python宽度学习训练后模型的持久化存储和快速调用

在模型训练完成后，我们需要对我们训练出来的模型进行持久性储存，这样既能将我们调参后得到的最佳模型进行存储，还可以方便后期同团队的人进行调用预测。

1.原理

此处用到的是sklearn库中的joblib包进行存储和加载

因为宽度学习的类属于自定义类，所以在调用时需要在调用的py文件中加入bls代码中的类（在bls代码中分别是node_generator, scaler, broadNet）

如果不加入这些类，由于宽度学习是未知自定义的模型的结构，joblib包将无法解析模型，出现报错：AttributeError: Can‘t get attribute ‘XXX‘ on ＜module ‘__main__‘ from XXX＞

2.核心代码

首先我们需要在训练模型后，对训练后的模型进行存储

核心代码

# bls模型训练
bls.fit(traindata, trainlabel)
# 存储训练后的模型
joblib.dump(bls,"model1.pkl")

然后再另一文件中加载模型文件——model1.pkl

核心代码

# 加载模型
BLS = joblib.load("model1.pkl")
# 用加载后的模型对测试集进行预测
predicts = BLS.predict(test_data)

3.完整代码

训练及存储模型宽度学习（bls）代码：

import numpy as np
from sklearn import preprocessing
import pandas as pd
from sklearn.model_selection import train_test_split
import datetime
import joblib# 准确度显示
def show_accuracy(predictLabel, Label):Label = np.ravel(Label).tolist()predictLabel = predictLabel.tolist()count = 0for i in range(len(Label)):if Label[i] == predictLabel[i]:count += 1return (round(count / len(Label), 5))# 线性/非线性变化
class node_generator(object):def __init__(self, isenhance=False):self.Wlist = []self.blist = []self.function_num = 0self.isenhance = isenhancedef sigmoid(self, x):return 1.0 / (1 + np.exp(-x))def relu(self, x):return np.maximum(x, 0)def tanh(self, x):return (np.exp(x) - np.exp(-x)) / (np.exp(x) + np.exp(-x))def linear(self, x):return xdef orth(self, W):"""orth是正交基的意思，求正交基可能是为了使增强节点彼此无关目前看来，这个函数应该配合下一个generator函数是生成权重的此函数传入的weights与传出的weights的shape是一样的。"""for i in range(0, W.shape[1]):w = np.mat(W[:, i].copy()).Tw_sum = 0for j in range(i):wj = np.mat(W[:, j].copy()).Tw_sum += (w.T.dot(wj))[0, 0] * wjw -= w_sumw = w / np.sqrt(w.T.dot(w))W[:, i] = np.ravel(w)return Wdef generator(self, shape, times):for i in range(times):W = 2 * np.random.random(size=shape) - 1if self.isenhance == True:W = self.orth(W)  # 只在增强层使用b = 2 * np.random.random() - 1yield (W, b)def generator_nodes(self, data, times, batchsize, function_num):# 按照bls的理论，mapping layer是输入乘以不同的权重加上不同的偏差之后得到的# 若干组，所以，权重是一个列表，每一个元素可作为权重与输入相乘self.Wlist = [elem[0] for elem in self.generator((data.shape[1], batchsize), times)]self.blist = [elem[1] for elem in self.generator((data.shape[1], batchsize), times)]self.function_num = {'linear': self.linear,'sigmoid': self.sigmoid,'tanh': self.tanh,'relu': self.relu}[function_num]  # 激活函数供不同的层选择# 下面就是先得到一组mapping nodes，再不断叠加，得到len(Wlist)组mapping nodesnodes = self.function_num(data.dot(self.Wlist[0]) + self.blist[0])for i in range(1, len(self.Wlist)):nodes = np.column_stack((nodes, self.function_num(data.dot(self.Wlist[i]) + self.blist[i])))return nodesdef transform(self, testdata):testnodes = self.function_num(testdata.dot(self.Wlist[0]) + self.blist[0])for i in range(1, len(self.Wlist)):testnodes = np.column_stack((testnodes, self.function_num(testdata.dot(self.Wlist[i]) + self.blist[i])))return testnodes# 归一化处理
class scaler:def __init__(self):self._mean = 0self._std = 0def fit_transform(self, traindata):self._mean = traindata.mean(axis=0)self._std = traindata.std(axis=0)return (traindata - self._mean) / (self._std + 0.001)def transform(self, testdata):return (testdata - self._mean) / (self._std + 0.001)# 宽度神经网络结构
