当我将 makeModel 函数中的输入数量从 3 改为 1 时，程序可以无错误运行，但实际上没有进行训练，准确率也没有变化。
import pandas as pdfrom numpy import loadtxtfrom sklearn.impute import SimpleImputerfrom sklearn.model_selection import train_test_splitfrom tensorflow.keras.models import Sequentialfrom tensorflow.keras.layers import Densefrom sklearn.preprocessing import LabelEncoderfrom sklearn.utils import shufflefrom sklearn.tree import DecisionTreeRegressor as dtrfrom sklearn.metrics import mean_absolute_error as maeimport numpy as npdef makeModel(num_inputs, num_classes, train_X, train_y): model = Sequential() model.add(Dense(8, input_dim=num_inputs, activation='relu')) model.add(Dense(10, activation='relu')) model.add(Dense(10, activation='relu')) model.add(Dense(10, activation='relu')) model.add(Dense(3, activation='softmax')) model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy']) model.fit(train_X, train_y, epochs=10, batch_size=10) return modellabel_encoder = LabelEncoder()iris_data = pd.read_csv("iris.csv")iris_data = shuffle(iris_data)iris_data['species'] = label_encoder.fit_transform(iris_data['species'])feature_columns = ['sepal_length', 'sepal_width', 'petal_length', 'petal_width']X = iris_data[feature_columns]y = iris_data['species']train_x, val_x, train_y, val_y = train_test_split(X, y, test_size=0.2)iris_model = makeModel(4, 3, train_x, train_y)回答：
LabelEncoder 将输入转换为编码值数组。例如，如果你的输入是 ["paris", "paris", "tokyo", "amsterdam"]，它们可以被编码为 [0, 0, 1, 2]。这不是 categorical_crossentropy 损失函数所期望的一热编码方案。如果你使用的是整数编码，你需要使用 sparse_categorical_crossentropy。
解决方法
将你的代码中的损失函数改为 sparse_categorical_crossentropy ：
model.compile(loss='sparse_categorical_crossentropy', optimizer='adam', metrics=['accuracy'])示例
def makeModel(num_inputs, num_classes, train_X, train_y): model = Sequential() model.add(Dense(8, input_dim=num_inputs, activation='relu')) model.add(Dense(10, activation='relu')) model.add(Dense(10, activation='relu')) model.add(Dense(10, activation='relu')) model.add(Dense(3, activation='softmax')) model.compile(loss='sparse_categorical_crossentropy', optimizer='adam', metrics=['accuracy']) model.fit(train_X, train_y, epochs=10, batch_size=10) return modellabel_encoder = LabelEncoder()iris = datasets.load_iris()y = iris.targety = label_encoder.fit_transform(y)train_x, val_x, train_y, val_y = train_test_split(iris.data, y, test_size=0.2)iris_model = makeModel(4, 3, train_x, train_y)