學習目標：
- 使用 TensorFlow
DNNRegressor類定義神經網絡 (NN) 及其隱藏層 - 訓練神經網絡學習數據集中的非線性規律，並實現比線性迴歸模型更好的效果
在之前的練習中，我們使用合成特徵來幫助模型學習非線性規律。
一組重要的非線性關係是緯度和經度的關係，但也可能存在其他非線性關係。
現在我們從之前練習中的邏輯迴歸任務回到標準的（線性）迴歸任務。也就是說，我們將直接預測 median_house_value。
from __future__ import print_function
import math
from IPython import display
from matplotlib import cm
from matplotlib import gridspec
from matplotlib import pyplot as plt
import numpy as np
import pandas as pd
from sklearn import metrics
import tensorflow as tf
from tensorflow.python.data import Dataset
tf.logging.set_verbosity(tf.logging.ERROR)
pd.options.display.max_rows = 10
pd.options.display.float_format = '{:.1f}'.format
california_housing_dataframe = pd.read_csv("https://download.mlcc.google.cn/mledu-datasets/california_housing_train.csv", sep=",")
california_housing_dataframe = california_housing_dataframe.reindex(
np.random.permutation(california_housing_dataframe.index))
def preprocess_features(california_housing_dataframe):
selected_features = california_housing_dataframe[
["latitude",
"longitude",
"housing_median_age",
"total_rooms",
"total_bedrooms",
"population",
"households",
"median_income"]]
processed_features = selected_features.copy()
# Create a synthetic feature.
processed_features["rooms_per_person"] = (
california_housing_dataframe["total_rooms"] /
california_housing_dataframe["population"])
return processed_features
def preprocess_targets(california_housing_dataframe):
output_targets = pd.DataFrame()
# Scale the target to be in units of thousands of dollars.
output_targets["median_house_value"] = (
california_housing_dataframe["median_house_value"] / 1000.0)
return output_targets
training_examples = preprocess_features(california_housing_dataframe.head(12000))
training_targets = preprocess_targets(california_housing_dataframe.head(12000))
# Choose the last 5000 (out of 17000) examples for validation.
validation_examples = preprocess_features(california_housing_dataframe.tail(5000))
validation_targets = preprocess_targets(california_housing_dataframe.tail(5000))
# Double-check that we've done the right thing.
print("Training examples summary:")
display.display(training_examples.describe())
print("Validation examples summary:")
display.display(validation_examples.describe())
print("Training targets summary:")
display.display(training_targets.describe())
print("Validation targets summary:")
display.display(validation_targets.describe())
def construct_feature_columns(input_features):
return set([tf.feature_column.numeric_column(
my_feature) for my_feature in input_features])
def my_input_fn(features,targets,batch_size=1,shuffle=True,num_epochs=None):
features={key:np.array(value) for key,value in dict(features).items()}
ds=Dataset.from_tensor_slices((features,targets))
ds=ds.batch(batch_size).repeat(num_epochs)
if shuffle:
ds=ds.shuffle(10000)
features,labels=ds.make_one_shot_iterator().get_next()
return features,labels
用 hidden_units 定義神經網絡的結構。hidden_units 參數會創建一個整數列表，其中每個整數對應一個隱藏層，表示其中的節點數。以下面的賦值爲例：
hidden_units=[3,10]
上述賦值爲神經網絡指定了兩個隱藏層：
- 第一個隱藏層包含 3 個節點。
- 第二個隱藏層包含 10 個節點。
如果我們想要添加更多層，可以向該列表添加更多整數。例如，hidden_units=[10,20,30,40] 會創建 4 個分別包含 10、20、30 和 40 個單元的隱藏層。
默認情況下，所有隱藏層都會使用 ReLu 激活函數，且是全連接層。
def train_nn_regression_model(
learning_rate,
steps,
batch_size,
hidden_units,
training_examples,
training_targets,
validation_examples,
validation_targets):
periods = 10
steps_per_period = steps / periods
# Create a DNNRegressor object.
my_optimizer = tf.train.GradientDescentOptimizer(learning_rate=learning_rate)
my_optimizer = tf.contrib.estimator.clip_gradients_by_norm(my_optimizer, 5.0)
dnn_regressor = tf.estimator.DNNRegressor(
feature_columns=construct_feature_columns(training_examples),
hidden_units=hidden_units,
optimizer=my_optimizer
)
# Create input functions.
training_input_fn = lambda: my_input_fn(training_examples,
training_targets["median_house_value"],
batch_size=batch_size)
predict_training_input_fn = lambda: my_input_fn(training_examples,
training_targets["median_house_value"],
num_epochs=1,
shuffle=False)
predict_validation_input_fn = lambda: my_input_fn(validation_examples,
validation_targets["median_house_value"],
num_epochs=1,
shuffle=False)
# Train the model, but do so inside a loop so that we can periodically assess
# loss metrics.
print("Training model...")