class broadNet(object):def __init__(self, map_num=10, enhance_num=10, map_function='linear', enhance_function='linear', batchsize='auto'):self.map_num = map_numself.enhance_num = enhance_numself.batchsize = batchsizeself.map_function = map_functionself.enhance_function = enhance_functionself.W = 0self.pseudoinverse = 0self.normalscaler = scaler()self.onehotencoder = preprocessing.OneHotEncoder(sparse=False)self.mapping_generator = node_generator()self.enhance_generator = node_generator(isenhance=True)def fit(self, data, label):if self.batchsize == 'auto':self.batchsize = data.shape[1]data = self.normalscaler.fit_transform(data)label = self.onehotencoder.fit_transform(np.mat(label).T)mappingdata = self.mapping_generator.generator_nodes(data, self.map_num, self.batchsize, self.map_function)enhancedata = self.enhance_generator.generator_nodes(mappingdata, self.enhance_num, self.batchsize,self.enhance_function)print('number of mapping nodes {0}, number of enhence nodes {1}'.format(mappingdata.shape[1],enhancedata.shape[1]))print('mapping nodes maxvalue {0} minvalue {1} '.format(round(np.max(mappingdata), 5),round(np.min(mappingdata), 5)))print('enhence nodes maxvalue {0} minvalue {1} '.format(round(np.max(enhancedata), 5),round(np.min(enhancedata), 5)))inputdata = np.column_stack((mappingdata, enhancedata))print('input shape ', inputdata.shape)pseudoinverse = np.linalg.pinv(inputdata)# 新的输入到输出的权重print('pseudoinverse shape:', pseudoinverse.shape)self.W = pseudoinverse.dot(label)def decode(self, Y_onehot):Y = []for i in range(Y_onehot.shape[0]):lis = np.ravel(Y_onehot[i, :]).tolist()Y.append(lis.index(max(lis)))return np.array(Y)def accuracy(self, predictlabel, label):label = np.ravel(label).tolist()predictlabel = predictlabel.tolist()count = 0for i in range(len(label)):if label[i] == predictlabel[i]:count += 1return (round(count / len(label), 5))def predict(self, testdata):testdata = self.normalscaler.transform(testdata)test_mappingdata = self.mapping_generator.transform(testdata)test_enhancedata = self.enhance_generator.transform(test_mappingdata)test_inputdata = np.column_stack((test_mappingdata, test_enhancedata))return self.decode(test_inputdata.dot(self.W))if __name__ == '__main__':# load the datatrain_data = pd.read_csv('../train.csv')test_data = pd.read_csv('../test.csv')samples_data = pd.read_csv('../sample_submission2.csv')le = preprocessing.LabelEncoder()#for item in train_data.columns:#    train_data[item] = le.fit_transform(train_data[item])label = train_data['label'].valuesdata = train_data.drop('label', axis=1)data = data.valuesprint(data.shape, max(label) + 1)traindata, testdata, trainlabel, testlabel = train_test_split(data, label, test_size=0.2, random_state=0)print(traindata.shape, trainlabel.shape, testdata.shape, testlabel.shape)bls = broadNet(map_num=32,enhance_num=33,map_function='sigmoid',enhance_function='sigmoid',batchsize=200)starttime = datetime.datetime.now()bls.fit(traindata, trainlabel)endtime = datetime.datetime.now()# 存储训练后模型joblib.dump(bls,"model1.pkl")print('the training time of BLS is {0} seconds'.format((endtime - starttime).total_seconds()))predictlabel = bls.predict(testdata)print(show_accuracy(predictlabel, testlabel))

