【机器学习】用 tensorflow 实现随机森林 RandomForest in tensorflow (mnist 示例)

随机森林

随机森林的基本知识，许多博客都已经讲的很清晰。 但是用 tensorflow 实现的代码却很少。
博主属于机器学习未入门级别，纯属分享一段可以运行的随机森林代码。若有错误，麻烦指出。

本代码实现用 mnist 示例，通过 tensorflow 实现随机森林。

import tensorflow as tf
import pandas as pd
import numpy as np
from tensorflow.contrib.tensor_forest.python import tensor_forest
from tensorflow.python.ops import resources

dataset = pd.read_csv('./data/train.csv').values

####################
# 读入数据进行数据处理
# 最终 X.shape 为 [?,784]
# y.shape 为 [?]，y 必须为数字标签，非 onehot标签。 
# 
# X 需要归一化处理
####################
# paremeters
num_steps = 5000    # 迭代次数
batch_size = 1000   # 每个处理批次的大小
num_classes = 10    # 类别的数目
num_features = 784  # 特征的数目
num_trees = 50      # 森林里树的个数
max_nodes = 800     # 每棵数的最大节点数目

# 定义 X 和 y
X = tf.placeholder(tf.float32, shape = [None, 784])
y = tf.placeholder(tf.int32, shape = [None])
  
# 定义随机森林的内部参数
hparams = tensor_forest.ForestHParams(num_classes = num_classes, num_features = num_features, num_trees = num_trees, max_nodes = max_nodes).fill()

# 创建随机森林运算图
forest_graph = tensor_forest.RandomForestGraphs(hparams)

# 定义训练运算图和 loss
train_op = forest_graph.training_graph(X, y)
loss_op = forest_graph.training_loss(X, y)
 
# 计算准确率
infer_op, _, _ = forest_graph.inference_graph(X)
correct_prediction = tf.equal(tf.argmax(infer_op, 1), tf.cast(y, tf.int64))
accuracy_op = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

# 定义图的初始化内容
init_var = tf.group(tf.global_variables_initializer(), resources.initialize_resources(resources.shared_resources()))

# 初始化
sess = tf.Session()
sess.run(init_var)


# 定义批次为随机 size 个训练数据
def next_batch(X_data, y_data, size):
    rand = np.arange(y_data.shape[0])
    np.random.shuffle(rand)
    return X_data[rand[0 : size]], y_data[rand[0 : size]]


# 训练过程
for i in range(1, num_steps + 1):
    batch = next_batch(X_train, y_train, batch_size)
    _, cost = sess.run([train_op, loss_op], feed_dict = {X: batch[0], y:batch[1]})
    
    if i % 100 == 0 or i == 1:
        train_acc = sess.run(accuracy_op, feed_dict={X: batch[0], y: batch[1]})
        val_acc = sess.run(accuracy_op, feed_dict={X: X_val, y: y_val})
        print('Step %i, Loss: %f, Train_Acc: %s, Test_acc: %s' % (i, cost, train_acc, val_acc))
 
print("Test Accuracy:", sess.run(accuracy_op, feed_dict={X:X_test y:y_test}))