tensorflow 高级api使用

1、定义模型

def my_model(features,labels,mode,params):#此处的features是真实的数据，不是特征列

return tf.estimator.EstimatorSpec(mode, loss=loss, train_op=train_op)

2、estimator包裹自定义模型

classifier = tf.estimator.Estimator(
  model_fn=my_model,
  params={
    'feature_columns': my_feature_columns,#包含特征列的列表
    'hidden_units': FLAGS.hidden_units.split(','),
    'learning_rate': FLAGS.learning_rate,
    'dropout_rate': FLAGS.dropout_rate
  },
  config=tf.estimator.RunConfig(model_dir=FLAGS.model_dir, save_checkpoints_steps=FLAGS.save_checkpoints_steps)
)

3、自定义生成数据的函数

这里需要注意的是：如果自定义模型时采用

net = fc.input_layer(features, params['feature_columns'])

输入数据时，而不是tf.placeholder的方式时，需要用以下方式自定义生成数据的函数

def parse_exmp(serial_exmp):
  click = fc.numeric_column("click", default_value=0, dtype=tf.int64)
  pay = fc.numeric_column("pay", default_value=0, dtype=tf.int64)
  fea_columns = [click, pay]
  fea_columns += my_feature_columns
  feature_spec = tf.feature_column.make_parse_example_spec(fea_columns)
  #把数据映射过来
  feats = tf.parse_single_example(serial_exmp, features=feature_spec)
  click = feats.pop('click')
  pay = feats.pop('pay')
  return feats, {'ctr': tf.to_float(click), 'cvr': tf.to_float(pay)}


def train_input_fn(filenames, batch_size, shuffle_buffer_size):
  #dataset = tf.data.TFRecordDataset(filenames)
  files = tf.data.Dataset.list_files(filenames)
  dataset = files.apply(tf.contrib.data.parallel_interleave(tf.data.TFRecordDataset, cycle_length=FLAGS.num_parallel_readers))
  # Shuffle, repeat, and batch the examples.
  if shuffle_buffer_size > 0:
    dataset = dataset.shuffle(shuffle_buffer_size)
  #dataset = dataset.apply(tf.contrib.data.map_and_batch(map_func=parse_exmp, batch_size=batch_size))
  #dataset = dataset.repeat().prefetch(1)
  dataset = dataset.map(parse_exmp, num_parallel_calls=8)
  dataset = dataset.repeat().batch(batch_size).prefetch(1)
  print(dataset.output_types)
  print(dataset.output_shapes)
  # Return the read end of the pipeline.
  return dataset

4、生成TrainSpec和Eval_Spec

#train_files 为tfrecord格式

train_spec = tf.estimator.TrainSpec(
  input_fn=lambda: train_input_fn(train_files, batch_size, shuffle_buffer_size),
  max_steps=FLAGS.train_steps
)

input_fn_for_eval = lambda: eval_input_fn(eval_files, batch_size)
eval_spec = tf.estimator.EvalSpec(input_fn=input_fn_for_eval, throttle_secs=600, steps=None)

5、训练并且评估

tf.estimator.train_and_evaluate(classifier, train_spec, eval_spec)

6、获得预测结果

results = classifier.evaluate(input_fn=input_fn_for_eval)