1 # Copyright 2016 The TensorFlow Authors. All Rights Reserved. 2 # 3 # Licensed under the Apache License, Version 2.0 (the "License"); 4 # you may not use this file except in compliance with the License. 5 # You may obtain a copy of the License at 6 # 7 # http://www.apache.org/licenses/LICENSE-2.0 8 # 9 # Unless required by applicable law or agreed to in writing, software 10 # distributed under the License is distributed on an "AS IS" BASIS, 11 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 12 # See the License for the specific language governing permissions and 13 # limitations under the License. 14 15 """Example of DNNClassifier for Iris plant dataset, with run config.""" 16 17 from __future__ import absolute_import 18 from __future__ import division 19 from __future__ import print_function 20 21 import numpy as np 22 from sklearn import datasets 23 from sklearn import metrics 24 from sklearn import model_selection 25 import tensorflow as tf 26 27 28 X_FEATURE = 'x' # Name of the input feature. 29 30 31 def main(unused_argv): 32 # Load dataset. 33 iris = datasets.load_iris() 34 x_train, x_test, y_train, y_test = model_selection.train_test_split( 35 iris.data, iris.target, test_size=0.2, random_state=42) 36 37 # You can define you configurations by providing a RunConfig object to 38 # estimator to control session configurations, e.g. tf_random_seed. 39 run_config = tf.estimator.RunConfig().replace(tf_random_seed=1) 40 41 # Build 3 layer DNN with 10, 20, 10 units respectively. 42 feature_columns = [ 43 tf.feature_column.numeric_column( 44 X_FEATURE, shape=np.array(x_train).shape[1:])] 45 classifier = tf.estimator.DNNClassifier( 46 feature_columns=feature_columns, hidden_units=[10, 20, 10], n_classes=3, 47 config=run_config) 48 49 # Train. 50 train_input_fn = tf.estimator.inputs.numpy_input_fn( 51 x={X_FEATURE: x_train}, y=y_train, num_epochs=None, shuffle=True) 52 classifier.train(input_fn=train_input_fn, steps=200) 53 54 # Predict. 55 test_input_fn = tf.estimator.inputs.numpy_input_fn( 56 x={X_FEATURE: x_test}, y=y_test, num_epochs=1, shuffle=False) 57 predictions = classifier.predict(input_fn=test_input_fn) 58 y_predicted = np.array(list(p['class_ids'] for p in predictions)) 59 y_predicted = y_predicted.reshape(np.array(y_test).shape) 60 61 # Score with sklearn. 62 score = metrics.accuracy_score(y_test, y_predicted) 63 print('Accuracy (sklearn): {0:f}'.format(score)) 64 65 # Score with tensorflow. 66 scores = classifier.evaluate(input_fn=test_input_fn) 67 print('Accuracy (tensorflow): {0:f}'.format(scores['accuracy'])) 68 69 70 if __name__ == '__main__': 71 tf.app.run() 72
