Eu uso um tensorflow para implementar um perceptron simples de múltiplas camadas para regressão. O código é modificado a partir do classificador mnist padrão, que apenas alterei o custo de saída para MSE (usetf.reduce_mean(tf.square(pred-y))) e algumas entradas, configurações de tamanho de saída. No entanto, se eu treinar a rede usando regressão, após várias épocas, o lote de saída será totalmente o mesmo. por exemplo:

target: 48.129, estimated: 42.634
target: 46.590, estimated: 42.634
target: 34.209, estimated: 42.634
target: 69.677, estimated: 42.634
......

Tentei diferentes tamanhos de lotes, diferentes inicialização, normalização de entrada usando sklearn.preprocessing.scale (meu intervalo de entradas é bem diferente). No entanto, nenhum deles funcionou. Eu também tentei um dos exemplos do sklearn do Tensorflow (Regressão de rede neural profunda com dados de Boston) Mas eu recebi outro erro na linha 40:

O objeto 'module' não tem atributo 'infer_real_valued_columns_from_input'

Alguém tem pistas sobre onde está o problema? Obrigado

Meu código está listado abaixo, pode ser um pouco longo, mas muito complicado:

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

import tensorflow as tf
from tensorflow.contrib import learn
import matplotlib.pyplot as plt

from sklearn.pipeline import Pipeline
from sklearn import datasets, linear_model
from sklearn import cross_validation
import numpy as np

boston = learn.datasets.load_dataset('boston')
x, y = boston.data, boston.target
X_train, X_test, Y_train, Y_test = cross_validation.train_test_split(
x, y, test_size=0.2, random_state=42)

total_len = X_train.shape[0]

# Parameters
learning_rate = 0.001
training_epochs = 500
batch_size = 10
display_step = 1
dropout_rate = 0.9
# Network Parameters
n_hidden_1 = 32 # 1st layer number of features
n_hidden_2 = 200 # 2nd layer number of features
n_hidden_3 = 200
n_hidden_4 = 256
n_input = X_train.shape[1]
n_classes = 1

# tf Graph input
x = tf.placeholder("float", [None, 13])
y = tf.placeholder("float", [None])

# Create model
def multilayer_perceptron(x, weights, biases):
    # Hidden layer with RELU activation
    layer_1 = tf.add(tf.matmul(x, weights['h1']), biases['b1'])
    layer_1 = tf.nn.relu(layer_1)

    # Hidden layer with RELU activation
    layer_2 = tf.add(tf.matmul(layer_1, weights['h2']), biases['b2'])
    layer_2 = tf.nn.relu(layer_2)

    # Hidden layer with RELU activation
    layer_3 = tf.add(tf.matmul(layer_2, weights['h3']), biases['b3'])
    layer_3 = tf.nn.relu(layer_3)

    # Hidden layer with RELU activation
    layer_4 = tf.add(tf.matmul(layer_3, weights['h4']), biases['b4'])
    layer_4 = tf.nn.relu(layer_4)

    # Output layer with linear activation
    out_layer = tf.matmul(layer_4, weights['out']) + biases['out']
    return out_layer

# Store layers weight & bias
weights = {
    'h1': tf.Variable(tf.random_normal([n_input, n_hidden_1], 0, 0.1)),
    'h2': tf.Variable(tf.random_normal([n_hidden_1, n_hidden_2], 0, 0.1)),
    'h3': tf.Variable(tf.random_normal([n_hidden_2, n_hidden_3], 0, 0.1)),
    'h4': tf.Variable(tf.random_normal([n_hidden_3, n_hidden_4], 0, 0.1)),
    'out': tf.Variable(tf.random_normal([n_hidden_4, n_classes], 0, 0.1))
}
biases = {
    'b1': tf.Variable(tf.random_normal([n_hidden_1], 0, 0.1)),
    'b2': tf.Variable(tf.random_normal([n_hidden_2], 0, 0.1)),
    'b3': tf.Variable(tf.random_normal([n_hidden_3], 0, 0.1)),
    'b4': tf.Variable(tf.random_normal([n_hidden_4], 0, 0.1)),
    'out': tf.Variable(tf.random_normal([n_classes], 0, 0.1))
}

# Construct model
pred = multilayer_perceptron(x, weights, biases)

# Define loss and optimizer
cost = tf.reduce_mean(tf.square(pred-y))
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)

# Launch the graph
with tf.Session() as sess:
    sess.run(tf.initialize_all_variables())

    # Training cycle
    for epoch in range(training_epochs):
        avg_cost = 0.
        total_batch = int(total_len/batch_size)
        # Loop over all batches
        for i in range(total_batch-1):
            batch_x = X_train[i*batch_size:(i+1)*batch_size]
            batch_y = Y_train[i*batch_size:(i+1)*batch_size]
            # Run optimization op (backprop) and cost op (to get loss value)
            _, c, p = sess.run([optimizer, cost, pred], feed_dict={x: batch_x,
                                                          y: batch_y})
            # Compute average loss
            avg_cost += c / total_batch

        # sample prediction
        label_value = batch_y
        estimate = p
        err = label_value-estimate
        print ("num batch:", total_batch)

        # Display logs per epoch step
        if epoch % display_step == 0:
            print ("Epoch:", '%04d' % (epoch+1), "cost=", \
                "{:.9f}".format(avg_cost))
            print ("[*]----------------------------")
            for i in xrange(3):
                print ("label value:", label_value[i], \
                    "estimated value:", estimate[i])
            print ("[*]============================")

    print ("Optimization Finished!")

    # Test model
    correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
    # Calculate accuracy
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))
    print ("Accuracy:", accuracy.eval({x: X_test, y: Y_test}))