调用自己训练的模型代码：

import numpy as np
from sklearn import preprocessing
import pandas as pd
import joblibclass node_generator(object):def __init__(self, isenhance=False):self.Wlist = []self.blist = []self.function_num = 0self.isenhance = isenhancedef sigmoid(self, x):return 1.0 / (1 + np.exp(-x))def relu(self, x):return np.maximum(x, 0)def tanh(self, x):return (np.exp(x) - np.exp(-x)) / (np.exp(x) + np.exp(-x))def linear(self, x):return xdef orth(self, W):"""orth是正交基的意思，求正交基可能是为了使增强节点彼此无关目前看来，这个函数应该配合下一个generator函数是生成权重的此函数传入的weights与传出的weights的shape是一样的。"""for i in range(0, W.shape[1]):w = np.mat(W[:, i].copy()).Tw_sum = 0for j in range(i):wj = np.mat(W[:, j].copy()).Tw_sum += (w.T.dot(wj))[0, 0] * wjw -= w_sumw = w / np.sqrt(w.T.dot(w))W[:, i] = np.ravel(w)return Wdef generator(self, shape, times):for i in range(times):W = 2 * np.random.random(size=shape) - 1if self.isenhance == True:W = self.orth(W)  # 只在增强层使用b = 2 * np.random.random() - 1yield (W, b)def generator_nodes(self, data, times, batchsize, function_num):# 按照bls的理论，mapping layer是输入乘以不同的权重加上不同的偏差之后得到的# 若干组，所以，权重是一个列表，每一个元素可作为权重与输入相乘self.Wlist = [elem[0] for elem in self.generator((data.shape[1], batchsize), times)]self.blist = [elem[1] for elem in self.generator((data.shape[1], batchsize), times)]self.function_num = {'linear': self.linear,'sigmoid': self.sigmoid,'tanh': self.tanh,'relu': self.relu}[function_num]  # 激活函数供不同的层选择# 下面就是先得到一组mapping nodes，再不断叠加，得到len(Wlist)组mapping nodesnodes = self.function_num(data.dot(self.Wlist[0]) + self.blist[0])for i in range(1, len(self.Wlist)):nodes = np.column_stack((nodes, self.function_num(data.dot(self.Wlist[i]) + self.blist[i])))return nodesdef transform(self, testdata):testnodes = self.function_num(testdata.dot(self.Wlist[0]) + self.blist[0])for i in range(1, len(self.Wlist)):testnodes = np.column_stack((testnodes, self.function_num(testdata.dot(self.Wlist[i]) + self.blist[i])))return testnodesclass scaler:def __init__(self):self._mean = 0self._std = 0def fit_transform(self, traindata):self._mean = traindata.mean(axis=0)self._std = traindata.std(axis=0)return (traindata - self._mean) / (self._std + 0.001)def transform(self, testdata):return (testdata - self._mean) / (self._std + 0.001)class broadNet(object):def __init__(self, map_num=10, enhance_num=10, map_function='linear', enhance_function='linear', batchsize='auto'):self.map_num = map_numself.enhance_num = enhance_numself.batchsize = batchsizeself.map_function = map_functionself.enhance_function = enhance_functionself.W = 0self.pseudoinverse = 0self.normalscaler = scaler()self.onehotencoder = preprocessing.OneHotEncoder(sparse=False)self.mapping_generator = node_generator()self.enhance_generator = node_generator(isenhance=True)def fit(self, data, label):if self.batchsize == 'auto':self.batchsize = data.shape[1]data = self.normalscaler.fit_transform(data)label = self.onehotencoder.fit_transform(np.mat(label).T)mappingdata = self.mapping_generator.generator_nodes(data, self.map_num, self.batchsize, self.map_function)enhancedata = self.enhance_generator.generator_nodes(mappingdata, self.enhance_num, self.batchsize,self.enhance_function)print('number of mapping nodes {0}, number of enhence nodes {1}'.format(mappingdata.shape[1],enhancedata.shape[1]))print('mapping nodes maxvalue {0} minvalue {1} '.format(round(np.max(mappingdata), 5),round(np.min(mappingdata), 5)))print('enhence nodes maxvalue {0} minvalue {1} '.format(round(np.max(enhancedata), 5),round(np.min(enhancedata), 5)))inputdata = np.column_stack((mappingdata, enhancedata))print('input shape ', inputdata.shape)pseudoinverse = np.linalg.pinv(inputdata)# 新的输入到输出的权重print('pseudoinverse shape:', pseudoinverse.shape)self.W = pseudoinverse.dot(label)def decode(self, Y_onehot):Y = []for i in range(Y_onehot.shape[0]):lis = np.ravel(Y_onehot[i, :]).tolist()Y.append(lis.index(max(lis)))return np.array(Y)def accuracy(self, predictlabel, label):label = np.ravel(label).tolist()predictlabel = predictlabel.tolist()count = 0for i in range(len(label)):if label[i] == predictlabel[i]:count += 1return (round(count / len(label), 5))def predict(self, testdata):testdata = self.normalscaler.transform(testdata)test_mappingdata = self.mapping_generator.transform(testdata)test_enhancedata = self.enhance_generator.transform(test_mappingdata)test_inputdata = np.column_stack((test_mappingdata, test_enhancedata))return self.decode(test_inputdata.dot(self.W))if __name__ == '__main__':test_data = pd.read_csv('../test.csv')samples_data = pd.read_csv('../sample_submission2.csv')# 加载训练好的模型BLS = joblib.load("model1.pkl")predicts = BLS.predict(test_data)# save as csv filesamples = samples_data['ImageId']result = {'ImageId':samples,'Label': predicts }result = pd.DataFrame(result)result.to_csv('../output/model8.csv', index=False)

调用后的模型对测试集进行预测的结果：

本文来自互联网用户投稿，文章观点仅代表作者本人，不代表本站立场，不承担相关法律责任。如若转载，请注明出处。 如若内容造成侵权/违法违规/事实不符，请点击【内容举报】进行投诉反馈！