如何在tensorflow示例中发布的示例中更改DNNRegressor中的hidden_​​units？

Question

在tensorflow程序“https://www.tensorflow.org/get_started/input_fn”的实践中，在DNNRegressor中将hidden_​​units从[10,10]修改为[10,20,10]时，python抛出错误。似乎hidden_​​units只能设置为[10,10]，我不知道为什么以及如何修改它。该计划如下：

＃

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import itertools

import pandas as pd
import tensorflow as tf

tf.logging.set_verbosity(tf.logging.INFO)

COLUMNS = ["crim", "zn", "indus", "nox", "rm", "age",
           "dis", "tax", "ptratio", "medv"]
FEATURES = ["crim", "zn", "indus", "nox", "rm",
            "age", "dis", "tax", "ptratio"]
LABEL = "medv"

training_set = pd.read_csv("boston_train.csv", skipinitialspace=True,
                           skiprows=1, names=COLUMNS)
test_set = pd.read_csv("boston_test.csv", skipinitialspace=True,
                       skiprows=1, names=COLUMNS)
prediction_set = pd.read_csv("boston_predict.csv", skipinitialspace=True,
                             skiprows=1, names=COLUMNS)

feature_cols = [tf.contrib.layers.real_valued_column(k)
                  for k in FEATURES]
# [_RealValuedColumn(column_name='crim', dimension=1,
#    default_value=None, dtype=tf.float32, normalizer=None) ...]
print('feature_cols: ', feature_cols)

regressor = tf.contrib.learn.DNNRegressor(feature_columns=feature_cols,
                                          hidden_units=[10, 20, 10],
                                          model_dir="/tmp/boston_model")

classifier = tf.contrib.learn.DNNClassifier(
    hidden_units=[10, 20, 40, 20, 10],
    n_classes=3,
    dropout=0.2,
    feature_columns=feature_columns
    )

def input_fn(data_set):
  feature_cols = {k: tf.constant(data_set[k].values)
                  for k in FEATURES}
  labels = tf.constant(data_set[LABEL].values)
  return feature_cols, labels

regressor.fit(input_fn=lambda: input_fn(training_set), steps=5000)

ev = regressor.evaluate(input_fn=lambda: input_fn(test_set), steps=1)
print('ev: ',ev)
loss_score = ev["loss"]
print("Loss: {0:f}".format(loss_score))

y = regressor.predict(input_fn=lambda: input_fn(prediction_set))
# .predict() returns an iterator; convert to a list and print predictions
predictions = list(itertools.islice(y, 6))
print ("Predictions: {}".format(str(predictions)))

＃

错误信息是：

NotFoundError (see above for traceback): Key dnn/hiddenlayer_2/weights/t_0/Adagrad not found in checkpoint
     [[Node: save/RestoreV2_11 = RestoreV2[dtypes=[DT_FLOAT], _device="/job:localhost/replica:0/task:0/cpu:0"](_recv_save/Const_0, save/RestoreV2_11/tensor_names, save/RestoreV2_11/shape_and_slices)]]

Answer 1

尝试删除目录/ tmp / boston_model并再次运行，或更改：

regressor = tf.contrib.learn.DNNRegressor(feature_columns=feature_cols,
                                      hidden_units=[10, 20, 10],
                                      model_dir="/tmp/boston_model")

到

regressor = tf.contrib.learn.DNNRegressor(feature_columns=feature_cols,
                                      hidden_units=[10, 20, 10])

然后又跑了。