print("RMSE (on training data):")
training_rmse = []
validation_rmse = []
for period in range (0, periods):
# Train the model, starting from the prior state.
dnn_regressor.train(
input_fn=training_input_fn,
steps=steps_per_period
)
# Take a break and compute predictions.
training_predictions = dnn_regressor.predict(input_fn=predict_training_input_fn)
training_predictions = np.array([item['predictions'][0] for item in training_predictions])
validation_predictions = dnn_regressor.predict(input_fn=predict_validation_input_fn)
validation_predictions = np.array([item['predictions'][0] for item in validation_predictions])
# Compute training and validation loss.
training_root_mean_squared_error = math.sqrt(
metrics.mean_squared_error(training_predictions, training_targets))
validation_root_mean_squared_error = math.sqrt(
metrics.mean_squared_error(validation_predictions, validation_targets))
# Occasionally print the current loss.
print(" period %02d : %0.2f" % (period, training_root_mean_squared_error))
# Add the loss metrics from this period to our list.
training_rmse.append(training_root_mean_squared_error)
validation_rmse.append(validation_root_mean_squared_error)
print("Model training finished.")
# Output a graph of loss metrics over periods.
plt.ylabel("RMSE")
plt.xlabel("Periods")
plt.title("Root Mean Squared Error vs. Periods")
plt.tight_layout()
plt.plot(training_rmse, label="training")
plt.plot(validation_rmse, label="validation")
plt.legend()
print("Final RMSE (on training data): %0.2f" % training_root_mean_squared_error)
print("Final RMSE (on validation data): %0.2f" % validation_root_mean_squared_error)
return dnn_regressor
訓練神經網絡模型
調整超參數，目標是將 RMSE 降到 110 以下。
運行以下代碼塊來訓練神經網絡模型。
我們已經知道，在使用了很多特徵的線性迴歸練習中，110 左右的 RMSE 已經是相當不錯的結果。我們將得到比它更好的結果。
在此練習中，您的任務是修改各種學習設置，以提高在驗證數據上的準確率。
對於神經網絡而言，過擬合是一種真正的潛在危險。您可以查看訓練數據損失與驗證數據損失之間的差值，以幫助判斷模型是否有過擬合的趨勢。如果差值開始變大，則通常可以肯定存在過擬合。
由於存在很多不同的可能設置，強烈建議您記錄每次試驗，以在開發流程中進行參考。
此外，獲得效果出色的設置後，嘗試多次運行該設置，看看結果的重複程度。由於神經網絡權重通常會初始化爲較小的隨機值，因此每次運行結果應該存在差異。
dnn_regressor = train_nn_regression_model(
learning_rate=0.01,
steps=500,
batch_size=10,
hidden_units=[10, 2],
training_examples=training_examples,
training_targets=training_targets,
validation_examples=validation_examples,
validation_targets=validation_targets)
dnn_regressor = train_nn_regression_model(
learning_rate=0.001,
steps=2000,
batch_size=100,
hidden_units=[10, 10],
training_examples=training_examples,
training_targets=training_targets,
validation_examples=validation_examples,
validation_targets=validation_targets)
用測試數據進行評估
確認您的驗證效果結果經受得住測試數據的檢驗。
獲得滿意的模型後，用測試數據評估該模型，以與驗證效果進行比較。
california_housing_test_data = pd.read_csv("https://download.mlcc.google.cn/mledu-datasets/california_housing_test.csv", sep=",")
test_examples = preprocess_features(california_housing_test_data)
test_targets = preprocess_targets(california_housing_test_data)
predict_testing_input_fn = lambda: my_input_fn(test_examples,
test_targets["median_house_value"],
num_epochs=1,
shuffle=False)
test_predictions = dnn_regressor.predict(input_fn=predict_testing_input_fn)
test_predictions = np.array([item['predictions'][0] for item in test_predictions])
root_mean_squared_error = math.sqrt(
metrics.mean_squared_error(test_predictions, test_targets))
print("Final RMSE (on test data): %0.2f" % root_mean_squared_error)
