import tensorflow as tf
import pandas as pd
import os
# 环境变量配置
os.environ['TF_XLA_FLAGS']='--tf_xla_enable_xla_devices'
os.environ['TF_FORCE_GPU_ALLOW_GROWTH']='true'# 数据的读取
data = pd.read_csv(r'dataset/dataset.csv')
x = data.iloc[:,1:-1]
y = data.iloc[:,-1]# 模型的建立
model = tf.keras.Sequential([
tf.keras.layers.Dense(10, input_shape=(3,), activation='relu'),
tf.keras.layers.Dense(1)])# 模型的相关配置
model.compile(
optimizer='adam',
loss='mse')# 模型的训练
history = model.fit(x, y, epochs=10000)# 模型的保存
model.save(r'model_data/model.h5')
四、预测代码如下:
import tensorflow as tf
import pandas as pd
import os
# 环境变量配置
os.environ['TF_XLA_FLAGS']='--tf_xla_enable_xla_devices'
os.environ['TF_FORCE_GPU_ALLOW_GROWTH']='true'# 数据的读取
data = pd.read_csv(r'dataset/dataset.csv')
test = data.iloc[:10,1:-1]
test_label = data.iloc[:10,-1]# 模型的构建
model = tf.keras.models.load_model(r'model_data/model.h5')# 模型的预测
pre_y = model.predict(test)print(pre_y)print(test_label)