ValueError：找不到可以处理 <numpy.ndarray> 的数据适配器

Question

我正在使用以下代码：

import tensorflow as tf
import numpy as np
in = np.array([10,20,30,40,50,60,70],dtype=float)
out = np.array([50,68,86,104,122,140,158],dtype=float)
ly1 = tf.keras.layers.Dense(units=3, input_shape=[1])
ly2 = tf.keras.layers.Dense(units=3)
result = tf.keras.layers.Dense(units=1)
model = tf.keras.models.Sequential([ly1,ly2,result])
model.compile(optimizer=tf.keras.optimizers.Adam(0.1),loss='mean_squared_error') 

错误出现在这一行：

train = model.fit(in, out, epochs=1000, verbose=3)

Answer 1

将变量in重命名为in1后，它开始工作，如下所示

import tensorflow as tf
print(tf.__version__)
import numpy as np
print(np.__version__)

in1 = np.array([10,20,30,40,50,60,70],dtype=float)
out = np.array([50,68,86,104,122,140,158],dtype=float)
ly1 = tf.keras.layers.Dense(units=3, input_shape=[1])
ly2 = tf.keras.layers.Dense(units=3)
result = tf.keras.layers.Dense(units=1)
model = tf.keras.models.Sequential([ly1,ly2,result])
model.compile(optimizer=tf.keras.optimizers.Adam(0.1),loss='mean_squared_error') 
train = model.fit(in1, out, epochs=10, verbose=1)

输出：

2.5.0
1.19.5
Epoch 1/10
1/1 [==============================] - 1s 553ms/step - loss: 4867.3154
Epoch 2/10
1/1 [==============================] - 0s 6ms/step - loss: 1008.0177
Epoch 3/10
1/1 [==============================] - 0s 5ms/step - loss: 454.7526
Epoch 4/10
1/1 [==============================] - 0s 4ms/step - loss: 1876.2919
Epoch 5/10
1/1 [==============================] - 0s 4ms/step - loss: 1345.9777
Epoch 6/10
1/1 [==============================] - 0s 4ms/step - loss: 392.1256
Epoch 7/10
1/1 [==============================] - 0s 6ms/step - loss: 235.7297
Epoch 8/10
1/1 [==============================] - 0s 5ms/step - loss: 699.6002
Epoch 9/10
1/1 [==============================] - 0s 5ms/step - loss: 981.0598
Epoch 10/10
1/1 [==============================] - 0s 7ms/step - loss: 822.2618

注意：我们不应该使用系统定义的关键字作为变量名