Question

我是tensorflow的新手，我一直在关注如何创建自己的估算器和训练模型的一些文档。我能够定义一个自定义模型并训练模型。但是当我尝试预测并读取值时，我无法读取estimator.predict返回的值。以下是我的代码示例

import numpy as np
import tensorflow as tf

# Declare list of features, we only have one real-valued feature
def model_fn(features, labels, mode):
  # Build a linear model and predict values
  W = tf.get_variable("W", [1], dtype=tf.float64)
  b = tf.get_variable("b", [1], dtype=tf.float64)
  y = W*features['x'] + b
  # Loss sub-graph
  loss = tf.reduce_sum(tf.square(y - labels))
  # Training sub-graph
  global_step = tf.train.get_global_step()
  optimizer = tf.train.GradientDescentOptimizer(0.01)
  train = tf.group(optimizer.minimize(loss),
                   tf.assign_add(global_step, 1))
  # EstimatorSpec connects subgraphs we built to the
  # appropriate functionality.
  return tf.estimator.EstimatorSpec(
      mode=mode,
      predictions=y,
      loss=loss,
      train_op=train)

estimator = tf.estimator.Estimator(model_fn=model_fn)
# define our data sets
x_train = np.array([1., 2., 3., 4.])
y_train = np.array([0., -1., -2., -3.])
x_eval = np.array([2., 5., 8., 1.])
y_eval = np.array([-1.01, -4.1, -7, 0.])
input_fn = tf.estimator.inputs.numpy_input_fn(
    {"x":x_train}, y_train, batch_size=4, num_epochs=None, shuffle=True)
train_input_fn = tf.estimator.inputs.numpy_input_fn(
    {"x":x_train}, y_train, batch_size=4, num_epochs=1000, shuffle=False)
eval_input_fn = tf.estimator.inputs.numpy_input_fn(
    {"x":x_eval}, y_eval, batch_size=4, num_epochs=1000, shuffle=False)

estimate_input_fn = tf.estimator.inputs.numpy_input_fn({"x": x_eval},shuffle=False)

# train
estimator.train(input_fn=input_fn, steps=1000)
# Here we evaluate how well our model did.
train_metrics = estimator.evaluate(input_fn=train_input_fn)
eval_metrics = estimator.evaluate(input_fn=eval_input_fn)

estimate_metrics = estimator.predict(input_fn=estimate_input_fn, predict_keys=['y'])

print(list(estimate_metrics))

这会抛出错误

Traceback (most recent call last):
  File "/Users/hdattada/PycharmProjects/TensorFlowIntro/custom_model_linear.py", line 49, in <module>
    predictions = list(itertools.islice(estimate_metrics, 15))
  File "/Users/hdattada/.virtualenvs/tensorflow/lib/python3.5/site-packages/tensorflow/python/estimator/estimator.py", line 339, in predict
    model_fn_lib.ModeKeys.PREDICT)
  File "/Users/hdattada/.virtualenvs/tensorflow/lib/python3.5/site-packages/tensorflow/python/estimator/estimator.py", line 545, in _call_model_fn
    features=features, labels=labels, **kwargs)
  File "/Users/hdattada/PycharmProjects/TensorFlowIntro/custom_model_linear.py", line 12, in model_fn
    loss = tf.reduce_sum(tf.square(y - labels))
  File "/Users/hdattada/.virtualenvs/tensorflow/lib/python3.5/site-packages/tensorflow/python/ops/math_ops.py", line 829, in binary_op_wrapper
    y = ops.convert_to_tensor(y, dtype=x.dtype.base_dtype, name="y")
  File "/Users/hdattada/.virtualenvs/tensorflow/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 676, in convert_to_tensor
    as_ref=False)
  File "/Users/hdattada/.virtualenvs/tensorflow/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 741, in internal_convert_to_tensor
    ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
  File "/Users/hdattada/.virtualenvs/tensorflow/lib/python3.5/site-packages/tensorflow/python/framework/constant_op.py", line 113, in _constant_tensor_conversion_function
    return constant(v, dtype=dtype, name=name)
  File "/Users/hdattada/.virtualenvs/tensorflow/lib/python3.5/site-packages/tensorflow/python/framework/constant_op.py", line 102, in constant
    tensor_util.make_tensor_proto(value, dtype=dtype, shape=shape, verify_shape=verify_shape))
  File "/Users/hdattada/.virtualenvs/tensorflow/lib/python3.5/site-packages/tensorflow/python/framework/tensor_util.py", line 364, in make_tensor_proto
    raise ValueError("None values not supported.")
ValueError: None values not supported.

我正在使用TF1.2，有人可以帮帮我吗？

Answer 1

调用Estimator.predict时labels将为无，因此调用model_fn时图表会出错，您可以按如下方式更改model_fn：

def model_fn(features, labels, mode):
  # Build a linear model and predict values
  W = tf.get_variable("W", [1], dtype=tf.float64)
  b = tf.get_variable("b", [1], dtype=tf.float64)
  y = W*features['x'] + b
  loss = None
  train = None
  if labels is not None:
      # Loss sub-graph
      loss = tf.reduce_sum(tf.square(y - labels))
      # Training sub-graph
      global_step = tf.train.get_global_step()
      optimizer = tf.train.GradientDescentOptimizer(0.01)
      train = tf.group(optimizer.minimize(loss),
                       tf.assign_add(global_step, 1))
  # EstimatorSpec connects subgraphs we built to the
  # appropriate functionality.
  return tf.estimator.EstimatorSpec(
      mode=mode,
      predictions={"y":y},
      loss=loss,
      train_op=train)

Answer 2

更好的解决方案是在计算mode或任何类似的model_fn之前，先检查您是否处于预测loss中，然后退出labels：

...
y = W*features['x'] + b

if mode == tf.estimator.ModeKeys.PREDICT:
    return tf.estimator.EstimatorSpec(mode=mode, predictions=y)

loss = tf.reduce_sum(tf.square(y - labels))
...

无法读取张量流

2 个答案: