计算机毕业设计 基于深度学习的商品销量LSTM时间序列预测 根据地点品牌时间等信息预测未来的商品销量 matplotlib统计图

import pandas as pd import numpy as np # 商品名 stuff_name = "Feeder40" # 读取数据 data = pd.read_csv("老师原数据.csv") # 时间里有0000异常值，删除 data = data[data["Date Invoiced"] != "0000-00-00"] # 转为时间类型 data["Date Invoiced"] = pd.to_datetime(data["Date Invoiced"]) data = data.set_index("Date Invoiced") data = data[data["Order Line SKU"] == stuff_name] # 按周分组计数 data = data.resample("W").sum()["Order Line Qty"] data = pd.DataFrame(data) # 标准化 from sklearn.preprocessing import StandardScaler ss_x = StandardScaler() data = ss_x.fit_transform(data) # 构造时间序列 def series_to_supervised(data, n_in, n_out, dropnan=True): n_vars = 1 if type(data) is list else data.shape[1] df = pd.DataFrame(data) cols, names = list(), list() for i in range(n_in, 0, -1): cols.append(df.shift(i)) names += [('var%d(t-%d)' % (j + 1, i)) for j in range(n_vars)] for i in range(0, n_out, 1): cols.append(df.shift(-i)) if i == 0: names += [('var%d(t)' % (j + 1)) for j in range(n_vars)] else: names += [('var%d(t+%d)' % (j + 1, i)) for j in range(n_vars)] agg = pd.concat(cols, axis=1) agg.columns = names if dropnan: agg.dropna(inplace=True) return agg # 前7天预测后1天 data_series = series_to_supervised(data, 7, 1) x = data_series.iloc[:, :-1] y = pd.DataFrame(data_series.iloc[:, -1]) '''from sklearn.preprocessing import StandardScaler ss_x = StandardScaler() x = ss_x.fit_transform(x) ss_y = StandardScaler() y = ss_y.fit_transform(y)''' # 10%训练集 from sklearn.model_selection import train_test_split x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.1) x_train = np.array(x_train).reshape(x_train.shape[0], 1, x_train.shape[1]) x_test = np.array(x_test).reshape(x_test.shape[0], 1, x_test.shape[1]) # 用tensorflow keras搭建lstm神经网络 from keras.callbacks import EarlyStopping from keras.layers import * from keras import Input from keras import Model es = EarlyStopping(monitor="val_loss", patience=5) x = Input(shape=(1, x.shape[1])) lstm_structure = LSTM(1, input_shape=(1, x.shape[1]))(x) lstm_structure = Dense(32, activation='relu')(lstm_structure) lstm_structure = Dense(32, activation='relu')(lstm_structure) output = Dense(1)(lstm_structure) model = Model(x, output) model.compile(optimizer='rmsprop', loss='mae', metrics=['mse']) history = model.fit(x_train, y_train, epochs=200, batch_size=5, validation_split=0.2, callbacks=[es]) y_pred = model.predict(x_test) # 画图 from matplotlib import pyplot as plt plt.plot(ss_x.inverse_transform(y_test)) plt.plot(ss_x.inverse_transform(y_pred.reshape(-1, 1))) '''plt.plot(np.array(y_test)) plt.plot(np.array(y_pred))''' plt.title(stuff_name) plt.xlabel("sample") plt.ylabel("amount") plt.legend(["true value", "predicted value"]) plt.savefig(stuff_name + ".png") plt.show() # 未来14天 future_days = 14 amounts = data.flatten() for i in range(future_days): last_days = amounts[-7:] next_day = model.predict(last_days.reshape(1, 1, -1)).flatten()[0] # next_day = ss_x.inverse_transform(next_day).round() amounts = np.append(amounts, next_day) result = list(ss_x.inverse_transform(amounts.reshape(-1, 1)).flatten()) result = [int(round(i)) for i in result] print(result[-future_days:])