos pemeriksaan dengan nama variabel daripada grafik objek yang dijelaskan dalam panduan Pos Pemeriksaan . tf.train.Checkpoint akan membaca pos pemeriksaan berbasis nama, tetapi nama variabel dapat berubah saat memindahkan bagian dari model di luar model_fn Penaksir. Untuk kompatibilitas ke depan, menyimpan pos pemeriksaan berbasis objek memudahkan untuk melatih model di dalam Penaksir dan kemudian menggunakannya di luar Penaksir.

import tensorflow.compat.v1 as tf_compat
def toy_dataset():
  inputs = tf.range(10.)[:, None]
  labels = inputs * 5. + tf.range(5.)[None, :]
  return tf.data.Dataset.from_tensor_slices(
    dict(x=inputs, y=labels)).repeat().batch(2)
class Net(tf.keras.Model):
  """A simple linear model."""

  def __init__(self):
    super(Net, self).__init__()
    self.l1 = tf.keras.layers.Dense(5)

  def call(self, x):
    return self.l1(x)
def model_fn(features, labels, mode):
  net = Net()
  opt = tf.keras.optimizers.Adam(0.1)
  ckpt = tf.train.Checkpoint(step=tf_compat.train.get_global_step(),
                             optimizer=opt, net=net)
  with tf.GradientTape() as tape:
    output = net(features['x'])
    loss = tf.reduce_mean(tf.abs(output - features['y']))
  variables = net.trainable_variables
  gradients = tape.gradient(loss, variables)
  return tf.estimator.EstimatorSpec(
    mode,
    loss=loss,
    train_op=tf.group(opt.apply_gradients(zip(gradients, variables)),
                      ckpt.step.assign_add(1)),
    # Tell the Estimator to save "ckpt" in an object-based format.
    scaffold=tf_compat.train.Scaffold(saver=ckpt))

tf.keras.backend.clear_session()
est = tf.estimator.Estimator(model_fn, './tf_estimator_example/')
est.train(toy_dataset, steps=10)
INFO:tensorflow:Using default config.
INFO:tensorflow:Using default config.
INFO:tensorflow:Using config: {'_model_dir': './tf_estimator_example/', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_checkpoint_save_graph_def': True, '_service': None, '_cluster_spec': ClusterSpec({}), '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}
INFO:tensorflow:Using config: {'_model_dir': './tf_estimator_example/', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_checkpoint_save_graph_def': True, '_service': None, '_cluster_spec': ClusterSpec({}), '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 0...
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 0...
INFO:tensorflow:Saving checkpoints for 0 into ./tf_estimator_example/model.ckpt.
INFO:tensorflow:Saving checkpoints for 0 into ./tf_estimator_example/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 0...
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 0...
INFO:tensorflow:loss = 4.659403, step = 0
INFO:tensorflow:loss = 4.659403, step = 0
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 10...
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 10...
INFO:tensorflow:Saving checkpoints for 10 into ./tf_estimator_example/model.ckpt.
INFO:tensorflow:Saving checkpoints for 10 into ./tf_estimator_example/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 10...
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 10...
INFO:tensorflow:Loss for final step: 39.58891.
INFO:tensorflow:Loss for final step: 39.58891.
<tensorflow_estimator.python.estimator.estimator.EstimatorV2 at 0x7f4b7c451fd0>

tf.train.Checkpoint kemudian dapat memuat pos pemeriksaan Estimator dari model_dir .

opt = tf.keras.optimizers.Adam(0.1)
net = Net()
ckpt = tf.train.Checkpoint(
  step=tf.Variable(1, dtype=tf.int64), optimizer=opt, net=net)
ckpt.restore(tf.train.latest_checkpoint('./tf_estimator_example/'))
ckpt.step.numpy()  # From est.train(..., steps=10)
10

Model Tersimpan dari Penaksir

Penaksir mengekspor SavedModels melalui tf.Estimator.export_saved_model .

