From 1aa5d08034160f1f6b551c92f71e6bca8d2c3e33 Mon Sep 17 00:00:00 2001 From: MuJ Date: Tue, 9 Jan 2024 20:08:48 +0800 Subject: [PATCH] =?UTF-8?q?=E9=AA=8C=E8=AF=81=E5=8F=AF=E8=83=BD=E6=98=AF?= =?UTF-8?q?=E6=95=B0=E6=8D=AE=E9=9B=86=E5=AD=98=E5=9C=A8=E9=97=AE=E9=A2=98?= =?UTF-8?q?=EF=BC=8C=E5=87=86=E5=A4=87=E5=A4=A7=E6=94=B9?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- main.py | 90 ++++++++++++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 76 insertions(+), 14 deletions(-) diff --git a/main.py b/main.py index eb5262f..b219909 100644 --- a/main.py +++ b/main.py @@ -1,14 +1,20 @@ import pandas as pd import numpy as np +import os +import warnings +warnings.filterwarnings('ignore') import tensorflow as tf +tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.ERROR) import matplotlib.pyplot as plt from keras import regularizers from keras.callbacks import TensorBoard, LearningRateScheduler from keras.layers import Dropout from sklearn.preprocessing import MinMaxScaler +from tqdm import tqdm def data_read(data_address): + df = pd.read_csv(data_address) label_mapping = {label: idx for idx, label in enumerate(df['Problem'].unique())} @@ -20,8 +26,6 @@ def data_read(data_address): X = np.array(df['Voltage']) Y = np.array(df['Problem']) - X = tf.nn.relu(X) - # 转换为时间序列数据格式 time_steps = 34 X_series, Y_series = [], [] @@ -32,20 +36,72 @@ def data_read(data_address): return np.array(X_series), np.array(Y_series) +# 编写绘图函数，画出训练集电压数据 +def plant_for_voltage(x_train_new, x_train_orginal, gamma, learning_rate, y_train, different_location): + # 绘制X_train的图形 + for i in different_location: + if i % 100 == 0: + plt.figure() + time_Step = list(range(0, 34)) + plt.plot(time_Step, + x_train_new[i]) + # 添加标题和标签 + plt.title(f'gamma:{gamma},learning_rate:{learning_rate},Y:{y_train[i]}') + plt.xlabel('time_step') + plt.ylabel('voltage') + try: + os.makedirs(f'Liu/picture/gamma{gamma} learningrate{learning_rate}') + except FileExistsError: + pass + plt.savefig(f'Liu/picture/gamma{gamma} learningrate{learning_rate}/X_train_new_{i}.png') + plt.close() + plt.clf() + + # 画出原始图像 + plt.figure() + time_Step = list(range(0, 34)) + plt.plot(time_Step, + x_train_orginal[i]) + # 添加标题和标签 + plt.title(f'gamma:{gamma},learning_rate:{learning_rate},Y:{y_train[i]}') + plt.xlabel('time_step') + plt.ylabel('voltage') + try: + os.makedirs(f'Liu/picture/gamma{gamma} learningrate{learning_rate}') + except FileExistsError: + pass + plt.savefig(f'Liu/picture/gamma{gamma} learningrate{learning_rate}/X_train_{i}.png') + plt.close() + plt.clf() + + + +def if_diff(a, b): + # 比较两个数组相同位置上的元素是否相等 + diff = np.where(a != b) + + list_diff = [] + + # 打印不同元素的索引及其对应的元素 + for i in range(len(diff[0])): + idx = (diff[0][i], diff[1][i]) + list_diff.append(idx[0]) + return list_diff + X_train, Y_train = data_read( 'Liu\data\VOLTAGE-QUALITY-CLASSIFICATION-MODEL--main\Voltage Quality.csv') X_test, Y_test = data_read( 'Liu/data/VOLTAGE-QUALITY-CLASSIFICATION-MODEL--main/Voltage Quality Test.csv') -# 归一化 -sc = MinMaxScaler(feature_range=(0, 1)) +# 初始化归一化模型 +sc = MinMaxScaler(feature_range=(-1, 1)) X_train = sc.fit_transform(X_train) X_test = sc.transform(X_test) + X_train.reshape(-1, 34, 1) X_test.reshape(-1, 34, 1) - np.random.seed(7) np.random.shuffle(X_train) np.random.seed(7) @@ -66,7 +122,7 @@ model = tf.keras.models.Sequential([ tf.keras.layers.SimpleRNN(100), Dropout(0.2), tf.keras.layers.Dense(n_classes, activation='relu', - kernel_regularizer=regularizers.l2(0.3)) # 适应多分类 + ) # kernel_regularizer=regularizers.l2(0.3) ]) # 编译模型 @@ -94,6 +150,7 @@ initial_weights = model.get_weights() log_dir = "logs/fit" tensorboard_callback = TensorBoard(log_dir=log_dir, histogram_freq=1) + # 制作扰动数据 # 转换X_train和Y_train为TensorFlow张量 @@ -119,7 +176,7 @@ accuracy_per_gamma = {} flattened_gradients = tf.reshape(gradients, [-1]) # 选择最大的γ * |X|个梯度 -for gamma in [0.05, 0.1, 0.2, 0.4]: +for gamma in tqdm([0.6, 0.7, 0.8, 0.9, 0.99]): num_gradients_to_select = int( gamma * tf.size(flattened_gradients, out_type=tf.dtypes.float32)) top_gradients_indices = tf.argsort(flattened_gradients, direction='DESCENDING')[ @@ -143,24 +200,29 @@ for gamma in [0.05, 0.1, 0.2, 0.4]: # 创建准确率列表 accuracy_list = [] - for learning_rate in [0.1, 0.2, 0.3, 0.4, 0.5]: + for learning_rate in tqdm([0.1, 0.2, 0.3, 0.4, 0.5]): # 应用学习率到梯度 - scaled_gradients = learning_rate * updated_gradients - + scaled_gradients = (learning_rate * 100) * updated_gradients # 使用缩放后的梯度更新X_train_tensor X_train_updated = tf.add(X_train_tensor, scaled_gradients) X_train_updated = X_train_updated.numpy() - + + list_diff = if_diff(X_train, X_train_updated) + + # 显示扰动数据和原始数据的可视化图像 + # plant_for_voltage(X_train_updated, X_train, gamma, learning_rate, Y_train, list_diff) + # Reset model weights to initial weights model.set_weights(initial_weights) # 训练模型，添加 TensorBoard 回调 - history = model.fit(X_train_updated, Y_train, epochs=1500, - batch_size=32, callbacks=[tensorboard_callback, lr_scheduler]) + history = model.fit(X_train_updated, Y_train, epochs=500, + batch_size=32, callbacks=[tensorboard_callback, lr_scheduler], verbose=0) # 评估模型 loss, accuracy = model.evaluate(X_test, Y_test) + print(f"Accuracy gamma: {gamma},learning:{learning_rate}", accuracy) # 记录准确率 accuracy_list.append(accuracy) @@ -177,7 +239,7 @@ learning_rates = [0.1, 0.2, 0.3, 0.4, 0.5] gammas = [0.05, 0.1, 0.2, 0.4] # 创建图像 -last_plt = plt.figure(figsize=(10, 6)) +plt.figure(figsize=(10, 6)) # 为每个gamma值绘制曲线 for gamma in gammas: