Python 3.x Tensorflow-如何获得预测函数结果的准确性？

我对Tensorflow完全陌生，我正在努力组建一支军队，这是我的第一个CNN

我最及时的问题是，我如何才能得到这个预测结果的准确性；我正在使用此功能：

cnn.predict (x)

我一直在研究，但还没有得到结果，你能指导我吗

为了更好地参考，以下是我的代码：

我的班级“分类”的代码是：

创建模型的代码如下所示：

from __future__ import absolute_import, division, print_function, unicode_literals
import sys
import os
from tensorflow.python.keras import backend as K
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Conv2D, Flatten, Dropout, MaxPooling2D
from tensorflow.keras.preprocessing.image import ImageDataGenerator

K.clear_session()

data_entrenamiento = './data/train'
data_validacion = './data/validation'


epocas=20
longitud, altura = 224, 224
batch_size = 32
pasos = 10
validation_steps = 3
filtrosConv1 = 32
filtrosConv2 = 64
tamano_filtro1 = (3, 3)
tamano_filtro2 = (2, 2)
tamano_pool = (2, 2)
clases = 2
lr = 0.0004
IMG_SIZE = 224
IMG_SHAPE = (longitud, altura, 3)

class myCallback(tf.keras.callbacks.Callback):
  def on_epoch_end(self, epoch, logs={}):
    if(logs.get('accuracy')>0.959):
      print("¡Alcanzó una precisión del 99% y canceló el entrenamiento!")
      self.model.stop_training = True

def modelo():
    vgg = tf.keras.applications.VGG16()
    vgg.summary()
    cnn=Sequential()
    for capa in vgg.layers:
         cnn.add(capa)
    cnn.pop()

    for layer in cnn.layers:
        layer.trainable=False
    cnn.add(Dense(2,activation='softmax'))   
    cnn.summary() 
    return cnn

def modelo():
    vgg=tf.keras.applications.VGG16()
    cnn=Sequential()
    for capa in vgg.layers:
        cnn.add(capa)
    cnn.pop()

    for layer in cnn.layers:
        layer.trainable=False
    cnn.add(Dense(2,activation='softmax'))
    cnn.summary()
    return cnn

entrenamiento_datagen = ImageDataGenerator(
    rescale=1. / 255)
    #shear_range=0.2,
    #zoom_range=0.2,
    #horizontal_flip=True)

test_datagen = ImageDataGenerator(rescale=1. / 255)

entrenamiento_generador = entrenamiento_datagen.flow_from_directory(
    data_entrenamiento,
    target_size=(altura, longitud),
    batch_size=batch_size,
    class_mode='categorical')

validacion_generador = test_datagen.flow_from_directory(
    data_validacion,
    target_size=(altura, longitud),
    batch_size=batch_size,
    class_mode='categorical')

cnn=modelo()

cnn.compile(loss='categorical_crossentropy',
            optimizer='adam',
            metrics=['accuracy'])

callbacks = myCallback()
history = cnn.fit_generator(
    entrenamiento_generador,
    steps_per_epoch=pasos,
    epochs=epocas,
    validation_data=validacion_generador,
    validation_steps=validation_steps,
    callbacks=[callbacks])

print('classes model: ', entrenamiento_generador.class_indices)

target_dir = './modelo/'
if not os.path.exists(target_dir):
  os.mkdir(target_dir)
cnn.save('./modelo/model.h5')
cnn.save_weights('./modelo/model_weights.h5')

---

我提前感谢您对我的模型构造提出的任何建议，如果我能以某种方式对其进行改进或修改，目标是对文档进行分类，到目前为止，通过这种配置，我获得了令人满意的结果。

只需使用

cnn.evaluate（x，y）

获取指标。但我如何定义“Y”或如何获取它？

Y

是基本事实。如果没有

y

，则无法计算精度

from __future__ import absolute_import, division, print_function, unicode_literals
import sys
import os
from tensorflow.python.keras import backend as K
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Conv2D, Flatten, Dropout, MaxPooling2D
from tensorflow.keras.preprocessing.image import ImageDataGenerator

K.clear_session()

data_entrenamiento = './data/train'
data_validacion = './data/validation'


epocas=20
longitud, altura = 224, 224
batch_size = 32
pasos = 10
validation_steps = 3
filtrosConv1 = 32
filtrosConv2 = 64
tamano_filtro1 = (3, 3)
tamano_filtro2 = (2, 2)
tamano_pool = (2, 2)
clases = 2
lr = 0.0004
IMG_SIZE = 224
IMG_SHAPE = (longitud, altura, 3)

class myCallback(tf.keras.callbacks.Callback):
  def on_epoch_end(self, epoch, logs={}):
    if(logs.get('accuracy')>0.959):
      print("¡Alcanzó una precisión del 99% y canceló el entrenamiento!")
      self.model.stop_training = True

def modelo():
    vgg = tf.keras.applications.VGG16()
    vgg.summary()
    cnn=Sequential()
    for capa in vgg.layers:
         cnn.add(capa)
    cnn.pop()

    for layer in cnn.layers:
        layer.trainable=False
    cnn.add(Dense(2,activation='softmax'))   
    cnn.summary() 
    return cnn

def modelo():
    vgg=tf.keras.applications.VGG16()
    cnn=Sequential()
    for capa in vgg.layers:
        cnn.add(capa)
    cnn.pop()

    for layer in cnn.layers:
        layer.trainable=False
    cnn.add(Dense(2,activation='softmax'))
    cnn.summary()
    return cnn

entrenamiento_datagen = ImageDataGenerator(
    rescale=1. / 255)
    #shear_range=0.2,
    #zoom_range=0.2,
    #horizontal_flip=True)

test_datagen = ImageDataGenerator(rescale=1. / 255)

entrenamiento_generador = entrenamiento_datagen.flow_from_directory(
    data_entrenamiento,
    target_size=(altura, longitud),
    batch_size=batch_size,
    class_mode='categorical')

validacion_generador = test_datagen.flow_from_directory(
    data_validacion,
    target_size=(altura, longitud),
    batch_size=batch_size,
    class_mode='categorical')

cnn=modelo()

cnn.compile(loss='categorical_crossentropy',
            optimizer='adam',
            metrics=['accuracy'])

callbacks = myCallback()
history = cnn.fit_generator(
    entrenamiento_generador,
    steps_per_epoch=pasos,
    epochs=epocas,
    validation_data=validacion_generador,
    validation_steps=validation_steps,
    callbacks=[callbacks])

print('classes model: ', entrenamiento_generador.class_indices)

target_dir = './modelo/'
if not os.path.exists(target_dir):
  os.mkdir(target_dir)
cnn.save('./modelo/model.h5')
cnn.save_weights('./modelo/model_weights.h5')