input_column = tf.feature_column.numeric_column("x")

estimator = tf.estimator.LinearClassifier(feature_columns=[input_column])

def input_fn():
  return tf.data.Dataset.from_tensor_slices(
    ({"x": [1., 2., 3., 4.]}, [1, 1, 0, 0])).repeat(200).shuffle(64).batch(16)
estimator.train(input_fn)
INFO:tensorflow:Using default config.
INFO:tensorflow:Using default config.
WARNING:tensorflow:Using temporary folder as model directory: /tmp/tmp30_d7xz6
WARNING:tensorflow:Using temporary folder as model directory: /tmp/tmp30_d7xz6
INFO:tensorflow:Using config: {'_model_dir': '/tmp/tmp30_d7xz6', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_checkpoint_save_graph_def': True, '_service': None, '_cluster_spec': ClusterSpec({}), '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}
INFO:tensorflow:Using config: {'_model_dir': '/tmp/tmp30_d7xz6', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_checkpoint_save_graph_def': True, '_service': None, '_cluster_spec': ClusterSpec({}), '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 0...
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 0...
INFO:tensorflow:Saving checkpoints for 0 into /tmp/tmp30_d7xz6/model.ckpt.
INFO:tensorflow:Saving checkpoints for 0 into /tmp/tmp30_d7xz6/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 0...
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 0...
INFO:tensorflow:loss = 0.6931472, step = 0
INFO:tensorflow:loss = 0.6931472, step = 0
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 50...
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 50...
INFO:tensorflow:Saving checkpoints for 50 into /tmp/tmp30_d7xz6/model.ckpt.
INFO:tensorflow:Saving checkpoints for 50 into /tmp/tmp30_d7xz6/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 50...
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 50...
INFO:tensorflow:Loss for final step: 0.4022895.
INFO:tensorflow:Loss for final step: 0.4022895.
<tensorflow_estimator.python.estimator.canned.linear.LinearClassifierV2 at 0x7f4b1c10fd10>

Untuk menyimpan Estimator , Anda perlu membuat serving_input_receiver . Fungsi ini membangun bagian dari tf.Graph yang mem-parsing data mentah yang diterima oleh SavedModel.

Modul tf.estimator.export berisi fungsi-fungsi untuk membantu membangun receivers ini.

Kode berikut membangun penerima, berdasarkan feature_columns , yang menerima buffer protokol tf.Example serial, yang sering digunakan dengan tf-serving .

tmpdir = tempfile.mkdtemp()

serving_input_fn = tf.estimator.export.build_parsing_serving_input_receiver_fn(
  tf.feature_column.make_parse_example_spec([input_column]))

estimator_base_path = os.path.join(tmpdir, 'from_estimator')
estimator_path = estimator.export_saved_model(estimator_base_path, serving_input_fn)
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Done calling model_fn.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.7/site-packages/tensorflow/python/saved_model/signature_def_utils_impl.py:145: 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.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.7/site-packages/tensorflow/python/saved_model/signature_def_utils_impl.py:145: 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:Signatures INCLUDED in export for Classify: ['serving_default', 'classification']
INFO:tensorflow:Signatures INCLUDED in export for Classify: ['serving_default', 'classification']
INFO:tensorflow:Signatures INCLUDED in export for Regress: ['regression']
INFO:tensorflow:Signatures INCLUDED in export for Regress: ['regression']
INFO:tensorflow:Signatures INCLUDED in export for Predict: ['predict']
INFO:tensorflow:Signatures INCLUDED in export for Predict: ['predict']
INFO:tensorflow:Signatures INCLUDED in export for Train: None
INFO:tensorflow:Signatures INCLUDED in export for Train: None
INFO:tensorflow:Signatures INCLUDED in export for Eval: None
INFO:tensorflow:Signatures INCLUDED in export for Eval: None
INFO:tensorflow:Restoring parameters from /tmp/tmp30_d7xz6/model.ckpt-50
INFO:tensorflow:Restoring parameters from /tmp/tmp30_d7xz6/model.ckpt-50
INFO:tensorflow:Assets added to graph.
INFO:tensorflow:Assets added to graph.
INFO:tensorflow:No assets to write.
INFO:tensorflow:No assets to write.
INFO:tensorflow:SavedModel written to: /tmp/tmpi_szzuj1/from_estimator/temp-1632343781/saved_model.pb
INFO:tensorflow:SavedModel written to: /tmp/tmpi_szzuj1/from_estimator/temp-1632343781/saved_model.pb

