talk python training

namespace into the client module, but does not import the following underscore "_" variable, if defined "__all__" The variables are read first. No matter how many variables are copied, the From statement always imports the entire module into memory. A potential threat from the from statement is that if a variable with the same name exists in the client, its value is replaced by the imported value, and the old variable value may be imported when used with reload. If you use from to get a specifi

Summarize some of the experiences of learning to use tensorflow during this time. The main scenario is to use the Python language to train a simple LR model and save the model in Savedmodel format, then restore the model in Python and the Java language to predict the results. (1) Training model Import TensorFlow as TF import numpy as NP #生成训练数据 x = Np.ndarray (Dt

been classified as a high school computer learning course, for today's college students, more free time, learning Python programming language in bridging the relevant skills, but also to master a new skill, for future work more helpful.V. Java developersCompared to Java,python in the past two years more popular, many Java developers aimed at this market, transformation to do

, Output) = Commands.getstatusoutput (' Cat/proc/cpuinfo ')>>> Print Outputprocessor:0Vendor_id:authenticamdCPU family:21... ...>>> Print Status0Note 1: The return value returned by using this method under Unix-like systems (status) is not equal to the return value after execution of the script or command, because the reason for calling Os.wait () is to understand the implementation of the system wait (). You need the correct return value (status), just move the return value to the right 8-bit o

In the model training, especially in the training set to do cross-validation, usually want to save the model, and then put on a separate test set test, the following is the Python training model to save and reuse.Scikit-learn already has the model persisted operation, the import joblib canfromimport joblibModel Save>>>

)Importsocketsys.setdefaultencoding ("UTF8") Socket.setdefaulttimeout (20) Base_url="http://www.zysj.com.cn/zhongyaofang/{}.html"Headers= {'user-agent':'mozilla/5.0 (Windows; U Windows NT 6.1; En-us; rv:1.9.1.6) gecko/20091201 firefox/3.5.6'@click. Command () @click. argument ('INPUT1') @click. Option ("--pos",'-pos')defQuery_tcm_info (input1,pos):"""The script would ignore the previous POS lines"""zhongyaofang_list=Open (INPUT1) POS=int (POS) num=0ifPOS: forIinchRange (0,pos): zhongyaofang_list

The time is divided into 4 weeks, all self-taught, only provide an outline. Applies to web orientation:1, Week1: read "Concise python tutorial", adapt to the Python development environment2, Week2: Write a crawler, need to know more about RE, urllib2, Sqlite3, Threading,queue and several other modules. Need to use on multi-threaded crawl, regular expression analysis, concurrent resource control, re-open the

effect of improving code reusability. >>> y1 = lineconfig (1, 1) # defines the line y = x + 1 >>> p1 = y1 (2) # Calculates the y-coordinate, gets x=2 when Y1 value GT;G t;> print (p1) 3 >>> lstpoint = [Y1 (x) for x in range (3)] # Calculate y-coordinate list >>> print (lstpoint) [1 , 2, 3];>> y2 = lineconfig (4, 5) # defines the line y = 4x + 5 >>> p2 = y2 (x=2) # Calculates the y-coordinate, gets x=2 when Y2 value >> > Print (p2) "Def Line (x): return a*x + B return linedef Newcounter (i=0)

#Coding=utf-8Importh5pyImportNumPy as NPImportCaffe#1. Import Datafilename ='Testdata.h5'F= H5py. File (filename,'R') N1= F.get ('Data') N1=Np.array (N1)Printn1[0]n2=f.get ('label_1d') N2=Np.array (n2) f.close ()#2. Importing models and Networksdeploy='Gesture_deploy.prototxt' #Deploy fileCaffe_model='Iter_iter_1000.caffemodel' #well-trained CaffemodelNET =caffe.net (Deploy,caffe_model,caffe. TEST) Count=0#statistics The number of predicted values and labels equalt=1000#T: Number of samples

12 binary, ASCII code, character encoding development2 List1) Create a list2) Increase, delete, change, check3) Slicing, looping, sorting#!-*-coding = Utf-8-*-L1 = []Print(L1) L2= ['a','b','C','D','e','F']Print(L2) L3=list ()Print(L3)Print(Type (L3))#NestingLa = ['ABC',['a','b','C']]Print(LA)#Index, subscriptL4 = ['x','y','Z', 1,2,3,4,5,'e','R','g','h']Print(l4[2])Print(l4[-1])#slicesPrint(L4[:4])#Step SizePrint(L4[::2])#index value foundPrint(L4.index (3))#Add, append to last note each function

