在TensorFlow中，我想创建一个逻辑回归模型，代价函数如下：

使用的数据集截图如下：

我的代码如下：

train_X = train_data[:, :-1] train_y = train_data[:, -1:] feature_num = len(train_X[0]) sample_num = len(train_X) print("Size of train_X: {}x{}".format(sample_num, feature_num)) print("Size of train_y: {}x{}".format(len(train_y), len(train_y[0]))) X = tf.placeholder(tf.float32) y = tf.placeholder(tf.float32) W = tf.Variable(tf.zeros([feature_num, 1])) b = tf.Variable([-.3]) db = tf.matmul(X, tf.reshape(W, [-1, 1])) + b hyp = tf.sigmoid(db) cost0 = y * tf.log(hyp) cost1 = (1 - y) * tf.log(1 - hyp) cost = (cost0 + cost1) / -sample_num loss = tf.reduce_sum(cost) optimizer = tf.train.GradientDescentOptimizer(0.1) train = optimizer.minimize(loss) init = tf.global_variables_initializer() sess = tf.Session() sess.run(init) print(0, sess.run(W).flatten(), sess.run(b).flatten()) sess.run(train, {X: train_X, y: train_y}) print(1, sess.run(W).flatten(), sess.run(b).flatten()) sess.run(train, {X: train_X, y: train_y}) print(2, sess.run(W).flatten(), sess.run(b).flatten())

运行结果截图如下：

可以看到，在迭代两次之后，得到的W和b都变成了nan，请问是哪里的问题？