Anda juga dapat memuat dan menjalankan model itu, dari python:

imported = tf.saved_model.load(estimator_path)

def predict(x):
  example = tf.train.Example()
  example.features.feature["x"].float_list.value.extend([x])
  return imported.signatures["predict"](
    examples=tf.constant([example.SerializeToString()]))
print(predict(1.5))
print(predict(3.5))
{'class_ids': <tf.Tensor: shape=(1, 1), dtype=int64, numpy=array([[1]])>, 'classes': <tf.Tensor: shape=(1, 1), dtype=string, numpy=array([[b'1']], dtype=object)>, 'all_classes': <tf.Tensor: shape=(1, 2), dtype=string, numpy=array([[b'0', b'1']], dtype=object)>, 'all_class_ids': <tf.Tensor: shape=(1, 2), dtype=int32, numpy=array([[0, 1]], dtype=int32)>, 'logits': <tf.Tensor: shape=(1, 1), dtype=float32, numpy=array([[0.2974025]], dtype=float32)>, 'logistic': <tf.Tensor: shape=(1, 1), dtype=float32, numpy=array([[0.5738074]], dtype=float32)>, 'probabilities': <tf.Tensor: shape=(1, 2), dtype=float32, numpy=array([[0.42619258, 0.5738074 ]], dtype=float32)>}
{'class_ids': <tf.Tensor: shape=(1, 1), dtype=int64, numpy=array([[0]])>, 'classes': <tf.Tensor: shape=(1, 1), dtype=string, numpy=array([[b'0']], dtype=object)>, 'all_classes': <tf.Tensor: shape=(1, 2), dtype=string, numpy=array([[b'0', b'1']], dtype=object)>, 'all_class_ids': <tf.Tensor: shape=(1, 2), dtype=int32, numpy=array([[0, 1]], dtype=int32)>, 'logits': <tf.Tensor: shape=(1, 1), dtype=float32, numpy=array([[-1.1919093]], dtype=float32)>, 'logistic': <tf.Tensor: shape=(1, 1), dtype=float32, numpy=array([[0.23291764]], dtype=float32)>, 'probabilities': <tf.Tensor: shape=(1, 2), dtype=float32, numpy=array([[0.7670824 , 0.23291762]], dtype=float32)>}

tf.estimator.export.build_raw_serving_input_receiver_fn memungkinkan Anda membuat fungsi input yang menggunakan tensor mentah daripada tf.train.Example s.

Menggunakan tf.distribute.Strategy dengan Estimator (Dukungan terbatas)

tf.estimator adalah API TensorFlow pelatihan terdistribusi yang awalnya mendukung pendekatan server parameter async. tf.estimator sekarang mendukung tf.distribute.Strategy . Jika Anda menggunakan tf.estimator , Anda dapat mengubah ke pelatihan terdistribusi dengan sedikit perubahan pada kode Anda. Dengan ini, pengguna Estimator sekarang dapat melakukan pelatihan terdistribusi secara sinkron pada beberapa GPU dan beberapa pekerja, serta menggunakan TPU. Namun, dukungan di Estimator ini terbatas. Lihat bagian Apa yang didukung sekarang di bawah ini untuk detail selengkapnya.

Menggunakan tf.distribute.Strategy dengan Estimator sedikit berbeda dengan kasus Keras. Alih-alih menggunakan strategy.scope , sekarang Anda meneruskan objek strategi ke RunConfig untuk Estimator.

Anda dapat merujuk ke panduan pelatihan terdistribusi untuk informasi lebih lanjut.