Open123456789f = open (' filename ', ' R ') # Read modef = open (' filename ', ' W ') # Write modef = open (' filename ', ' a ') # Append modeNote: RB is read in binaryNow you think it's useless, right, I think so.ButIn the future use of the socket, the transfer of files, read and write with the binary formRB and WB can transfer files more quickly　　Read content modeF.read () # reads the entire file at a time, the file is not applicableF.readline () # reads only one row at a time, which accounts

list is the current directory, then the order of the modules is searched by the Sys.path list, and the current directory is found, and one does not continue to be searched.Tab Completion Code#!/usr/bin/env Python # python startup file import sysimport readlineimport rlcompleterimport atexitimport os# tab complet Ion Readline.parse_and_bind (' Tab:complete ') # history file Histfile = Os.path.join (os.envir

Re moduleRe.math Match from scratchRe.searchStructure: Re.math (R ' ^c ', a) non-conforming return noneOriginal character:. Any character[] or [a-z,a-z,b]\d Digital \d Non-digital\s Space \s non-whitespace\w Letter \w Non-letter ([a-za-z0-9])^ Beginning $ endGrouping (A|B)Number of Matches* {0,~}+{1,~}? {0,1}{M}/{m,n}Return Object properties: Group (), groups (), span (), string ()Script#!/usr/bin/python#-*-coding:utf-8-*- from __future__ Importprint_

1. The higher order function receives the function as the parameter, returns the function.2. Function closure3. Receive a function as a parameter, wrap it, and return a wrapper function#!/usr/env/python#-*-coding:utf-8-*- from __future__ Importprint_functiondefF1 (x):returnX*2defNew_fn (f):deffn (x):Print("Pager"+f.__name__+'()') returnf (x)returnFNA= NEW_FN (F1) (2)Print(a)Print("##################") b= F1 (2)Print(b) @new_fndefF1 (x):returnX*

parameter list according to function F. For example:>>> L = [1,2,3,4,5]>>> map (lambda x:x*x, L)[1,4,9,16,25]The reduce (f, list) function is similar to map, whose argument f function receives two parameters, and all elements in the returned list are the result of the cumulative calculation of all elements in the parameter list according to function f, such as additive/multiplicative:>>> L = [1,2,3,4,5]>>> reduce (lambda x,y:x+y, l)15>>> reduce ( Lambda x,y:x*y, l)120Then using these two functi

Python 1 day training 102 ---- create a simple blog (below), python102 ---- Continue to change the blog and continue to change the blog in the previous article. My mother tongue is not swollen in English. I have a lot of English on my blog! The change is simple. You can make a quick change to see if the Django management tool is translated into the language you want. Only in settings. add 'django. middle

Using the functions in Sklearn makes it easy to divide the data into trainset and TestsetThe function is Sklearn.cross_validation.train_test_split and uses the following:Import NumPy as NP from Import train_test_split>>> X, y = Np.arange (Ten). Reshape ((5, 2)), Range (5)>>>1 ],[2, 3],[4, 5],[6, 7],[8, 9]])>>>1, 2, 3, 4]>>> X_train, X_test, y_train, y_test = train_test_split (... X, y, test_size=0.33, random_state=42) ... >>> X_trainarray ([[4, 51],[6, 7]])>>> y_train[2, 0, 3 ]>>> x_testarray (

'].append (i) Print dic#用默认字典方式Import collections#coding:utf-8li=[11, 33,44,55,66,77,88,99,90] #用默认字典方法提前定义value的数据类型dic = Collections.defaultdict (list) for value in Li:if value>66:dic[' K1 '].append (value) else:dic[' K2 ']. Append (value) print dicQueue#队列分为双向队列和单向队列, all queues are process-safe.Two-way queue is the queue on both sides can be increased or decreased operation, is FIFO modeOne-way queues can increment and decrement operations only one side, so a one-way queue is a stack.This a

can do to set the% (Levelno) S: Print the value of the log level% (levelname) S: Print log level name% (pathname) s: Prints the path of the currently executing program, which is actually sys.argv[0]% (filename) s: Prints the current name of the executing program% (funcName) s: Print the current function of the log% (Lineno) d: Print the current line number of the log% (asctime) s: Time to print logs% (thread) d: Print thread ID% (threadname) s: Print thread name% (process) d: Print process ID%

formatImport re#1l=[' 2016-05-23 ', ' 2016-06-11 ', ' 2016-07-12 ']l1= ' 2016-05-23 ' Print (Re.sub (' (\d{4})-(\d{2})-(\d{2}) ', R ' \2 /\3/\1 ', L1))ResultsThe second method ofImport re#1l=[' 2016-05-23 ', ' 2016-06-11 ', ' 2016-07-12 ']l1= ' 2016-05-23 ' Print (Re.sub (? P10. String concatenationThe first of these methodsBut this method wastes memory, here is the second method, with the join11, adjust the center of the string, align left and rights= ' abc ' Print (s.ljust) print (S.ljust (+,