Python Keras中由列表组成的列表的正确输入形状是什么？

我有一个由10个数字组成的小列表。像

[[1,2,3,4,5,6,7,8,9,10],[0,0,0,0,0,0,0,0,0,0],[1,1,1....]].

我应该将什么形状应用于我的模型？以下是我目前掌握的代码：

''' 将open（'games_from_file_1.json'，'r'）作为f： data1=json.load（f）

'''

我得到了50%的准确率。。。所以我想可能是因为它的形状

---

我也有65000个较小的列表，我想用于培训，但当我运行该脚本时，在epochs文本下会出现：2032/2032。。。不应该有65000/65000吗？

默认情况下，培训分32批进行，由model.fit方法的batch_size参数控制。因此2032/2032是正确的（65000/32）。

with open('games_from_file_3.json', 'r') as f:
  data3 = json.load(f)

data = {}
data['games'] = data1['games']  + data3['games']
pprint(data['games'][0])

final_data = pop_or_add_afk(data)

train_samples = []
train_labels = []


for i in range(65000):

  temp_list = [None] * 10

  temp_list[0] = final_data['games'][i]['team_1']['Top']
  temp_list[1] = final_data['games'][i]['team_1']['Mid']
  temp_list[2] = final_data['games'][i]['team_1']['Jg']
  temp_list[3] = final_data['games'][i]['team_1']['ADC']
  temp_list[4] = final_data['games'][i]['team_1']['Sup']

  temp_list[5] = final_data['games'][i]['team_2']['Top']
  temp_list[6] = final_data['games'][i]['team_2']['Mid']
  temp_list[7] = final_data['games'][i]['team_2']['Jg']
  temp_list[8] = final_data['games'][i]['team_2']['ADC']
  temp_list[9] = final_data['games'][i]['team_2']['Sup']


  train_samples.append(temp_list)
  train_labels.append(final_data['games'][i]['win'])

train_labels = np.array(train_labels)
train_samples = np.array(train_samples)

model = keras.Sequential([
    keras.layers.Dense(units=10, input_shape=(65000,10), activation = 'relu'), 
    keras.layers.Dense(units=10, activation='relu'),  # hidden layer (2)
    keras.layers.Dense(units=3, activation='softmax')
])

model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

model.fit(train_samples, train_labels, epochs = 30)