Python 我的Keras序列模型的精确度为0

我正在尝试建立一个模型来预测英超比赛的结果。当尝试这样做时，该模型在训练时给我50-66的准确度，在使用一些测试数据进行测试时给我0的准确度。 我刚开始使用keras和tensorflow，所以我为错误的代码道歉

from tensorflow import keras
import numpy as np
import pandas as pd
from tensorflow.python.keras.layers.core import Dropout

x_train = np.random.randint(0, 43, size=(4000, 2))
x_test = np.random.randint(0, 43, size=(2000, 2))
y_train = np.random.randint(0, 2, size=(4000))
y_test = np.random.randint(0, 2, size=(2000))

model = keras.models.Sequential(
    [
        keras.layers.Embedding(5000, 43),
        keras.layers.Conv1D(
            filters=43, kernel_size=3, padding="same", activation="relu"
        ),
        keras.layers.MaxPooling1D(pool_size=2),
        keras.layers.LSTM(100),
        keras.layers.Dropout(0.2),
        keras.layers.Dense(1, activation="sigmoid"),
    ]
)

model.compile(optimizer="adam", loss="binary_crossentropy", metrics=["accuracy"])
model.fit(x_train, y_train, epochs=5)
acc = model.evaluate(x_test)
print(acc)

这是输出：

----

Epoch 1/5
125/125 [==============================] - 1s 4ms/step - loss: 0.6716 - accuracy: 0.5663

Epoch 2/5
125/125 [==============================] - 1s 4ms/step - loss: 0.6320 - accuracy: 0.6357

Epoch 3/5
125/125 [==============================] - 1s 4ms/step - loss: 0.6244 - accuracy: 0.6457

Epoch 4/5
125/125 [==============================] - 1s 5ms/step - loss: 0.6208 - accuracy: 0.6578

Epoch 5/5
125/125 [==============================] - 1s 6ms/step - loss: 0.6183 - accuracy: 0.6585

**89/89 [==============================] - 0s 1ms/step - loss: 0.0000e+00 - accuracy: 0.0000e+00**

**[0.0, 0.0]**

---

您需要为评估提供目标数据。i、 e

acc = model.evaluate(x_test,y_test)

---

如果没有相应的数据，则无法预制作。至少先尝试使用随机生成的输入或公开可用的数据集来重现这一点。好的，再重复一遍