correct_prediction = tf.equal(predictions["classes"], tf.argmax(y, axis=1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
TP = tf.metrics.true_positives(labels=tf.argmax(y, axis=1), predictions=predictions["classes"])
FP = tf.metrics.false_positives(labels=tf.argmax(y, axis=1), predictions=predictions["classes"])
TN = tf.metrics.true_negatives(labels=tf.argmax(y, axis=1), predictions=predictions["classes"])
FN = tf.metrics.false_negatives(labels=tf.argmax(y, axis=1), predictions=predictions["classes"])
recall = tf.metrics.recall(labels=tf.argmax(y, axis=1), predictions=predictions["classes"])
tf_accuracy = tf.metrics.accuracy(labels=tf.argmax(y, axis=1), predictions=predictions["classes"])
# ------------------------------------------------------------------------------------------
# ------------------------------ train model ------------------------------
mydata_train = DataSet(data_train, label_train)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
accuracys=[]
begin_time = datetime.now()
for i in range(train_iter):
# print('iter:', i)
batch = mydata_train.next_batch(128) # 此处的batch是由[128,1600]和[128,]组成的tuple，batch[0]就是tuple
labels = labels_transform(batch[1], classes_num)
if (i + 1) % 20 == 0:
train_accuracy = sess.run(accuracy, feed_dict={_X: batch[0], y: labels,
keep_prob: 1.0, batch_size: _batch_size})
accuracys.append(train_accuracy)
print("\nthe %dth loop,training accuracy:%f" % (i + 1, train_accuracy))
predictions=predictions["classes"])
# ------------------------------------------------------------------------------------------
# ------------------------------ train model ------------------------------
print("\n" + "=" * 50 + "Benign Trainging" + "=" * 50)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
_batch_size = 128
mydata_train = DataSet(data_train, label_train)
statr = time.time()
accuracys = []
begin_time = datetime.now()
for i in range(train_iter):
batch = mydata_train.next_batch(_batch_size)
labels = labels_transform(batch[1], class_num)
if (i + 1) % 20 == 0:
train_accuracy = sess.run(accuracy,
feed_dict={
_X: batch[0],
y: labels,
keep_prob: 1.0,
batch_size: _batch_size
})
print("\n" + "=" * 50 + "Benign Trainging" + "=" * 50)
# config = tf.ConfigProto()
# config.gpu_options.allow_growth = True
#
# sess = tf.Session(config=config)
##### errors_impl.InternalError: Failed to create session. 看看这里为什么报错
sess = tf.Session()
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
_batch_size = 128
mydata_train = DataSet(data_train, label_train)
accuracys = []
begin_time = datetime.now()
for i in range(train_iter):
batch = mydata_train.next_batch(_batch_size)
labels = labels_transform(batch[1], class_num)
if (i + 1) % 100 == 0:
train_accuracy = sess.run(accuracy,
feed_dict={
_X: batch[0],
y: labels,
keep_prob: 1.0,
batch_size: _batch_size
})