python machine learning cookbook chris albon

Smooth python and cookbook learning notes (5), pythoncookbook1. Random selection Use the random module to generate random numbers in python. 1. Randomly select elements from the sequence and use random. choice () >>> import random>>> values = [1, 2, 3, 4, 5, 6]>>> random.choice(values)3>>> random.choice(values)3>>> ran

the number of days in different months. DateTime cannot process months.>>> fromDateutil.relativedeltaImportRelativedelta>>> A = DateTime (2017, 9, 8)>>> A + relativedelta (Months=1) Datetime.datetime (2017, 10, 8, 0, 0)>>> A + relativedelta (months=4) Datetime.datetime (2018, 1, 8, 0, 0)>>> B = DateTime (2017, 11, 11)>>> d = B-a>>>Ddatetime.timedelta (64)>>> d =Relativedelta (b, a)>>>Drelativedelta (Months=+2, Days=+3)>>>d.months2>>>d.days33. Convert string to time, use Datetime.strptime (), co

Smooth python and cookbook learning notes (2), pythoncookbook1. Value assignment of the packet splitting and decompression sequence of tuples Any sequence (or iteratable object) can be decompressed and assigned to multiple variables through a simple assignment statement. The only premise is that the number of variables must be the same as the number of sequential

Smooth python and cookbook learning notes (9), pythoncookbook1. Reduce the number of parameters of callable objects and use functools. partial to freeze Parameters Use functools. partial () to fix one or more values to reduce the call parameters. >>> Def spam (a, B, c, d ):... print (a, B, c, d)...> from functools import partial >>> s1 = partial (spam, 1) # set t

1. Tuple unpacking and decompression sequence assignment　　Any sequence (or an iterative object) can be extracted and assigned to multiple variables by a simple assignment statement. The only prerequisite is that the number of variables must be the same as the number of elements in the sequence.　　1. Parallel Assignment:>>> x = (1, 2>>> A, b = x #>>> a1>>> b 22. Use the * operator to disassemble an iterative object as a function parameter:>>> Divmod (8) # 20 for the remainder of 8, 2 * 8 + 4

input iteration type, so the length of the Cartesian product list is equal to the product of the length of the input variable.1. Calculating Cartesian product using list derivation>>> colors = ['Black',' White'] >>> sizes = ['S','M','L'] >>> tshirts = [(color, size) forColorinchColors forSizeinchSizes]>>> Tshirts[('Black','S'), ('Black','M'), ('Black','L'), (' White','S'), (' White','M'), (' White','L')] >>> forColorinchcolors: # Use for loop is the same effect ... forSizeinchSizes: ...Print((c

operation time complexity is O (log N), where n is the size of the heap, so even when N is very large, they are still running fast. In the above code, the queue contains a tuple (-priority, index, item). The goal of a negative priority is to make the element sort from highest to lowest priority. This sort of heap is the opposite of regular sort by priority from low to high. The role of the index variable is to ensure that the same priority elements are sorted correctly. By saving an ever-increa

Class and object of the 8th chapter 2016.5.38.1 Changing the object's string display __str__ and __repr__%s and%r, mentioned Eval, I didn't use it.8.2 Formatting a custom string __format__8.3 Let the object support context management, __enter__ and __exit__, you can use the WITH8.4 Ways to save memory when creating a large number of objects __slot__,__slot__ is more of a memory-optimized tool than a wrapper tool to prevent users from adding new properties to an instance.8.5 Encapsulating propert

1. The default parameters of the function must be immutableIf the default parameter of a function is a mutable object, then the default parameter is modified outside the function and affects the function itself.def spam (A, b=None): # B to be an immutable parameter, you cannot use a variable parameter such as an empty list [] ... if is None: ... = []...2. Anonymous functions1. You can use anonymous functions when you can't think of a function name or want a short operationLambda

[Machine learning algorithm-python implementation] matrix denoising and normalization, python Machine Learning1. The background project is required. We plan to use python to implement matrix denoising and normalization. The numpy

This article focuses on the contents of the 1.2Python libraries and functions in the first chapter of the Python Machine learning Time Guide. Learn the workflow of machine learning.I. Acquisition and inspection of dataRequests getting dataPandans processing Data1 ImportOS2 ImportPandas as PD3 ImportRequests4 5PATH = R'

Python Machine Learning Theory and Practice (5) Support Vector Machine and python Learning Theory Support vector machine-SVM must be familiar with