05 2019 档案
摘要:1、知识点 2、代码 3、目标函数 4、优化目标
阅读全文
摘要:1、获取resource目录下的template路径 String path = Thread.currentThread().getContextClassLoader().getResource("").getPath()+"template";
阅读全文
摘要:1、知识点 2、将数据写入TFRecords import tensorflow as tf import os os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2' FLAGS = tf.app.flags.FLAGS tf.app.flags.DEFINE_strin
阅读全文
摘要:1、知识点 2、代码 3、分布式架构图
阅读全文
摘要:1、知识点 2、代码 3、发展历程 4、卷积与池化输出矩阵维度计算公式 5、损失计算-交叉熵损失公式 6、SoftMax回归计算公式 7、激活函数-Relu
阅读全文
摘要:1、知识点 2、代码
阅读全文
摘要:1、知识点 2、代码
阅读全文
摘要:1、知识点 2、代码
阅读全文
摘要:1、知识点 """ 模拟一个y = 0.7x+0.8的案例 报警: 1、initialize_all_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02
阅读全文
摘要:内容:包含tensorflow变量作用域、tensorboard收集、模型保存与加载、自定义命令行参数 1、知识点 2、代码
阅读全文
摘要:1、知识点 2、代码
阅读全文
摘要:声明:由于业务场景需要,所以根据一个网友的完成的。 1、既然要使用PDF模板填充,那么就需要制作PDF模板,可以使用Adobe Acrobat DC,下载地址:https://carrot.ctfile.com/dir/11269771-27158812-194d66/29433907/ (使用特别
阅读全文
摘要:知识点: 代码: 轮廓系数:
阅读全文
摘要:# coding=utf-8 from sklearn.datasets import load_iris, fetch_20newsgroups, load_boston from sklearn.model_selection import train_test_split, GridSearchCV from sklearn.neighbors import KNeighborsClas...
阅读全文
摘要:# coding = utf-8 from sklearn import datasets from sklearn.datasets import load_iris from sklearn.model_selection import train_test_split ''' 算法是核心,数据和计算是基础 定位: 1、分析数据 2、分析业务 3、应用常见算法进行对比...
阅读全文
摘要:# coding=utf-8 from sklearn.preprocessing import MinMaxScaler from sklearn.preprocessing import StandardScaler from sklearn.preprocessing import Imputer import numpy as np ''' 特征预处理:通过特定的统计方法,将数据转换成...
阅读全文
摘要:# coding = utf-8 from sklearn.feature_selection import VarianceThreshold from sklearn.decomposition import PCA ''' 数据降维:特征的数量减少(即columns减少) 1、特征选择原因:
阅读全文
摘要:# coding=utf-8 from sklearn.feature_extraction.text import CountVectorizer from sklearn.feature_extraction.dict_vectorizer import DictVectorizer from sklearn.feature_extraction.text import TfidfVecto...
阅读全文
摘要:1、data_range生成时间范围 b)将时间字符串转为时间序列 使用pandas提供的方法把时间字符串转化为时间序列 df["timeStamp"] = pd.to_datetime(df["timeStamp"],format=""),其中format参数大部分情况下可以不用写 c)DataF
阅读全文
摘要:1、pandas对缺失数据的处理 判断数据是否为NaN:pd.isnull(df),pd.notnull(df) 处理方式1:删除NaN所在的行列dropna (axis=0, how='any', inplace=False) 处理方式2:填充数据,t.fillna(t.mean()),t.fia
阅读全文
摘要:知识点 1、DataFrame创建,可以通过index和columns指定索引名称 2、DataFrame基础属性和整体情况查询 3、通过pd.sort_values(by="Count_AnimalName",ascending=False).head(5)排序获取次数最高的排名数据 4、切片与索
阅读全文
摘要:知识点 1、Series创建 2、Series索引与切片 a)索引:一个的时候直接传入序号或者index,多个的时候传入序号或者index的列表 b)切片:直接传入start end或者步长即可 3、Series其他属性与方法 4、Series读取外部数据
阅读全文
摘要:# coding=utf-8 import numpy as np ''' 填充nan的数据,为该列的平均值 ''' def fill_ndarray(t1): for i in range(t1.shape[1]): temp_col = t1[:,i] #取每一列 print(temp_col) nan_num =np.count_...
阅读全文
摘要:a = np.array([[11, 12, 13, 14, 15], [16, 17, 18, 19, 20], [21, 22, 23, 24, 25], [26, 27, 28 ,29, 30], [31, 32, 33, 34, 35]]) print(type(a)) # >>><clas
阅读全文
摘要:# coding=utf-8 import numpy as np import random ############一维数组创建############# #方式一 t1 = np.array([1,2,3],dtype=float) print(t1,t1.dtype,t1.shape) #方式二 t2 = np.array(range(5),dtype=float) print(t...
阅读全文
摘要:1、知识点 2、案例 3、案例二 多条形图
阅读全文
摘要:1、案例一 2、案例二 3、案例三 4、案例四 绘制多个直线图
阅读全文
摘要:数据模型原型如下: 1、表输入,针对最新的数据输入的表 2、目标表,需要更新的表 3、两个表都需要进行排序操作 4、合并,根据id进行合并 5、数据同步(包括更新、插入、删除) 6、点击运行,就可以实现数据同步了。
阅读全文
摘要:1、Kettle的Job使用十分简单,这里也只是演示比较简单的操作,创建Job 2、点击转换,然后点击浏览,选择转换对象 3、执行按钮,运行该转换 4、如果需要长期的进行定时转换,可以在Job中的start控件进行配置
阅读全文
摘要:1、安装JDK 2、下载Kettle 3、将文件解压,点击Spoon.bat文件,启动kettle. 4、创建资源库 点击Connet,选择Pentaho Repository 注意: 1、在使用mysql进行连接测试的时候,一定要找到相应的驱动包,我的是5.1.20jar包这个可以百度找一下, 或
阅读全文