我自己编写了一个TensorFlow类，如下所示，但在尝试在refine_init_weight函数中手动将一些权重设置为零时遇到了问题。在这个函数中，我尝试将所有低于某个值的数字设置为零，并观察准确率的变化。问题是，当我重新运行self.sess.run(self.accuracy, feed_dict={self.var_X: self.X_test, self.var_Y: self.y_test})时，似乎其值并未相应改变。我只是在想，在这种情况下，我应该在哪里更改符号变量（准确率取决于我更改的权重）？
import tensorflow as tffrom nncomponents import * from helpers import * from sda import StackedDenoisingAutoencoderclass DeepFeatureSelection: def __init__(self, X_train, X_test, y_train, y_test, weight_init='sda', hidden_dims=[100, 100, 100], epochs=1000, lambda1=0.001, lambda2=1.0, alpha1=0.001, alpha2=0.0, learning_rate=0.1, optimizer='FTRL'): # 初始化输入层 # 获取输入X的维度 n_sample, n_feat = X_train.shape n_classes = len(np.unique(y_train)) self.epochs = epochs # 存储原始值 self.X_train = X_train self.y_train = one_hot(y_train) self.X_test = X_test self.y_test = one_hot(y_test) # 创建长度未定的两个变量 self.var_X = tf.placeholder(dtype=tf.float32, shape=[None, n_feat], name='x') self.var_Y = tf.placeholder(dtype=tf.float32, shape=[None, n_classes], name='y') self.input_layer = One2OneInputLayer(self.var_X) self.hidden_layers = [] layer_input = self.input_layer.output # 初始化网络权重 weights, biases = init_layer_weight(hidden_dims, X_train, weight_init) print(type(weights[0])) # 创建隐藏层 for init_w,init_b in zip(weights, biases): self.hidden_layers.append(DenseLayer(layer_input, init_w, init_b)) layer_input = self.hidden_layers[-1].output # 最终分类层，传递变量Y self.softmax_layer = SoftmaxLayer(self.hidden_layers[-1].output, n_classes, self.var_Y) n_hidden = len(hidden_dims) # 输入层的系数正则化项 self.L1_input = tf.reduce_sum(tf.abs(self.input_layer.w)) self.L2_input = tf.nn.l2_loss(self.input_layer.w) # 隐藏层的权重正则化项 L1s = [] L2_sqrs = [] for i in xrange(n_hidden): L1s.append(tf.reduce_sum(tf.abs(self.hidden_layers[i].w))) L2_sqrs.append(tf.nn.l2_loss(self.hidden_layers[i].w)) L1s.append(tf.reduce_sum(tf.abs(self.softmax_layer.w))) L2_sqrs.append(tf.nn.l2_loss(self.softmax_layer.w)) self.L1 = tf.add_n(L1s) self.L2_sqr = tf.add_n(L2_sqrs) # 包含两个正则化项的成本 self.cost = self.softmax_layer.cost \ + lambda1*(1.0-lambda2)*0.5*self.L2_input + lambda1*lambda2*self.L1_input \ + alpha1*(1.0-alpha2)*0.5 * self.L2_sqr + alpha1*alpha2*self.L1 # 使用FTRL优化器以产生更多的零# self.optimizer = tf.train.FtrlOptimizer(learning_rate=learning_rate).minimize(self.cost) self.optimizer = optimize(self.cost, learning_rate, optimizer) self.accuracy = self.softmax_layer.accuracy self.y = self.softmax_layer.y def train(self, batch_size=100): sess = tf.Session() self.sess = sess sess.run(tf.initialize_all_variables()) for i in xrange(self.epochs): x_batch, y_batch = get_batch(self.X_train, self.y_train, batch_size) sess.run(self.optimizer, feed_dict={self.var_X: x_batch, self.var_Y: y_batch}) if i % 2 == 0: l = sess.run(self.cost, feed_dict={self.var_X: x_batch, self.var_Y: y_batch}) print('epoch {0}: global loss = {1}'.format(i, l)) self.selected_w = sess.run(self.input_layer.w) print("训练准确率:",sess.run(self.accuracy, feed_dict={self.var_X: self.X_train, self.var_Y: self.y_train})) print("测试准确率:",sess.run(self.accuracy, feed_dict={self.var_X: self.X_test, self.var_Y: self.y_test})) print(self.selected_w) print(len(self.selected_w[self.selected_w==0])) print("最终测试准确率:",sess.run(self.accuracy, feed_dict={self.var_X: self.X_test, self.var_Y: self.y_test})) def refine_init_weight(self, threshold=0.001): refined_w = np.copy(self.selected_w) refined_w[refined_w < threshold] = 0 self.input_layer.w.assign(refined_w) print("测试准确率（优化后）:",self.sess.run(self.accuracy, feed_dict={self.var_X: self.X_test, self.var_Y: self.y_test}))回答：
(我只是重新发布我的评论作为答案)
你需要运行你创建的赋值操作，否则它只是被添加到图中而从未执行。
assign_op = self.input_layer.w.assign(refined_w)self.sess.run(assign_op)如果你想在TensorFlow中这样做，你可以使用tf.greater和tf.less创建一个权重变量的布尔掩码，将这个掩码转换为tf.float32，然后与权重数组相乘。
