我必须承认我是初学者，但在StackOverflow或Github中提出的任何问题上，我无法找到问题的答案。请帮忙
错误：
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-33-a9ebf2ee96c3> in <module>()
19 _, c = sess.run([train_op, loss_op], feed_dict={v_u: np.array(onehot_user),
20 v_i: np.array(onehot_item),
---> 21 label:label_list})
22 # Compute average loss
23 avg_cost += c / total_batch
~/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py in run(self, fetches, feed_dict, options, run_metadata)
893 try:
894 result = self._run(None, fetches, feed_dict, options_ptr,
--> 895 run_metadata_ptr)
896 if run_metadata:
897 proto_data = tf_session.TF_GetBuffer(run_metadata_ptr)
~/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py in _run(self, handle, fetches, feed_dict, options, run_metadata)
1091 feed_handles[subfeed_t] = subfeed_val
1092 else:
-> 1093 np_val = np.asarray(subfeed_val, dtype=subfeed_dtype)
1094
1095 if (not is_tensor_handle_feed and
~/anaconda3/lib/python3.6/site-packages/numpy/core/numeric.py in asarray(a, dtype, order)
529
530 """
--> 531 return array(a, dtype, copy=False, order=order)
532
533
ValueError: setting an array element with a sequence.
# Loading data
t1 = time()
dataset = Dataset("/Users/shishir/Documents/botconnect_Playground/NCF_Prototype/Data/ml-1m")
train, testRatings, testNegatives = dataset.trainMatrix, dataset.testRatings, dataset.testNegatives
num_users, num_items = train.shape
print("Load data done [%.1f s]. #user=%d, #item=%d, #train=%d, #test=%d"
%(time()-t1, num_users, num_items, train.nnz, len(testRatings)))
“Train”是一个看起来像的矩阵：[（user_id，item_id）implicit_rating ....]
[ (0, 0) 1.0
(0, 1) 1.0
(0, 2) 1.0
(0, 3) 1.0
(0, 4) 1.0
(0, 5) 1.0
(0, 6) 1.0
(0, 7) 1.0
(0, 8) 1.0
(0, 9) 1.0
(0, 10) 1.0
(0, 11) 1.0
(0, 12) 1.0
(0, 13) 1.0
(0, 14) 1.0
(0, 15) 1.0
(0, 16) 1.0
(0, 17) 1.0
(0, 18) 1.0
(0, 19) 1.0
(0, 20) 1.0
(0, 21) 1.0
(0, 22) 1.0
(0, 23) 1.0
(0, 24) 1.0
: :
(0, 28) 1.0
(0, 29) 1.0
(0, 30) 1.0
(0, 31) 1.0
(0, 32) 1.0
(0, 33) 1.0
(0, 34) 1.0
(0, 35) 1.0
(0, 36) 1.0
(0, 37) 1.0
(0, 38) 1.0
(0, 39) 1.0
(0, 40) 1.0.....]
