def Normalization(data): data_mean = data.mean() data_std = data.std() data = data - data_mean data = data / data_std return data
时间: 2023-10-08 10:09:40 浏览: 57
这段代码实现了数据标准化(Normalization)的功能,具体解释如下:
- 输入参数 data 是一个 numpy 数组,用于存储需要进行标准化的数据。
- data_mean = data.mean():计算数据的均值。
- data_std = data.std():计算数据的标准差。
- data = data - data_mean:将数据减去均值,使得数据的中心在 0 附近。
- data = data / data_std:将数据除以标准差,使得数据的范围在 -1 到 1 之间。
- 返回结果数据 data,它的每个元素都是标准化后的数值。
相关问题
arr0 = np.array([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24]) arr1 = np.array([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24]) arr3 = np.array(input("请输入连续24个月的配件销售数据,元素之间用空格隔开:").split(), dtype=float) data_array = np.vstack((arr1, arr3)) data_matrix = data_array.T data = pd.DataFrame(data_matrix, columns=['month', 'sales']) sales = data['sales'].values.astype(np.float32) sales_mean = sales.mean() sales_std = sales.std() sales = abs(sales - sales_mean) / sales_std train_data = sales[:-1] test_data = sales[-12:] def create_model(): model = tf.keras.Sequential() model.add(layers.Input(shape=(11, 1))) model.add(layers.Conv1D(filters=32, kernel_size=2, padding='causal', activation='relu')) model.add(layers.BatchNormalization()) model.add(layers.Conv1D(filters=64, kernel_size=2, padding='causal', activation='relu')) model.add(layers.BatchNormalization()) model.add(layers.Conv1D(filters=128, kernel_size=2, padding='causal', activation='relu')) model.add(layers.BatchNormalization()) model.add(layers.Conv1D(filters=256, kernel_size=2, padding='causal', activation='relu')) model.add(layers.BatchNormalization()) model.add(layers.Conv1D(filters=512, kernel_size=2, padding='causal', activation='relu')) model.add(layers.BatchNormalization()) model.add(layers.Dense(1, activation='linear')) return model model = create_model() BATCH_SIZE = 16 BUFFER_SIZE = 100 train_dataset = tf.data.Dataset.from_tensor_slices(train_data) train_dataset = train_dataset.window(11, shift=1, drop_remainder=True) train_dataset = train_dataset.flat_map(lambda window: window.batch(11)) train_dataset = train_dataset.map(lambda window: (window[:-1], window[-1:])) train_dataset = train_dataset.shuffle(BUFFER_SIZE).batch(BATCH_SIZE).prefetch(1) model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.001), loss='mse') history = model.fit(train_dataset, epochs=100, verbose=0) test_input = test_data[:-1] test_input = np.reshape(test_input, (1, 11, 1)) predicted_sales = model.predict(test_input)[0][0] * sales_std + sales_mean test_prediction = model.predict(test_input) y_test=test_data[1:12] y_pred=test_prediction y_pred = test_prediction.ravel() print("预测下一个月的销量为:", predicted_sales),如何将以下代码稍作修改插入到上面的最后,def comput_acc(real,predict,level): num_error=0 for i in range(len(real)): if abs(real[i]-predict[i])/real[i]>level: num_error+=1 return 1-num_error/len(real) a=np.array(test_data[label]) real_y=a real_predict=test_predict print("置信水平:{},预测准确率:{}".format(0.2,round(comput_acc(real_y,real_predict,0.2)* 100,2)),"%")
你可以在最后添加如下代码实现 `comput_acc()` 函数的功能:
```
label = 0 # 定义标签
a = np.array(test_data[label]) # 获取测试数据的标签列
real_y = a[1:] # 实际销售数据