Berikut adalah cuplikan kode yang menunjukkan ini dengan Estimator LinearRegressor dan MirroredStrategy yang dibuat sebelumnya :

mirrored_strategy = tf.distribute.MirroredStrategy()
config = tf.estimator.RunConfig(
    train_distribute=mirrored_strategy, eval_distribute=mirrored_strategy)
regressor = tf.estimator.LinearRegressor(
    feature_columns=[tf.feature_column.numeric_column('feats')],
    optimizer='SGD',
    config=config)
INFO:tensorflow:Using MirroredStrategy with devices ('/job:localhost/replica:0/task:0/device:GPU:0',)
INFO:tensorflow:Using MirroredStrategy with devices ('/job:localhost/replica:0/task:0/device:GPU:0',)
INFO:tensorflow:Initializing RunConfig with distribution strategies.
INFO:tensorflow:Initializing RunConfig with distribution strategies.
INFO:tensorflow:Not using Distribute Coordinator.
INFO:tensorflow:Not using Distribute Coordinator.
WARNING:tensorflow:Using temporary folder as model directory: /tmp/tmpftw63jyd
WARNING:tensorflow:Using temporary folder as model directory: /tmp/tmpftw63jyd
INFO:tensorflow:Using config: {'_model_dir': '/tmp/tmpftw63jyd', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': <tensorflow.python.distribute.mirrored_strategy.MirroredStrategy object at 0x7f4b0c04c050>, '_device_fn': None, '_protocol': None, '_eval_distribute': <tensorflow.python.distribute.mirrored_strategy.MirroredStrategy object at 0x7f4b0c04c050>, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_checkpoint_save_graph_def': True, '_service': None, '_cluster_spec': ClusterSpec({}), '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1, '_distribute_coordinator_mode': None}
INFO:tensorflow:Using config: {'_model_dir': '/tmp/tmpftw63jyd', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': <tensorflow.python.distribute.mirrored_strategy.MirroredStrategy object at 0x7f4b0c04c050>, '_device_fn': None, '_protocol': None, '_eval_distribute': <tensorflow.python.distribute.mirrored_strategy.MirroredStrategy object at 0x7f4b0c04c050>, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_checkpoint_save_graph_def': True, '_service': None, '_cluster_spec': ClusterSpec({}), '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1, '_distribute_coordinator_mode': None}

Di sini, Anda menggunakan Estimator yang dibuat sebelumnya, tetapi kode yang sama juga berfungsi dengan Estimator khusus. train_distribute menentukan bagaimana pelatihan akan didistribusikan, dan eval_distribute menentukan bagaimana evaluasi akan didistribusikan. Ini adalah perbedaan lain dari Keras di mana Anda menggunakan strategi yang sama untuk pelatihan dan evaluasi.

Sekarang Anda dapat melatih dan mengevaluasi Estimator ini dengan fungsi input:

def input_fn():
  dataset = tf.data.Dataset.from_tensors(({"feats":[1.]}, [1.]))
  return dataset.repeat(1000).batch(10)
regressor.train(input_fn=input_fn, steps=10)
regressor.evaluate(input_fn=input_fn, steps=10)
INFO:tensorflow:Calling model_fn.
/tmpfs/src/tf_docs_env/lib/python3.7/site-packages/tensorflow/python/data/ops/dataset_ops.py:374: UserWarning: To make it possible to preserve tf.data options across serialization boundaries, their implementation has moved to be part of the TensorFlow graph. As a consequence, the options value is in general no longer known at graph construction time. Invoking this method in graph mode retains the legacy behavior of the original implementation, but note that the returned value might not reflect the actual value of the options.
  warnings.warn("To make it possible to preserve tf.data options across "
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Create CheckpointSaverHook.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.7/site-packages/tensorflow_estimator/python/estimator/util.py:95: DistributedIteratorV1.initialize (from tensorflow.python.distribute.input_lib) is deprecated and will be removed in a future version.
Instructions for updating:
Use the iterator's `initializer` property instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.7/site-packages/tensorflow_estimator/python/estimator/util.py:95: DistributedIteratorV1.initialize (from tensorflow.python.distribute.input_lib) is deprecated and will be removed in a future version.
Instructions for updating:
Use the iterator's `initializer` property instead.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 0...
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 0...
INFO:tensorflow:Saving checkpoints for 0 into /tmp/tmpftw63jyd/model.ckpt.
INFO:tensorflow:Saving checkpoints for 0 into /tmp/tmpftw63jyd/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 0...
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 0...
2021-09-22 20:49:45.706166: W tensorflow/core/grappler/utils/graph_view.cc:836] No registered 'MultiDeviceIteratorFromStringHandle' OpKernel for GPU devices compatible with node { {node MultiDeviceIteratorFromStringHandle} }
    .  Registered:  device='CPU'