构建MLP的代码：
with tf.name_scope('MLP'):
# num_layer = [20,10]
# Network Parameters
n_input = 1 #Dimensions of concatenated input vector
n_hidden_1 = 40 # 1st layer number of neurons
n_hidden_2 = 5 # 2nd layer number of neurons
n_out = 1
latent_dim=1
#Input Variables
with tf.name_scope('Input'):
label = tf.placeholder(tf.float32, name='label1')
v_u = tf.placeholder(tf.int32, [1,num_users], name='user_input')
v_i = tf.placeholder(tf.int32, [1,num_items], name='item_input')
with tf.name_scope('Embeddings'):
P_M = tf.truncated_normal([num_users,n_input], stddev=0.1)
tf.Variable(P_M,name="P_M")
embedded_users = tf.matmul(tf.to_float(v_u),P_M)
Q_M = tf.truncated_normal([num_items,n_input], stddev=0.1)
tf.Variable(Q_M,name="Q_M")
embedded_items = tf.matmul( tf.to_float(v_i),Q_M)
# The 0-th layer is the concatenation of embedding layers
concat_vector = tf.concat([embedded_users, embedded_items], 0)
with tf.name_scope("MLP_Layers"):
# Parameters
learning_rate = 0.01
training_epochs = 15
batch_size = 100
display_step = 1
# Store layers weight & bias
weights = {
'h1': tf.Variable(tf.random_normal([n_input, n_hidden_1]), name='Weight1'),
'h2': tf.Variable(tf.random_normal([n_hidden_1, n_hidden_2]), name='Weight2'),
'out': tf.Variable(tf.random_normal([n_hidden_2, n_out]), name='Weight_Output')}
biases = {
'b1': tf.Variable(tf.random_normal([n_hidden_1]), name='Bias1'),
'b2': tf.Variable(tf.random_normal([n_hidden_2]), name='Bias2'),
'out': tf.Variable(tf.random_normal([n_out]), name='Bias_Output')}
with tf.name_scope("Model"):
# Create model
def mlp(x):
layer_1 = tf.add(tf.matmul(x, weights['h1']), biases['b1'], name='Layer1')
layer_1_out = 1/(1 + tf.exp(-layer_1))
layer_2 = tf.add(tf.matmul(layer_1_out, weights['h2']), biases['b2'], name='Layer2')
layer_2_out = 1/(1 + tf.exp(-layer_2))
final_layer = tf.add(tf.matmul(layer_2, weights['out']),biases['out'], name='Output_Layer')
out_layer = 1/(1 + tf.exp(-final_layer))
return out_layer
从训练矩阵构建输入的代码：
def get_training_data(train, num_negatives):
user_input, item_input, labels = [],[],[]
num_users = train.shape[0]
for (u, i) in train.keys():
# positive instance
user_input.append(u)
item_input.append(i)
labels.append(1)
# negative instances
for t in range(num_negatives):
j = np.random.randint(num_items)
while (u,j) in train.keys():
j = np.random.randint(num_items)
user_input.append(u)
item_input.append(j)
labels.append(0)
return user_input, item_input, labels
构建模型：
# Build model
prediction = mlp(concat_vector)
init = tf.global_variables_initializer()
# Define loss and optimizer
with tf.name_scope('cross_entropy'):
loss_op = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=prediction, labels=label))
tf.summary.scalar('cross_entropy', loss_op)
with tf.name_scope('train'):
optimizer = tf.train.AdamOptimizer(learning_rate=0.15)
train_op = optimizer.minimize(loss_op)
# Initializing the variables
init = tf.global_variables_initializer()
最后培训（导致错误的代码）
num_epochs=1
num_negatives=3
# Training model
with tf.Session() as sess:
for epoch in range(num_epochs):
t1 = time()
# Generate training instances
user_input, item_input, labels = get_training_data(train, num_negatives)
onehot_user = (tf.one_hot(user_input,len(user_input)))
onehot_item = (tf.one_hot(item_input,len(item_input)))
label_list = np.array(labels)
total_batch = int(len(user_input))
# Training
for i in range(total_batch):
# Run optimization op (backprop) and cost op (to get loss value)
_, c = sess.run([train_op, loss_op], feed_dict={v_u: np.array(onehot_user),
v_i: np.array(onehot_item),
label:label_list})
# Compute average loss
avg_cost += c / total_batch
print("Batch:", '%04d' % (i),"Completed")
# Display logs per epoch step
if epoch % display_step == 0:
print("Epoch:", '%04d' % (epoch+1), "cost={:.9f}".format(avg_cost))
print("Optimization Finished!")
t2 = time()
merged = tf.summary.merge_all()
writer = tf.summary.FileWriter('logs', sess.graph)
任何调试的想法??
提前谢谢！
答案 0 :(得分：0)
您的label_list变量是一个整数数组，但label占位符需要tf.float32个值。如果你切换其中一种类型，它应该可以正常工作。