real_predict = y_pred[:-1] # 预测销售数据
print("置信水平:{},预测准确率:{}".format(0.2, round(comput_acc(real_y, real_predict, 0.2) * 100, 2)), "%")
```
这段代码的作用是,首先定义标签为 0,即选取测试数据中的第一列数据(即销售数据)。然后,使用 `np.array()` 函数将该列数据转换为 numpy 数组 `a`。接着,将 `a` 中的第二个元素到最后一个元素赋值给 `real_y`,这里是因为真实销售数据的第一个元素已经作为测试数据输入了模型,所以预测值中不包含该元素。然后,将 `y_pred` 中的第一个元素到倒数第二个元素赋值给 `real_predict`,这是因为预测值中的最后一个元素已经与真实值的最后一个元素相对应。最后,调用 `comput_acc()` 函数计算预测准确率,并将结果打印输出。
import matplotlib.pyplot as plt import pandas as pd from keras.models import Sequential from keras import layers from keras import regularizers import os import keras import keras.backend as K import numpy as np from keras.callbacks import LearningRateScheduler data = "data.csv" df = pd.read_csv(data, header=0, index_col=0) df1 = df.drop(["y"], axis=1) lbls = df["y"].values - 1 wave = np.zeros((11500, 178)) z = 0 for index, row in df1.iterrows(): wave[z, :] = row z+=1 mean = wave.mean(axis=0) wave -= mean std = wave.std(axis=0) wave /= std def one_hot(y): lbl = np.zeros(5) lbl[y] = 1 return lbl target = [] for value in lbls: target.append(one_hot(value)) target = np.array(target) wave = np.expand_dims(wave, axis=-1) model = Sequential() model.add(layers.Conv1D(64, 15, strides=2, input_shape=(178, 1), use_bias=False)) model.add(layers.ReLU()) model.add(layers.Conv1D(64, 3)) model.add(layers.Conv1D(64, 3, strides=2)) model.add(layers.BatchNormalization()) model.add(layers.Dropout(0.5)) model.add(layers.Conv1D(64, 3)) model.add(layers.Conv1D(64, 3, strides=2)) model.add(layers.BatchNormalization()) model.add(layers.LSTM(64, dropout=0.5, return_sequences=True)) model.add(layers.LSTM(64, dropout=0.5, return_sequences=True)) model.add(layers.LSTM(32)) model.add(layers.Dropout(0.5)) model.add(layers.Dense(5, activation="softmax")) model.summary() save_path = './keras_model3.h5' if os.path.isfile(save_path): model.load_weights(save_path) print('reloaded.') adam = keras.optimizers.adam() model.compile(optimizer=adam, loss="categorical_crossentropy", metrics=["acc"]) # 计算学习率 def lr_scheduler(epoch): # 每隔100个epoch,学习率减小为原来的0.5 if epoch % 100 == 0 and epoch != 0: lr = K.get_value(model.optimizer.lr) K.set_value(model.optimizer.lr, lr * 0.5) print("lr changed to {}".format(lr * 0.5)) return K.get_value(model.optimizer.lr) lrate = LearningRateScheduler(lr_scheduler) history = model.fit(wave, target, epochs=400, batch_size=128, validation_split=0.2, verbose=2, callbacks=[lrate]) model.save_weights(save_path) print(history.history.keys()) # summarize history for accuracy plt.plot(history.history['acc']) plt.plot(history.history['val_acc']) plt.title('model accuracy') plt.ylabel('accuracy') plt.xlabel('epoch') plt.legend(['train', 'test'], loc='upper left') plt.show() # summarize history for loss plt.plot(history.history['loss']) plt.plot(history.history['val_loss']) plt.title('model loss') plt.ylabel('loss') plt.xlabel('epoch') plt.legend(['train', 'test'], loc='upper left') plt.show()
这段代码是用来训练一个深度学习模型的代码,使用了Keras和Tensorflow进行模型的搭建、训练和保存。该模型包含了卷积层、批量归一化层、Dropout层、LSTM层和全连接层,并且最终使用softmax激活函数得到预测的类别。通过对数据进行处理,将输入数据按照一定的规范进行规范化,将标签转换成了一个one-hot编码的形式。最后,该代码通过在数据上进行训练,得到一个可以用来做预测的深度学习模型。