2021-09-22 20:49:45.707521: W tensorflow/core/grappler/utils/graph_view.cc:836] No registered 'MultiDeviceIteratorGetNextFromShard' OpKernel for GPU devices compatible with node { {node MultiDeviceIteratorGetNextFromShard} }
    .  Registered:  device='CPU'
INFO:tensorflow:loss = 1.0, step = 0
INFO:tensorflow:loss = 1.0, step = 0
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 10...
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 10...
INFO:tensorflow:Saving checkpoints for 10 into /tmp/tmpftw63jyd/model.ckpt.
INFO:tensorflow:Saving checkpoints for 10 into /tmp/tmpftw63jyd/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 10...
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 10...
INFO:tensorflow:Loss for final step: 2.877698e-13.
INFO:tensorflow:Loss for final step: 2.877698e-13.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /replica:0/task:0/device:CPU:0 then broadcast to ('/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Starting evaluation at 2021-09-22T20:49:46
INFO:tensorflow:Starting evaluation at 2021-09-22T20:49:46
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /tmp/tmpftw63jyd/model.ckpt-10
INFO:tensorflow:Restoring parameters from /tmp/tmpftw63jyd/model.ckpt-10
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Done running local_init_op.
2021-09-22 20:49:46.680821: W tensorflow/core/grappler/utils/graph_view.cc:836] No registered 'MultiDeviceIteratorFromStringHandle' OpKernel for GPU devices compatible with node { {node MultiDeviceIteratorFromStringHandle} }
    .  Registered:  device='CPU'

2021-09-22 20:49:46.682161: W tensorflow/core/grappler/utils/graph_view.cc:836] No registered 'MultiDeviceIteratorGetNextFromShard' OpKernel for GPU devices compatible with node { {node MultiDeviceIteratorGetNextFromShard} }
    .  Registered:  device='CPU'
INFO:tensorflow:Evaluation [1/10]
INFO:tensorflow:Evaluation [1/10]
INFO:tensorflow:Evaluation [2/10]
INFO:tensorflow:Evaluation [2/10]
INFO:tensorflow:Evaluation [3/10]
INFO:tensorflow:Evaluation [3/10]
INFO:tensorflow:Evaluation [4/10]
INFO:tensorflow:Evaluation [4/10]
INFO:tensorflow:Evaluation [5/10]
INFO:tensorflow:Evaluation [5/10]
INFO:tensorflow:Evaluation [6/10]
INFO:tensorflow:Evaluation [6/10]
INFO:tensorflow:Evaluation [7/10]
INFO:tensorflow:Evaluation [7/10]
INFO:tensorflow:Evaluation [8/10]
INFO:tensorflow:Evaluation [8/10]
INFO:tensorflow:Evaluation [9/10]
INFO:tensorflow:Evaluation [9/10]
INFO:tensorflow:Evaluation [10/10]
INFO:tensorflow:Evaluation [10/10]
INFO:tensorflow:Inference Time : 0.26514s
INFO:tensorflow:Inference Time : 0.26514s
INFO:tensorflow:Finished evaluation at 2021-09-22-20:49:46
INFO:tensorflow:Finished evaluation at 2021-09-22-20:49:46
INFO:tensorflow:Saving dict for global step 10: average_loss = 1.4210855e-14, global_step = 10, label/mean = 1.0, loss = 1.4210855e-14, prediction/mean = 0.99999994
INFO:tensorflow:Saving dict for global step 10: average_loss = 1.4210855e-14, global_step = 10, label/mean = 1.0, loss = 1.4210855e-14, prediction/mean = 0.99999994
INFO:tensorflow:Saving 'checkpoint_path' summary for global step 10: /tmp/tmpftw63jyd/model.ckpt-10
INFO:tensorflow:Saving 'checkpoint_path' summary for global step 10: /tmp/tmpftw63jyd/model.ckpt-10
{'average_loss': 1.4210855e-14,
 'label/mean': 1.0,
 'loss': 1.4210855e-14,
 'prediction/mean': 0.99999994,
 'global_step': 10}

Perbedaan lain yang perlu disorot di sini antara Estimator dan Keras adalah penanganan input. Di Keras, setiap kumpulan dataset dibagi secara otomatis di beberapa replika. Namun, di Estimator, Anda tidak melakukan pemisahan batch otomatis, atau secara otomatis melakukan sharding data ke pekerja yang berbeda. Anda memiliki kontrol penuh atas bagaimana Anda ingin data Anda didistribusikan ke seluruh pekerja dan perangkat, dan Anda harus memberikan input_fn untuk menentukan cara mendistribusikan data Anda.

input_fn Anda dipanggil sekali per pekerja, sehingga memberikan satu set data per pekerja. Kemudian satu batch dari dataset tersebut diumpankan ke satu replika pada pekerja itu, sehingga mengkonsumsi N batch untuk N replika pada 1 pekerja. Dengan kata lain, kumpulan data yang dikembalikan oleh input_fn harus menyediakan kumpulan ukuran PER_REPLICA_BATCH_SIZE . Dan ukuran batch global untuk satu langkah dapat diperoleh sebagai PER_REPLICA_BATCH_SIZE * strategy.num_replicas_in_sync .

Saat melakukan pelatihan multi-pekerja, Anda harus membagi data Anda ke seluruh pekerja, atau mengacak dengan benih acak pada masing-masing pekerja. Anda dapat memeriksa contoh bagaimana melakukannya di pelatihan Multi-pekerja dengan tutorial Pengukur .

Demikian pula, Anda juga dapat menggunakan strategi multi-pekerja dan server parameter. Kodenya tetap sama, tetapi Anda perlu menggunakan tf.estimator.train_and_evaluate , dan menyetel variabel lingkungan TF_CONFIG untuk setiap biner yang berjalan di cluster Anda.

Apa yang didukung sekarang?

Dukungan terbatas untuk pelatihan dengan Estimator menggunakan semua strategi kecuali TPUStrategy . Pelatihan dasar dan evaluasi harus bekerja, tetapi sejumlah fitur lanjutan seperti v1.train.Scaffold tidak. Mungkin juga ada sejumlah bug dalam integrasi ini dan tidak ada rencana untuk secara aktif meningkatkan dukungan ini (fokusnya adalah pada Keras dan dukungan loop pelatihan khusus). Jika memungkinkan, Anda sebaiknya menggunakan tf.distribute dengan API tersebut.

Pelatihan API Strategi Tercermin strategi TPUS MultiWorkerMirroredStrategy Strategi Penyimpanan Pusat ParameterServerStrategy
API Penaksir Dukungan terbatas Tidak didukung Dukungan terbatas Dukungan terbatas Dukungan terbatas

Contoh dan tutorial

Berikut adalah beberapa contoh ujung ke ujung yang menunjukkan cara menggunakan berbagai strategi dengan Estimator:

  1. Tutorial Pelatihan Multi-Pekerja dengan Penaksir menunjukkan bagaimana Anda dapat berlatih dengan banyak pekerja menggunakan MultiWorkerMirroredStrategy pada set data MNIST.
  2. Contoh end-to-end menjalankan pelatihan multi-pekerja dengan strategi distribusi di tensorflow/ecosystem menggunakan template Kubernetes. Itu dimulai dengan model Keras dan mengonversinya menjadi Penaksir menggunakan API tf.keras.estimator.model_to_estimator .
  3. Model ResNet50 resmi, yang dapat dilatih menggunakan MirroredStrategy atau MultiWorkerMirroredStrategy .