基于Python的水位预测系统源代码+预测模型文件

import numpy as np import pandas as pd from sklearn.preprocessing import MinMaxScaler import tensorflow as tf import sys # 用来接收参数 0:py文件名 1:CSV文件,2: 模型的类别 javaToPy = sys.argv[0:] # csvPath = "D:\\Program\\workspace\\python\\实训材料\\fallraw_7041JA26clear.csv" # 数据文件 # 加载数据集 pre_len = 24 * 2+24 df = pd.read_csv(javaToPy[1], header=0, index_col=0) # df = pd.read_csv(csvPath, header=0, index_col=0) df = df.iloc[-pre_len:-24,:] # 数据处理 x_pre_set = df.iloc[:, :-1].copy() y_pre_set = df.iloc[:, -1:].copy() sc1 = MinMaxScaler(feature_range=(0, 1)) x_set = sc1.fit_transform(x_pre_set) sc2 = MinMaxScaler(feature_range=(0, 1)) y_set = sc2.fit_transform(y_pre_set) # 多维预测分块-------------------------------------------------- def data_sblocck(x_set, sblock): x_train = [] for i in range(sblock, len(x_set)): x_train.append(x_set[i - sblock:i, :]) x_train = np.array(x_train) return x_train sblock = 2 * 24 # 训练的步数 小时数 x_pre = data_sblocck(x_set, sblock) # 加载模型 if javaToPy[2] == '单层LSTM': # 单层lstm 模型 model = tf.keras.models.load_model(r'D:\Program\workspace\java\Idea\water levels\python\model\1lstm.h5') elif javaToPy[2] == '双层LSTM': # 双层lstm 模型 model = tf.keras.models.load_model(r'D:\Program\workspace\java\Idea\water levels\python\model\2lstm.h5') elif javaToPy[2] == 'GRU': # GRU 模型 model = tf.keras.models.load_model(r'D:\Program\workspace\java\Idea\water levels\python\model\GRU.h5') elif javaToPy[2] == 'simpleRNN': # SimpleRNN 模型 model = tf.keras.models.load_model(r'D:\Program\workspace\java\Idea\water levels\python\model\simpleRNN.h5') else: model = tf.keras.models.load_model(r'D:\Program\workspace\java\Idea\water levels\python\model\BiRNN.h5') # model = tf.keras.models.load_model(r'D:\Program\workspace\java\Idea\water levels\python\model\lstm.h5') # model.summary() # 模型结构 # 数据预测 y_res = model.predict(x_pre) y_result = sc2.inverse_transform(y_res) y_result = y_result.reshape(1, len(y_result)) # y_result 为 预测数据 print(y_result)